A jury found the defendants, Violet Amirault (Amirault) and Cheryl Amirault LeFave (LeFave), guilty of indecent assault and battery on a child under the age of fourteen and rape of a child under the age of sixteen.2 The convictions involved four child-victims, both female and male. On appeal, the defendants claim that the trial judge erred in (1) permitting testimony on the subject of child pornography; (2) admitting evidence of fresh complaint; (3) determining that the child-witnesses were competent to testify; (4) denying the defendants’ motion for required findings of not guilty on two indictments; and (5) denying their motion for mistrial based on the prosecutor’s closing argument. We transferred the case here on our own motion. We conclude that there was no reversible error. We affirm.

We summarize the evidence presented at trial. The defendant Amirault was the owner and director of the Fells Acres Day School (Fells Acres) in Malden. Amirault’s daughter, the defendant LeFave, worked at Fells Acres as a teacher. We previously have described the general manner in which Fells Acres was operated. See Commonwealth v. Amirault, 404 Mass. 221, 224 (1989). When the rest of the school’s population went on field trips, some children occasionally remained at Fells Acres in the care of Amirault, LeFave, or Gerald “Tooky” Amirault, who also worked at the school and is Amirault’s son and LeFave’s brother. The *929four victims in this case were among the children who occasionally were left at the school during those field trips.

On September 2, 1984, a mother of a Fells Acres child made an allegation of sexual abuse against Gerald Amirault.3 After Fells Acres was closed in September, 1984, a parents’ meeting was held at the Malden police station, where parents were informed of the symptoms of sexual abuse and directed to question their children about a magic room, a secret room, and a clown.

The children’s descriptions of the sexual abuse varied in some details, but were similar over all. The three female victims each testified that the defendants took them separately to the “magic room,” described as a bathroom on the second floor of the school with a small, child-sized door. Two girls testified that both of the defendants touched her naked “bum” and vagina with their (the defendants’) hands or a purple “magic wand.” One of the girls testified that only LeFave touched her vagina and “bum.” One child was forced to touch both of the defendants’ “bums” and to kiss their vaginas. One of the girls testified that both of the defendants inserted their fingers, a thermometer, and the “magic wand” into her vagina or “bum.” A second girl testified that LeFave inserted a thermometer and a pencil into her vagina and rectum.

The male victim testified that the defendants took him to a garage at Amirault’s house. There, Amirault “stuck a stick up [his] bum” and LeFave “put her mouth on his private spot [penis].” He stated that the defendants hurt birds and squirrels in front of the children and the teachers; that they gave him white pills which he never took; and that they tied him, naked, to a tree in front of the other children and teachers.

Each child testified that she or he was photographed in the nude by one of the defendants while the abuse occurred. Two of the children described a black camera on a tripod. The *930other two described a black camera with pictures that came out of its front. While at home, one child spontaneously demonstrated “sexy” poses in the nude for her mother.

All of the children stated that they were threatened by one or both of the defendants that if they told anyone about the incidents, they would be sent away from home or killed by their mothers or that the defendants would kill the children’s parents. A child who attended Fells Acres but who was not a victim testified that he overhead Amirault tell the male victim that “you better keep quiet or else I’ll cut your mother’s arm off.”

The parents of the child-witnesses testified about their children’s behavior while, or shortly after, attending Fells Acres. All four of the child-victims began to demonstrate pronounced sexual behavior. Two of the children made sex-' ual advances toward their mothers. Most of the children began nightly bedwetting, had nightmares, and developed a fear of being alone. Several reverted to baby talk, screamed and cried when bathed, exhibited fear at the sight of LeFave, and complained of pain in their genital areas.

Dr. Renee Brant, a child psychiatrist testifying for the Commonwealth, stated, in part, that sexualized behavior4 is a symptom of sexual abuse. Trauma specific behaviors include sexualized behavior such as masturbation, inserting objects into body cavities, approaching others in a sexually provocative manner, or asking others to touch their genitals. Trauma nonspecific behaviors include complaints of stomachaches, headaches, nightmares, sleep difficulties, aggressiveness, lack of trust, nervousness, baby talk, bedwetting, withdrawal, lying, separation anxiety, and hyper-vigilance.

