FELDMAN, Justice

Kevin Scott Roscoe (defendant) was convicted of first degree murder, kidnapping, and two counts of child molesting. A.R.S. §§ 13-1105, 13-1304, and 13-1410. The trial court found the killing to have been committed in an especially cruel, heinous or depraved manner and, finding no mitigating circumstances, sentenced defendant to death for the murder. The court also imposed aggravated, consecutive sentences of twenty-one years in prison for the kidnapping and fourteen years for each count of child molesting. This court has jurisdiction pursuant to A.R.S. § 13-4031.

FACTS

The victim, Laura Dunn, left her home in the early evening of May 13, 1982 to look for a missing cat. When she failed to return, her family commenced a search and her mother found her bicycle along the edge of the road. The following day Laura's nude body was found face down in a remote desert area some twelve miles from her home. She had been sexually molested and then strangled.

Defendant, on probation for an offense committed in California, became a suspect. Under questioning, he told the police that he was familiar with the area, had been in the area on the evening that Laura disappeared, but had been visiting a friend, had then gone to a party, and had no knowledge of the killing. In fact, a sheriff's deputy searching for Laura had come upon defendant and his friends at a party at a nearby lake on the night of the crime. However, defendant's statement with respect to the time of his arrival at the party was not corroborated by his friends, who placed his arrival considerably later than the time he claimed in his statement.

A witness identified defendant's car as similar to one he had seen driving away at a high speed from the general area of the crime at about 7:30 p.m. on the night Laura was killed. He stated that the driver was a young white male with long brown hair not unlike the defendant's. In defendant's car police found some household carpeting which was similar in fiber structure to carpet pile found near Laura's body. Also found in defendant's car were hairs from a human head that proved to be similar to those of the victim. The car seat had traces of human blood on it. Scientific analysis of enzymes revealed that semen found in the victim's mouth was consistent with the hypothesis that defendant was the attacker. Pubic hairs found on the front of Laura's blouse were of a type consistent with defendant's pubic hair.

In addition to these circumstances linking defendant to the crime, the state adduced evidence of a prior bad act committed by defendant as well as evidence of defendant's identification through use of a trained dog. After forty minutes of deliberation, the jury found defendant guilty as charged.

On appeal, we must decide the following questions:

1. Did the trial judge err in admitting evidence of a prior bad act?

2. Should evidence of dog scenting have been admitted at trial?

3. Were defendant's statements of June 18, 1982 properly admitted at trial?

4. Did reversible error occur in the admission of autopsy photographs?

5. Should a new trial be ordered because the state implied that appellant was responsible for another murder?

6. Should the photographic identification of defendant's car have been suppressed?

7. Was defendant denied effective assistance of counsel at sentencing?

8. Was there sufficient evidence to support two counts of child molesting?

9. Is Arizona's death penalty constitutional?

10. Was the death penalty properly imposed in this case?

Admissibility of Prior Bad Act

Defendant claims that the trial court erred in admitting evidence of a crime he committed in California on March 21, 1981. Defendant had entered a plea of guilty to a charge of assault with a deadly weapon and been sentenced to prison in California. He was released after serving six months and had moved to the Phoenix area shortly before the events which formed the basis for the present case. The state called Cheryl Clark, the victim of the California crime, as a witness in the case at bench. She testified that defendant had sexually attacked her, choked her, and abandoned her in a remote location.

Defendant admitted from the stand that he had committed the crime, but claims that the trial court erred in admitting the evidence over objection. The general rule, of course, is that evidence of prior bad acts is inadmissible to prove the bad character of the perpetrator. State v. Moore, 108 Ariz. 215, 495 P.2d 445 (1972); Rule 404(b), Ariz.R.Evid., 17A A.R.S. The rule provides for certain exceptions:

[such evidence may] be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.

Id. (Emphasis added.)