Dr. Jean Emans, a pediatric gynecologist, examined the three female victims who testified in this case and made positive findings as to all of them. All suffered vulvitis. See Amirault, supra at 226. Two had redness around their labia *931and one child also had cracking and Assuring around her labia and a small bump on her hymen. Dr. Emans testified that these findings were significant because they are more commonly seen in sexually abused children than in nonabused children and that it was unusual to have found vulvitis in three girls under the age of six who attended the same school.

In their defense, the defendants offered testimony from twelve teachers and teachers’ aides employed at Fells Acres. Those employees stated that they never heard of a magic room nor did they see any indication of sexual abuse. One said that she never returned from a field trip to find the defendants alone with children. Each denied seeing the male tied naked to a tree or that animals were harmed.

Defense witness Dr. William Erickson, a psychiatrist, testified that some of the behaviors seen in sexually abused children can be attributed with equal frequency to causes other than sexual abuse. He testified that the most common source of excessive sexual behavior in children is observed behavior in the home. He also commented on videotaped interviews of the three female victims and stated that, in his opinion, the interviewer made no effort to distinguish between fact and fantasy. A psychologist, Dr. Sherry Skidmore, reviewed one taped interview and testified that the interview procedure was flawed. The pediatrician of one female victim, Dr. Andrew Guthrie, testified that his examination of the child in September, 1984, when the school was closed, showed no signs of vulvitis.

1. The admission of inspector Dunn’s testimony on child pornography. The last prosecution witness was John Dunn, a United States postal inspector specializing in the investigation of child pornography. Dunn testified that he had seen approximately one thousand pornographic photographs of male and female children between the ages of three and six years old. He stated that “still” photographs included the thirty-five millimeter variety and that “Polaroid” types were “very common.”

*932Dunn described the types of poses and props or paraphernalia generally seen in child pornography. As he testified, he had in front of him on a podium, out of the jury’s view, photocopies of pornographic photographs involving children that had been seized in unrelated cases. Those photocopies were not introduced in evidence or displayed to the jury. Dunn was allowed to refer to particular photographs during his testimony. Defense counsel repeatedly objected to Dunn’s testimony and the use of the photographs. As Dunn’s descriptions of a particular photograph became more detailed, defense counsel made an oral motion for a mistrial. The judge denied the motion.

Dunn testified that child pornography usually depicts nude or partially nude children posed in a manner focusing on a genital area. Testifying from his general knowledge and by direct reference to the photographs, Dunn described young children engaged in a variety of sexual acts with both male and female adults. Although he described poses and props similar to those mentioned by the children here (sticks, pencils, thermometers), he also described poses and objects that were not alleged to have been involved in this case.

Dunn briefly described the underground market for child pornography and testified that there is a market for the types of photographs he had described. He stated that child pornography involving prepubescent children is not readily available to members of the public and that it is extremely difficult for investigators to locate pornographic photographs or to identify a child in a particular photograph.

On cross-examination, Dunn testified that, to his knowledge, none of the photographs in front of him depicted either of the defendants or any child known to have attended Fells Acres. Dunn testified that he spoke with other postal inspectors’ offices and State authorities to determine whether the defendants or any Fells Acres children appeared in any photograph. To his knowledge, no such photographs had ever been found.

The defendants claim that they were prejudiced by Dunn’s testimony, because the evidence (1) was “completely irrelevant”*9335 to the purpose for which it was admitted and (2) its admission was tantamount to the admission of “other crimes” evidence and hence prejudicial error. The defendants also allege error in admitting Dunn’s testimony as to child pornography because the subject was within the jurors’ knowledge and therefore it was error to admit expert testimony. They also allege error in the instructions.

The record shows that the Commonwealth offered Dunn’s testimony to establish a motive for the abuse.6 The defendants concede (1) that, as a general rule, the Commonwealth is entitled to introduce all relevant evidence of motive, see Commonwealth v. Weichell, 390 Mass. 62, 73 (1983), cert. denied, 465 U.S. 1032 (1984); Commonwealth v. Borodine, 371 Mass. 1, 8 (1976), cert. denied, 429 U.S. 1049 (1977); (2) that “Dunn’s testimony was not offered as evidence of acts committed by the defendants”; and (3) that the Commonwealth’s theory at trial was “that the children here were sexually molested for the purpose of obtaining pornographic photographs.” “The purpose for which evidence is offered is determinative of its relevancy.” Commonwealth v. Jewett, 392 Mass. 558, 564 (1984). See Commonwealth v. Soares, 384 Mass. 149, 161 (1981). The Commonwealth asserts that Dunn’s testimony provided a possible motive for the sexual abuse and thus was relevant to the issue of motive and, hence, admissible on that issue. See Weichell, supra; Com*934monwealth v. St. Germain, 381 Mass. 256, 271 (1980); Commonwealth v. Brown, 376 Mass; 156, 164 (1978); Commonwealth v. Borodine, 371 Mass. 1, 8 (1976).

Dunn’s testimony suggested that, generally, one motive for sexually abusing children is the production of pornographic photographs for distribution in an underground market. “Proof of motive need not be by direct evidence but may be based on inferences which could reasonably be drawn from the circumstances. . . . [E]vidence that ‘merely suggests rather than “clearly shows” a motive for the crime may still be ruled admissible.’ ” (Citations omitted.) Weichell, supra at 73-74, quoting St. Germain, supra at 271. Although the defendants correctly note that no photographs connected to them were found, they concede that “[e]vidence to support this theory of motive came from [the four] children who testified that their pictures were taken coincident with the alleged acts of abuse.” The children testified that they were photographed while posed in the nude, often with a variety of objects inserted in their vaginas or rectums.7 Dunn’s testimony was relevant to suggest a motive for photographing the children in these various poses. The judge did not abuse his discretion by admitting Dunn’s testimony as relevant to the issue of motive.8 See St. Germain, supra at 271.9

*935The defendants next claim that “the admission of Dunn’s testimony was tantamount to the admission of ‘other crimes’ evidence not involving these defendants” and suggest that it was error to admit his testimony because there was no connection between the defendants or the crimes here and the “bad acts evidence.” We conclude that there was no error. The judge instructed the jury on the limited purposes for which he admitted all of the picture taking testimony: to establish motive. Dunn’s testimony was relevant and admissible for that purpose. If “evidence [here, testimony regarding the taking of pictures of the children in various poses] ... is connected with the facts of the case, it may be admitted to establish ‘knowledge, intent, motive, method, material to proof of the crime charged.’ ” Commonwealth v. Imbruglia, 377 Mass. 682, 695 (1979), quoting Commonwealth v. Murphy, 282 Mass. 593, 598 (1933). The testimony of each of the child victims that they were photographed by the defendants while the defendants sexually abused them was sufficient to suggest the defendants had a motive for their conduct.

*936The cases that the defendants rely on do not advance their claim. In Commonwealth v. Tobin, 392 Mass. 604 (1984), this court concluded that “evidence of the relationship between . . . other participants in the second enterprise was not relevant to their involvement in the earlier and completely separate scheme involving Tobin.” Id. at 615-616 (finding no reversible error in improperly admitting evidence of the second enterprise). Here, there were no allegations that other persons were involved in the defendants’ scheme to sexually abuse the children and Dunn’s testimony was not offered for that purpose. The defendants also rely on Commonwealth v. Szemetum, 3 Mass. App. Ct. 651 (1975), which was concerned with the imputation of knowledge, or criminal intent, by association with unrelated arrests. The factual similarities between Szemetum and the present case and, hence, its precedential value, are tenuous at best. Dunn’s testimony was not offered to prove criminal intent or to show that the defendants had knowledge that they or any other persons were engaged in criminal activity.