To the enumerated exceptions contained in Rule 404(b), Arizona courts have added a further exception that such evidence is admissible to show emotional propensity for sexual aberration. State ex rel. LaSota v. Corcoran, 119 Ariz. 573, 583 P.2d 229 (1978); State v. McFarlin, 110 Ariz. 225, 517 P.2d 87 (1973). In the case at bench the prosecution offered the evidence both to prove emotional propensity and to prove identity by showing modus operandi. However, the trial court admitted the evidence only for the latter reason. We review its admission only in that context.

The modus operandi exception was discussed by this court in State v. Jackson, 124 Ariz. 202, 603 P.2d 94 (1979). We held there that evidence of a prior bad act is admissible to identify the defendant as the person who committed the crime for which he is being tried. Id. at 204, 603 P.2d at 96; State v. Moore, supra. When the evidence is used for such a purpose, the state must establish first that the defendant actually was the person who committed the prior act. That part of the problem does not confront us in this case because defendant admitted the prior act.

The second requirement of the modus operandi exception is that the prior acts must be "so unusual and distinctive as to be like a signature." McCormick on Evidence § 190, at 560 (3d ed. 1984). While identity in every particular is not required, there must be similarities between the offenses in those important aspects "when normally there could be expected to be found differences." Jackson, 124 Ariz. at 204, 604 P.2d at 96. The similarity requirement is usually more stringent where evidence of a prior bad act is used to prove identity than where such evidence is offered to prove emotional propensity or state of mind. United States v. Myers, 550 F.2d 1036, 1045 (5th Cir. 1977).

Thus, the modus operandi exception is applied to sex offenses where an adequate foundation is made showing that the prior offense was not too remote in time, was similar to the offense charged and was committed with a person similar to the prosecuting witness in the case being tried. People v. Goodson, 80 Cal.App.3d 290, 292, 145 Cal.Rptr. 489, 491 (1978); People v. Kelly, 66 Cal.2d 232, 57 Cal.Rptr. 363, 371, 424 P.2d 947, 956 (1967). In the case at bench, the two offenses were not remote; the offense charged was committed within six months after defendant's release from prison for the prior offense. Being cognizant of the extreme danger of prejudice from the admission of this type of testimony, we have carefully reviewed the similarity of the offenses and concluded that there are many significant areas of similarity. Both victims had their clothes removed. The blouses of both were torn open in front. The clothes of both were found near their bodies, the shoes of both standing side-by-side. Both victims were left in remote areas. There was evidence that both victims were gagged, a sock being used on each occasion. Both had their hands tied behind their backs, one with socks and the other with her shoelaces. The knots used were similar. Both victims were choked, one with shoelaces and one with elastic hairbands. There was evidence that both oral and vaginal sex were forced on both. A vehicle was involved in both crimes.

We do not believe that these are similarities which are common to all child molestations. Cf. People v. Cook, 11 Ill. Dec. 792, 53 Ill. App.3d 997, 369 N.E.2d 246 (1977). We think, instead, that there are so many similarities between the two offenses that it could fairly be inferred that the known perpetrator of the first offense was probably the same person who committed the second. We believe that the reasonableness of that inference is the true test for admissibility under the modus operandi exception. See United States v. Cavallino, 498 F.2d 1200, 1207 (5th Cir. 1974); Bradley v. United States, 433 F.2d 1113, 1120-21 (D.C. Cir. 1969).

We acknowledge that there is one marked dissimilarity between the two offenses. Laura Dunn was not quite eight years old at the time she was killed while Cheryl Clark was seventeen at the time of the crime perpetrated upon her. We do not find this dissimilarity to be so significant as to preclude admission, especially in view of evidence that Cheryl Clark appeared thirteen rather than seventeen at the time she was attacked, and that the two girls were similar in hair color and complexion. Absolute identity in every detail cannot be expected. Where an overwhelming number of significant similarities exist, the evidence of the prior act may be admitted; the major dissimilarity, and others here present, go to the weight of the evidence. We hold that the trial court did not err.