The defendants also argue that “[t]he Court’s instruction regarding the use of the evidence of pornography for purposes of corroboration, invited the jury to conclude that the defendants were part of a common scheme to create and distribute child pornography.” This was not the basis of counsel’s objection at trial.10 Defense counsel’s objection at the *937conclusion of the instructions clearly indicated that his objection was based on the use of one child’s testimony to corroborate the other, and not that the instructions referred to Dunn’s testimony. Commonwealth v. Drew, 397 Mass. 65, 80 (1986).11

The judge specifically instructed the jury that the defendants were not charged with child pornography and that no testimony was being admitted as evidence of that crime. In his instruction, the judge was referring to the testimony by the children that their pictures had been taken in the course of the rapes and indecent assault and battery incidents. See note 11, supra. The instruction at issue did not in any way single out or highlight Dunn’s testimony or the subject of child pornography. Rather, the judge referred generally to “testimony with respect to the taking of pictures of the children,” which was concerned primarily with the testimony regarding the taking of photographs given by the four children. Moreover, the instruction on “common scheme” addressed the credibility of the children and the use of one child’s testimony to corroborate another child’s testimony.12 Defense *938counsel clearly understood the instructions that way. See note 11, supra. That instruction referred to a plan or scheme involving picture taking in the rape and indecent assault and battery cases and did not in any way allude to a scheme “to create and distribute child pornography.” In sum, none of the instructions invited the jury to consider Dunn’s testimony as evidence of “other crimes” or of a common scheme to engage in child pornography.

2. The prosecutor’s closing argument. The defendants argue that comments made by the prosecutor in closing “were impermissible attempts to invoke sympathy and emotion of the jurors” and were “so prejudicial as to require reversal.” We do not agree.13

In their summations, both defense attorneys pursued that theory by attacking the credibility of the victims. The defendants emphasized that the victims delayed in disclosing the sexual abuse and attacked the prosecution’s theory that the defendants’ threats caused those delays. During their summations, the defendants’ attorneys theorized that, as a result of repeated leading and suggestive questions by parents, police, social workers, doctors, and prosecutors, the victims fabricated the allegations of abuse.

*939The Commonwealth claims that “the prosecutor did nothing more than fairly respond to the defendants’ summation by offering an explanation for the victim [s’] delay in disclosing the abuse, and asking the jury to evaluate [the victims’] testimony concerning the threat in light of their own experience with children of similar age.” The Commonwealth argues that, “[h]aving forcefully used the fact of the victims’ delayed disclosures and questioning by different people to suggest that there were no threats and sexual abuse, the defendants should not now be allowed to argue that the prosecutor was foreclosed from responding with equal force in his summation that the victims’ delay in disclosing the sexual abuse was explained by, and consistent with, being threatened by the defendants.” We agree with the Commonwealth. A “defense counsel’s argument may justify a particular rebuttal from the prosecutor . . . .” Commonwealth v. Kozec, 399 Mass. 514, 519 (1987). The prosecutor’s comments were reasonably “within the prosecutor’s right of retaliatory reply.” Commonwealth v. Prendergast, 385 Mass. 625, 633 (1982), and cases cited. See Commonwealth v. Earltop, 372 Mass. 199, 206 (1977) (Hennessey, C.J., concurring).

The judge denied the defendants’ motion for mistrial made immediately following the prosecutor’s summation. The judge gave a curative instruction, as requested by the defendants.14 The defendants now argue that that instruction “[did] little to alleviate the prejudicial effect of the comments.” We do not agree. The judge’s instructions specifically addressed the comments and clearly directed jurors not to be influenced by their status as parents and grandparents. “We are unwill*940ing to assume that the jury would not have heeded these instructions.” Commonwealth v. Errington, 390 Mass. 875, 882 (1984), and cases cited. See Commonwealth v. Amirault, 404 Mass. 221, 240 (1989).

The defendants also argue that the statements here “are extremely similar” to the closing remarks in Commonwealth v. Sevieri, 21 Mass. App. Ct. 745, 753-754 (1986). We do not agree. In Sevieri, while describing each step taken by an assailant during an assault, the prosecutor repeatedly directed the jurors to place themselves in the position of the victim. The remarks here were not reasonably susceptible of being interpreted as an appeal to the jurors to place themselves in the victim’s position. Moreover, the remarks in Sevieri were not made in response to the defendant’s closing argument, as occurred here.

3. The admission of fresh complaint testimony. The defendant LeFave challenges the admission of fresh complaint testimony from the mother of one of the children.16 The mother testified that, when interviewed by a social worker approximately eighteen months after leaving Fells Acres, her daughter initially denied being abused, but that within minutes of that interview’s completion, the child told her mother that a clown put a pencil, a thermometer, and his “pee-pee” in the child’s “bum-bum” and “pee-pee.” The mother testified that, at a later time, her daughter stated that LeFave touched her in the vagina and “bum-bum” with a pencil and a thermometer and took photographs of the naked child. The defendant LeFave asserts that the judge abused his discretion in admitting the fresh complaint testimony concerning the clown, arguing that that testimony did not corroborate the child’s allegations with respect to LeFave. The defendant LeFave concludes that, because the complaint “was in no manner corroborative of the victim’s testimony at trial and was inflammatory and prejudicial, the trial court abused its discretion in admitting evidence of the . . . fresh complaint.” *941We conclude that the judge did not err by admitting the “whole statement including the details” for corroborative purposes only. Commonwealth v. Hanger, 357 Mass. 464, 466 (1970). Commonwealth v. Bailey, 370 Mass. 388, 396 (1976).

In sexual assault cases, evidence of a fresh complaint is admitted “for the more general purpose of confirming the testimony of the . . . [victim].” Hanger, supra at 466, quoting Commonwealth v. Cleary, 172 Mass. 175, 176-177 (1898). On both direct and cross-examination, the child was asked about her interview with the social worker and the child testified that she talked about a clown. The fact that the mother’s fresh complaint testimony was more detailed about the clown than the child’s testimony did not render her fresh complaint evidence inadmissible. See Bailey, supra at 396.

During the mother’s testimony, the judge instructed the jury that fresh complaint testimony was being admitted only for corroborative purposes. He repeated that instruction in his final charge to the jury. The defendant LeFave does not claim error in the instructions. Compare Bailey, supra, with Commonwealth v. Spare, 353 Mass. 263, 266 (1967). “The testimony complained of assumed particular relevance when considered in light of the various attempts by the defendant’s trial counsel to insinuate by questioning and argument that the testimony of the victims had been influenced by parents, social workers and members of the prosecution team. The cases decided in the wake of Commonwealth v. Bailey, 370 Mass. 388, 391-397 (1976), do not support the notion that fresh complaint evidence should be excluded in cases such as the present.” Commonwealth v. Baran, 21 Mass. App. Ct. 989, 991 (1986), and cases cited.

4. Competency of the child-witnesses. In Commonwealth v. Tatisos, 238 Mass. 322, 325 (1921), this court stated that, in order to determine competency of a child-witness, “ ‘[t]he ultimate test . . . must depend upon the existence of [an] understanding sufficient to comprehend the difference between truth and falsehood, the wickedness of the latter and *942the obligation and duty to tell the truth, and, in a general way, belief that failure to perform the obligation will result in punishment.’ ” Commonwealth v. Reid, 400 Mass. 534, 542 (1987), quoting Tatisos, supra. The defendants argue that the judge’s examination of the four child-witnesses “failed to completely satisfy the . . . [Tatisos] test since the inquiries failed to elicit responses demonstrating each child’s awareness that a failure to tell the truth would result in punishment or other consequences, [and as] a result, the [judge’s] determination that each child was [competent was] based on an incorrect application of the relevant legal standard.” They claim that “[h]ere there is no indication that the four children . . . possessed a general understanding of the consequences of telling a lie.”