The Dog Scent Evidence

Following a pretrial evidentiary hearing to determine admissibility and over strenuous objection by defense counsel, the trial court admitted evidence that defendant had been identified by a dog using a scent identification process. Essentially, this evidence showed that Harass II, a pedigreed German Shepherd, had been used to perform certain identification tests. All of the tests were run blind — the dog's handler, John Preston, was not told in advance which article or location was connected with the crime. The test protocol was as follows:

1. The dog was given the victim's scent, obtained from her clothing; when so scented the dog was taken to a line-up of five cars, one of which was the defendant's, for the purpose of identifying the victim's scent in one of those cars, if possible. The handler testified that the dog "alerted" to the victim's scent at a car which was later identified as the defendant's. According to the handler, the dog's reaction indicated that the victim's scent was present in several areas of defendant's car.

2. The dog was scented on clothes belonging to defendant and then taken to the general area where the victim's bicycle had been found; the dog was put on search of the area and alerted at the place where the bicycle had been found. According to Preston, the dog's reaction indicated that defendant had been present in the area where the bike was found. Similarly, the dog alerted to the defendant's scent in the area where the body was found.

3. After having been again scented on defendant's clothes, the dog was taken to a room where five articles of clothing were laid out. After having been ordered to search, the dog alerted at the clothing which had been taken from the victim's body. According to Preston, this indicated that the defendant's scent was present on that article of clothing.

4. After having been put on defendant's scent, the dog was taken to a "line-up" of five bicycles. The dog alerted at the bicycle which had belonged to the victim, thus indicating that the defendant's scent was present on that bicycle.

Defendant has numerous objections to the admission of this testimony. Primarily, he claims that the testimony was inadmissible absent a showing of general acceptance of such procedures in the scientific community. In particular, he claims that there is no scientific acceptance for the hypothesis that even a trained dog can perform scent identification in addition to tracking, and that there is a total absence of scientific agreement that any dog can either track or identify after a long hiatus between the time the scent is laid down on the scene and the time at which the dog is put on the scent. The state argues that proof of acceptance by the scientific community is not required.

The test developed in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923) postulates that before the court may admit evidence based upon

well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs.

Id. at 1014.

The rationale for this standard is probably the fear that jurors will accord scientific evidence too much weight because of its "aura of special reliability and trustworthiness" and will fail to consider the possibility that evidence based upon the particular scientific principle in question may be incorrect. Chapman v. State, 638 P.2d 1280, 1290 (Wyo. 1982). However, the Frye rule is not applicable to all expert evidence; in many instances expert testimony is admissible in the absence of proof or findings of general acceptance, provided that there is sufficient individualized foundation. See State ex rel Collins v. Superior Court, 132 Ariz. 180, 196-99, 644 P.2d 1266, 1282-85 (1982)(supp.op.). In Collins we reiterated that the Frye standard

recognizes that the ultimate criterion in determining admissibility is the perceived probative value of the proffered evidence weighed against its likelihood of undue prejudice.

Id. at 199 n. 4, 644 P.2d 1266. We rejected McCormick's suggestion [McCormick on Evidence § 203, at 491 (2d ed. 1972)] that the Frye test be discarded and replaced by a case-by-case analysis under Rule 403 of the Rules of Evidence. Id. In concluding that the Frye test was applicable to hypnotically induced recall evidence, we relied upon the following:

If there is a discernible thread distinguishing what is rigorously scrutinized as scientific evidence before admission [application of the Frye rule] from what is more generously received as relevant expert opinion [without testing under the Frye rule], it is that any technique that in its application is likely to have an enormous effect in resolving quickly a matter of controversy must be demonstrably reliable. Where, on the other hand, an expert opinion only helps a trier to interpret the evidence or is susceptible to evaluation from the trier's own knowledge, it will be received on a lesser showing of scientific certainty. Because "science" is often accepted in our society as synonymous with truth, there is a substantial risk of overweighting by the jury. The rules concerning scientific evidence [including the Frye rule] are aimed at that risk.

1 M. Udall J. Livermore, Law of Evidence § 102, at 212 (2nd ed. 1982), quoted in Collins, supra, 132 Ariz. at 199, 644 P.2d at 1285. We also stated in Collins that the

Frye test . . ., with all its shortcomings, serves the "salutary purpose of preventing the jury from being misled by unproven and ultimately unsound scientific methods."