In determining competency, “[t]he judge is afforded wide discretion — indeed, is obliged — to tailor the competency inquiry to the particular circumstances and intellect of the witnesses.” Commonwealth v. Brusgulis, 398 Mass. 325, 329-330 (1986). “Much which cannot be reproduced by the printed word depends on the child’s appearance and manner. It is seldom that the discretion of the trial judge can be revised; its exercise must have been clearly erroneous to justify such action.” Tatisos, supra at 325, and cases cited. See Reid, supra at 542. Although the judge could have asked a specific question about whether the children understood the consequences of telling a lie, the record supports his competency findings. The children’s answers to the judge’s questions indicated that they were aware of the difference between a truthful statement and a lie and that they understood the importance of telling the truth in court. There was no error. See Commonwealth v. Dockham, 405 Mass. 618, 624 (1989); Commonwealth v. Bergstrom, 402 Mass. 534, 540 n.7 (1988); Commonwealth v. Welcome, 348 Mass. 68, 70 (1964); Tatisos, supra at 325-326; Commonwealth v. Corbett, 26 Mass. App. Ct. 773, 776 (1989).

5. Denial of the defendants’ motions for required findings of not guilty on two indictments. Amirault and LeFave were both indicted for the rape and indecent assault and battery of *943the male victim. At trial, the male victim testified to a single incident where Amirault placed a stick in his rectum and LeFave put her mouth on his penis while Gerald Amirault took pictures. Defense counsels’ motions for a required finding of not guilty on the indictments charging the defendants with indecent assault and battery were denied. The defendants claim that the judge committed error by denying the motions, arguing that the victim described a single act that could not support a verdict of guilty on charges of both rape and indecent assault and battery.

The defendants acknowledge that “to determine whether a defendant may be convicted of two statutory offenses arising from a single incident, ‘the long-prevailing test in this Commonwealth is whether each crime requires proof of an additional fact that the other does not.’ ” Commonwealth v. Crocker, 384 Mass. 353, 357 (1981), quoting Commonwealth v. Jones, 382 Mass. 387 (1981). Morey v. Commonwealth, 108 Mass. 433, 434 (1871). At the time of the offense here, indecent assault and battery of a child required proof of lack of consent or lack of capacity to consent, an additional fact not required in statutory rape. See Commonwealth v. Reid, 400 Mass. 534, 541 (1987). Thus, we concluded that “indecent assault and battery of a child . . . was not a lesser included offense within the crime of rape of a child.” Id. The defendants do not cite to nor attempt to distinguish Reid. Rather, they “urge [this] court to adopt the view of Commonwealth v. Jones,” 382 Mass. 387 (1981). That case is of no aid to the defendants because it does not adopt a “view” that is any different from Reid. In Jones, we disallowed multiple punishments because neither offense there required proof of an additional fact that the other did not. Id. at 393-394.

Judgments affirmed.

(dissenting). The court today condones the admission in evidence of highly inflammatory and prejudicial *944testimony offered in support of the dubious proposition that the testimony related to the defendants’ “motive.” Thus, I cannot join in part 1 of the court’s opinion. The crimes of which the defendants have been convicted are indeed heinous. Nevertheless, the facts of this case should not drive the court to a distortion of sound evidentiary principles nurtured and developed over many years as an essential guarantee of a fair trial. We should not allow outrage at the nature of the alleged crimes or sympathy for the alleged victims to cloud our constitutional duty to determine whether there was prejudicial error in this case.

The testimony of the expert witness, United States Postal Inspector Dunn, was not legally relevant to the indictments before the jury. The defendants were charged with indecent assault and battery on children under the age of fourteen and rape of children under the age of sixteen. The Commonwealth offered as evidence of the defendants’ motive Dunn’s detailed descriptions of pornographic photographs of children taken by persons not in any way connected to the defendants. Dunn also described objects, not mentioned by the children in this case, such as gun barrels, scissors, and dildoes, which were inserted into the genitalia of minors who were completely unrelated to this case.1 Additionally, Dunn was allowed to engage in a generalized discussion before the jury of the child pornography industry.