Collins, supra (quoting People v. Shirley, 31 Cal.3d 18, 53, 181 Cal.Rptr. 243, 263, 641 P.2d 775, 795 (1982)).

We do not believe that the Frye test is applicable to evidence of dogtracking or scenting. The evidence here was not bottomed on any scientific theory. In fact, it appears that no one knows exactly how or why some dogs are able to track or scent, or the degree to which they are able to do so. No attempt was made to impress the jury with the infallibility of some general scientific technique or theory. Rather, this evidence was offered on the basis that it is common knowledge that some dogs, when properly trained and handled, can discriminate between human odors. Preston's testimony was premised upon this simple idea and was not offered as a product of the application of some accepted scientific process, principle, technique or device. It was offered as Preston's opinion of the meaning of his dog's reaction; that opinion was based upon Preston's training of and experience with the dog. The weight of the evidence did not hinge upon the validity or accuracy of some scientific principle; rather, it hinged on Preston's credibility, the accuracy of his past observation of the dog's performance, the extent of the training he had given the dog, and the reliability of his interpretations of the dog's reactions. It was not the theories of Newton, Einstein or Freud which gave the evidence weight; if so, the Frye test should have been applied. It was, rather, Preston's knowledge, experience and integrity which would give the evidence weight and it was Preston who was available for cross-examination. His credentials, his experience, his motives and his integrity were effectively probed and tested. Determination of these issues does not depend on science; it is the exclusive province of the jury. See State v. Wanczyk, 196 N.J. Super. 397, 399, 482 A.2d 964, 966 (1984).

Recent cases support the conclusion that the Frye rule is inapplicable. In United States v. McNiece, 558 F. Supp. 612, 615 (E.D.N.Y. 1983) the court indicated that there was a "lesser potential prejudicial impact" of dog identification evidence than one would expect from the "seemingly flawless evidence" based upon mechanical or scientific instruments. Thus, the court held that courts "need not apply as strict a standard" when considering the admission of scent identification. Id. In United States v. Gates, 680 F.2d 1117 (6th Cir. 1982), the court allowed the testimony without considering the Frye rule. In People v. Craig, 150 Cal.Rptr. 676, 86 Cal.App.3d 905 (1978), the California appellate court held that the rule requiring general scientific acceptance was inapplicable to dog tracking evidence. The court noted that the rule dealt with "the demonstrable reliability of a newly developed technique," while dog tracking testimony was based upon a foundation of individual ability of the dog and its handler and the consequent reliability of the results. Id. at 682, 86 Cal.App.3d at 914.

Dog tracking evidence has been considered in approximately thirty-five or thirty-six states. It has been admitted, with proper foundation made on an individualized, case-by-case basis, in thirty of those states. See Terrell v. State, 3 Md. App. 340, 239 A.2d 128 (1968) for a history of dog tracking cases and review of the authorities; see also Annot., 18 A.L.R.3d 1221 (1968). In none of those states has it been thought necessary to adopt the Frye rule as a preliminary standard for determining foundation. In those few states that do not permit the testimony, inadmissibility is not based upon an application of the Frye rule; instead, the courts base their ruling either on the fear that the jurors will be misled by folklore superstitions that attach to bloodhounds and their ability to track ( see People v. McPherson, 85 Mich. App. 341, 345, 271 N.W.2d 228, 230 (1978)) or upon the very lack of a scientific basis for such evidence ( see Terrell v. State, 239 A.2d at 136).

The majority view is that admission of evidence based upon dog scenting or tracking is proper on individualized, foundational showings made on a case-by-case basis rather than on a general application of the Frye test. A proper foundation should show:

1) that the dog is of . . . [a breed and pedigree] characterized by acuteness of scent and power of discrimination, (2) that the dog has been accustomed and trained to pursue the human track, (3) that the dog has been found by experience in actual cases to be reliable in such tracking, (4) that the dog was placed on the trail at a spot where the alleged participant or participants in the crime were known to have been, and (5) that the dog was placed on the trail within the period of his efficiency.

United States v. Gates, 680 F.2d at 1119, citing People v. Centolella, 61 Misc.2d 726, 305 N.Y.S.2d 460 (Oneida County Ct. 1969); State v. Wanczyk, supra. In Gates the court considered whether these foundational requirements had been met by evidence that was almost identical to that presented to the court in the case at bench. In Gates, as in the case at bench, Harass II had been used for "scent identification" rather than tracking. The foundation shows that Harass II is a German shepherd that received extensive training for a period of four years in Germany and was then purchased by his American handler, John Preston. Preston testified in the case at bench to the years of experience that he has had with the dog. They have worked on hundreds of cases involving tracking, and hundreds more involving scent identification. Preston testified about the procedures he employs with the dog and the factors (such as "blind line-ups") which prevent the handler from giving cues to the dog. While much of this testimony was attacked, we believe the cross-examination goes to the weight of the evidence and not to its admissibility. We believe the foundation here was adequate to allow the court to exercise its discretion under Rule 403 of the Rules of Evidence and allow the jury to hear the testimony. The matters which defendant claims destroy reliability, such as the delay in putting the dog on the scent, the different alert responses, and the dog's inexperience with bicycle identification and the like, go to the weight of the testimony and not to its admissibility. We note, also, that all cases which have considered identification testimony from this particular dog have reached the same conclusion. See United States v. Gates, supra; United States v. McNiece, supra; Dedge v. State, 442 So.2d 429 (Fla.App. 1983); Epperly v. Commonwealth, 224 Va. 214, 294 S.E.2d 882 (Va. 1982).

We hold, therefore, that dog tracking or identification evidence is admissible in Arizona upon a proper foundational showing that the breeding, training, performance and handling of the particular dog warrants an inference that the results obtained from use of that dog are reliable. State v. Coleman, 122 Ariz. 130, 593 P.2d 684 (App. 1978), aff'd on other grounds 122 Ariz. 99, 593 P.2d 653 (1979). We are aware that many of the cases, and at least two of those involving Harass II, have indicated the advisability of instructing the jury to be cautious in evaluating and using such evidence. See Gates, 680 F.2d at 1119; McNiece, 558 F. Supp at 616-17. We express no opinion on the propriety or advisability of such an instruction, but we do caution trial judges to use care in determining admissibility under Rule 403 of the Rules of Evidence. There is some likelihood that jurors may give such evidence considerable weight; consequently, care should be taken to see that the foundation does indicate that the results from use of the dog are reliable. Demonstrations, in the courtroom or on film, to verify the dog's abilities might be advisable. In the case at bench, the foundational requirement was met. The question of reliability and integrity was fully explored. More importantly, the several different tests utilizing the dog produced consistent results. In each test, the dog "identified" the defendant as the perpetrator. Such consistency can be explained only by remarkable coincidence or manipulation of the tests by law enforcement officers. The latter was suggested to the jury by cross-examination and was evidently rejected by them. Although it would have been better if the trial court had required some independent verification of the dog's abilities, we cannot hold that it erred in admitting Preston's testimony. The Statement

Having been given Miranda warnings, defendant voluntarily came to the police station for an interview on June 17, 1982. He denied any involvement in the murder and gave the police a detailed description of his whereabouts on that day. His statement regarding his arrival time at his friend's house before the party near the lake was not supported by his friends. On the following day, the police appeared at defendant's place of employment and presented him with a court order requiring that he submit to temporary detention for the purpose of giving samples of hair and blood, a procedure permitted by A.R.S. § 13-3905. Before accompanying the police, defendant asked to call his lawyer and was told that the lawyer could not interfere with the court order. See State v. Curiel, 130 Ariz. 176, 634 P.2d 988 (App. 1981) (holding that there is no right to counsel during such temporary detention). After defendant had submitted the required samples, he was questioned again by the police. A Miranda warning was not given on this occasion. Defendant made a statement in which he accounted for the time discrepancies by explaining that he had stopped at a convenience store and had had car trouble on the way back. The place where he said he stopped his car to check on the stuck throttle and where he then spent several minutes happened to be directly across a field from the point of abduction. Thus, in his June 18 statement, defendant admitted having been near the scene of the crime. Defendant claims that admission of the statement violates his sixth amendment right to counsel. See Edwards v. Arizona, 451 U.S. 477, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981). Defendant concedes that counsel need not be appointed upon execution of an order under A.R.S. § 13-3905, but claims that he did have the right to call his own attorney. See McNutt v. Superior Court, 133 Ariz. 7, 648 P.2d 122 (1982).