“In order to be considered relevant, ‘the evidence must have rendered the desired inference more probable than it would have been without it.’ ” Commonwealth v. Fayerweather, 406 Mass. 78, 83 (1989), quoting Commonwealth v. Copeland, 375 Mass. 438, 443 (1978). See Green v. Richmond, 369 Mass. 47, 59 (1975). Testimony concerning photographs depicting unnamed and unidentified children in pornographic poses is not relevant to whether these defendants in this particular case were more likely to have had the mo-*945live of committing the acts with which they were charged. The photographs about which Dunn testified were not related to the defendants in any way; they were not in evidence. They were not taken by the defendants and had not been in the possession of the defendants. There was no evidence to show that the children depicted in the photographs were the children involved in this case. And there was no evidence that the children in this case were forced to pose with gun barrels, scissors, dildoes, fruit, vegetables, or animals.

The Commonwealth apparently desired the jury to infer that, because pornographers often take photographs of children in poses similar to those described by the children involved in this case, these defendants took pictures of the victims for the purpose of distributing and marketing the photographs in the pornographic trade. Without any supporting evidence, that connection is so remote that it cannot be deemed to pass the threshold test of relevance. See Commonwealth v. Fayerweather, supra at 83. There was no evidence presented that the defendants were engaging in the pornography trade. No pornographic photographs allegedly taken by the defendants were presented in evidence. In short, Dunn’s testimony was irrelevant to show motive.2 See Commonwealth v. Lamrini, 392 Mass. 427, 433-435 (1984).

*946Even assuming Dunn’s testimony about pornography was marginally relevant, its probative value was far outweighed by its highly prejudicial effect.3 Dunn testified about the contents of several unidentified photographs in painstaking detail.4 He alluded to props and poses in photographs of unknown origin similar to pictures described by the children as taken by the defendants in this case.5 Dunn also described the “underground market” for child pornography. Dunn testified that it is very difficult to identify a child-victim from a particular piece of pornography. He finally testified that there is no central source in this country to help identify a child-victim in a particular photograph.

Dunn’s testimony suggested to the jury that these very photographs could have been photographs of the children in this case. Without any supporting evidence, and placed just beyond the view of the jury, these photographs were used by the witness in the presence of the jury. This testimony and Dunn’s related testimony served more to inflame the passions of the jury than to assist them in deciding an issue in this case.

*947The defendants were not charged with violating the laws prohibiting child pornography. Yet Dunn’s testimony suggested to the jury that the defendants had violated the pornography laws — based not on the acts of the defendants but on the acts of unnamed and unknown third parties.6 Even if the jury did not infer that the photographs described depicted the children in this case, the testimony was nonetheless highly prejudicial. Testimony concerning child abuse perpetrated by anonymous persons on anonymous children could have improperly played on the jury’s sympathy for the children in this case. In addition, there was inherent prejudice in Dunn’s allusion to objects, including scissors, dildoes, vibrators, and gun barrels, which were inserted into the genitalia of minors depicted in photographs unconnected with this case, and in his discussion of sexual poses with animals, also not mentioned by the children in this case. This testimony served only to stir the indignation and disgust of the jury, distracting them from the relevant, admissible evidence in this case.

The judge compounded the prejudice in his instruction to the jury on “common scheme.” The judge, while instructing the jury that the defendants were not charged with the crime of violating pornography laws, nevertheless instructed the jury that they could consider “the taking of pictures of the children” because “it may have shown a common scheme.” The court today states that the judge was referring to the testimony by the children that their pictures had been taken during the course of the abusive conduct. The court does not *948— but should — ask what a reasonable jury could glean from the charge. See Commonwealth v. Claudio, 405 Mass. 481, 484 (1989). The instruction given can be viewed reasonably as a reference to Dunn’s expert testimony which, after all, dealt exclusively with the taking of pictures of children. The jury could have understood participation in the underground pornography industry, described by Dunn, as the “common scheme” to which the judge alluded. The judge stated to the jury that the testimony “was allowed in for the limited purpose of establishing what might have been a motive.” This appears to be a direct reference to Dunn’s testimony about pornography.7 In my view, this is one of those rare cases where the trial judge abused his discretion by admitting in evidence testimony which was not only irrelevant but which was highly prejudicial. See Commonwealth v. Richmond, 371 Mass. 563, 565-566 (1976). For this reason, and because the evidence of guilt was highly dependent on the jury’s assessment of the credibility of witnesses, I would reverse the convictions. I dissent.