We do not believe the court erred in admitting the statement. First, we note that the statement was not admitted over objection. It was, in fact, not offered by the state at all but by defense counsel. Nor was this a case where the defense attempted to "draw the sting" by offering evidence after the trial court had overruled the defendant's motion to suppress. The statement in question provided an explanation for the discrepancies in arrival time between the testimony of the alibi witnesses and defendant's first statement (that of June 17) which he gave after his Miranda warnings and without invoking any of his rights. The June 18 statement thus had substantial exculpatory value and was offered by the defense for that reason.

Further, even if the statement had been offered by the state, we see no error in its use. At the suppression hearing, defendant admitted that at the time he made the statement, he remembered the Miranda rights which had been read to him on the day before, knew that he was not under arrest, knew that he could leave the police station, and knew that he could call a lawyer. He made the statement because he was afraid that if he did not do so, the police would arrest him.

The procedures authorized by A.R.S. § 13-3905 have been held constitutional. State v. Grijalva, 111 Ariz. 476, 533 P.2d 533, cert. denied, 423 U.S. 873, 96 S.Ct. 141, 46 L.Ed.2d 104 (1975). Defendant's request to call his lawyer came in connection with the execution of the order for temporary detention. The officers did not tell him that he had no right to talk to a lawyer, but only that a lawyer could not prevent the execution of the order. This was not a misstatement of the law. State v. Curiel, supra. At the time of the second statement defendant knew that he was not under arrest, had the right not to talk and the right to leave if he wished; therefore, the interview conducted at the police station does not rise "to the level of a custodial interrogation." State v. Garrison, 120 Ariz. 255, 256, 585 P.2d 563, 564 (1978). The fear of arrest is not alone sufficient to "apply [the] coercive pressures" that make the interview "equivalent to custodial questioning." United States v. Pratt, 645 F.2d 89, 90 (1st. Cir. 1981).

Gruesome Pictures

Defendant claims that the trial court erred in admitting two autopsy photographs of the victim. These photographs showed the face and the neck, with the ligature strangulation marks being clearly visible. In State v. Chapple, 135 Ariz. 281, 660 P.2d 1208 (1983), we held that where offered exhibits have a capacity to incite passion or inflame the jury,

the court must go beyond the question of relevancy and consider whether the probative value of the exhibit outweighs the danger of prejudice created by admission.

Id. at 288, 660 P.2d at 1215. We do not find the photographs at issue here to be particularly gruesome. Cf. State v. Chapple, supra. They are certainly not gruesome enough to create a risk of inflaming the jury, particularly where the crime committed was so atrocious that photographs could add little to the repugnance felt by anyone who heard the testimony. State v. Piskorski, 177 Conn. 677, 419 A.2d 866, 880-81 (1979). More importantly, the photographs tend to show similarity between the ligature marks on the victim in California and those on the victim in the present case. We find no abuse of discretion in their admission.

Defendant also alleges that the prosecutor held up one of the photographs at issue here while he made this closing remark to the jury:

Who was it that turned this beautiful little girl into this asphyxiated condition?

It is by no means clear from the record that the photograph was actually used in this manner. No objection was raised at the time. Later, in his motion for a new trial, defendant alleged that such use had occurred. (Instruments at 71 g-h.) This allegation was not specifically rebutted in the state's response to the motion. Given the equivocal nature of the facts, we cannot say that reversible error occurred. Our conclusion is strengthened by the observation that the photographs are not particularly gruesome. Cf. State v. Clawson, 270 S.E.2d 659, 673-74 (W.Va. 1980) (utilization in closing argument of extremely gruesome pictures of headless, decomposed torsos held reversible error). Given the strength of the other evidence, the prosecutor's actions certainly did not "rise to the level of substantial prejudice." Pittman v. United States, 375 A.2d 16, 19 (D.C.App. 1977).

The Refusal To Grant A Mistrial

Defendant claims that the court abused its discretion in failing to grant his motion for a mistrial. The motion was made after a prosecution witness, answering a question about the area where the prior bad act had been committed, indicated that another murder had previously occurred in the area. Defense counsel objected, pointing out that there was no foundation for such testimony, and the court sustained the objection. There was no further mention of this subject. Defendant moved for a mistrial, arguing that the testimony implied that defendant had been involved in a prior murder. The court denied the motion. Testimony of this type merits the granting of a mistrial only where it might have materially influenced the jury. State v. Grier, 129 Ariz. 279, 630 P.2d 575 (App. 1981). We do not believe that the testimony in question was so closely connected to defendant that we can say that it had any material impact upon the jury. Id. Defendant did not request a curative instruction, and we do not believe the trial court abused its discretion in failing to grant the mistrial. See State v. Christensen, 129 Ariz. 32, 38, 628 P.2d 580, 586 (1981).

Identification Of Defendant's Vehicle

A witness identified defendant's automobile and testified that he had seen the car leaving the area of the crime at a high speed. The witness had originally told the police that the car he had seen was a 1978 or 1979 Ford. He later identified defendant's 1967 Rambler from a picture shown to him by the police. Before showing him the picture, the police told the witness to disregard the condition of the car and its wheels. The reason for these instructions, though not explained to the witness, was that defendant had sold the car about a week after the crime and the new buyer had cleaned up the outside and repainted the wheels.

Defendant claims that the police should have utilized a line-up procedure for the identification. We have mandated the use of such procedures with respect to eyewitness identifications. State v. Dessureault, 104 Ariz. 380, 453 P.2d 951 (1969). Defendant asks that we extend this rule to identification of inanimate, physical evidence. We decline to do so. By the great weight of authority, the right to pretrial identification procedures is inapplicable to items of physical evidence. See People v. Coston, 40 Colo. App. 205, 576 P.2d 182 (1978), aff'd., 633 P.2d 470 (1981); Klase v. State, 346 A.2d 160 (Del. 1975); State v. Bruns, 304 N.W.2d 217 (Iowa 1981). Factors surrounding automobile identifications, such as the suggestiveness of the proceedings, affect only the weight, and not the admissibility, of the evidence. Buchanan v. State, 561 P.2d 1197 (Alaska 1977). As the court stated in State v. King, 31 Wn. App. 56, 639 P.2d 809 (1982):

The famous trilogy of cases that point out the dangers of a show up identification only address their concerns to the proper identification of the person (citations omitted). . . . [A] witness who identifies a piece of clothing which is essentially the same in appearance as hundreds of thousands of others is not only more likely to change his mind, but his credibility before the trier of fact may be readily challenged.

Id. at 59, 639 P.2d at 811-812. We think the same reasoning is applicable to identification of cars. The differences between a 1967 Rambler and a 1978 Ford are easy to convey to the jury. Reasonable cross-examination could easily have elicited the admission that when his memory was unaided by police the witness recalled having seen a 1978 Ford and that after seeing photographs of a single car, he changed his mind and identified a 1967 Rambler. We believe that it was for the jury to decide whether the identification in court was dependable and accurate.

Effective Assistance of Counsel

Defendant claims that he was denied effective assistance of counsel during the sentencing hearing. He contends that his trial counsel failed to argue the mitigating factors of age (defendant was twenty at the time of the crime) and close family ties, and also that counsel failed to argue that there was no proof of defendant's intent to kill.

The standard of "minimal competence" was established by this court in State v. Watson, 134 Ariz. 1, 653 P.2d 351 (1982). In State v. Carriger, 132 Ariz. 301, 645 P.2d 816 (1982), we held that the requirement of effective assistance of counsel was applicable to the sentencing phase of a capital case. We stated that defense counsel falls below the standard of minimal competence where he fails to challenge the admission of aggravating evidence when it is reasonably possible to do so or fails to present available mitigating evidence. Id. at 304, 645 P.2d at 819.

In the case at bench, defense counsel did not argue either defendant's youth or his close ties with family and friends as mitigating circumstances. However, these matters were clearly contained in the information before the trial judge. No doubt there are situations where trial counsel's failure to point out the obvious would amount to a lack of minimal competence, but even if we were to assume that there was a breach of the minimum standard of competence here, there would be no reason to reverse. We have recently held that the Watson rule will not require reversal unless the defendant can show a reasonable probability that a different result would have been reached had his counsel's performance been up to the required standard. State v. Lee, 142 Ariz. 210, 689 P.2d 153 (1984). We do not believe that such a probability could be established here. The information with regard to both of these mitigating circumstances was contained in the file and was known to the trial judge; he stated that he had considered all the mitigating factors and evidence in the file but that these were not "sufficiently substantial to call for leniency." Special Verdict, filed February 25, 1983. See A.R.S. § 13-703(E). The circumstance most important from the standpoint of mitigation was age, and the judge specifically stated that he had considered and rejected it as a mitigating factor. In light of this, we cannot believe that there is any probability that further argument by defense counsel would have produced a different result. We have held that while age is a relevant mitigating factor, it cannot be considered in a vacuum. Defendant's criminal character and history must also be considered. State v. Williams, 134 Ariz. 411, 413-14, 656 P.2d 1272, 1274-75 (1982); see also State v. Johnson, 131 Ariz. 299, 305, 640 P.2d 861, 867 (1982).

We turn, then, to the assertion that counsel fell below the minimal standard when he failed to argue that defendant's intent to kill had not been established. Under the facts of this case, we are not persuaded that the failure to argue the lack of intent to kill falls below the standard. Intent must usually be inferred from the nature of the act committed or performed. Here defendant gagged the victim, took her elastic hair band, choked her with it and left her in a remote area in the early evening with no prospect of being found until the morning. The ligature was left so tight that it took the victim only ten minutes to die. These facts make any argument as to defendant's benign intent totally untenable. Defense counsel is not required to argue the absurd or the impossible. Even if he were, it is hard to see how, under these facts, any argument regarding lack of intent would have prevailed. State v. Lee, supra.

We hold that the case need not be reversed for ineffective assistance of counsel.

The Child Molesting Counts

Defendant claims that the evidence was insufficient to support his conviction of two counts of child molesting. The indictment charged two counts of touching the "private parts" of a child. A.R.S. § 13-1410. The physician who performed the autopsy could not state whether the victim's vagina had been penetrated more than once. However, there was evidence of oral sexual contact. Conceding that the vaginal penetration will support conviction of one count of child molesting, defendant argues that the evidence of oral sexual contact cannot support the other count because the counts charged only touching the "private parts" of the child.

We note, however, that the jury was instructed that the crime occurs when a defendant touches a child's "private parts" or causes a child to touch his. Thus, both theories available to support the charge were before the jury. No objections were made to the jury instruction. There was evidence that both types of child molestation occurred in this case. Defendant did not object to evidence of oral sexual contact on grounds of irrelevancy; neither did he object to his indictment on two counts of child molesting until the case was before this court. The objection comes too late. State v. Bailey, 125 Ariz. 263, 609 P.2d 78 (App. 1980). The indictment is deemed amended to conform to the evidence adduced at trial. Rule 13.5(b), Ariz.R.Crim. Proc., 17 A.R.S. Other Issues

Defendant claims that the Arizona death penalty procedure violates the Sixth, Eighth and Fourteenth Amendments to the United States Constitution because it constitutes cruel and unusual punishment, denies defendant the right to jury trial on the issue of sentencing, places the burden on defendant to prove mitigating circumstances, fails to offer guidelines for weighing aggravating and mitigating circumstances, and does not sufficiently define the words used in A.R.S. § 13-703. We have previously rejected similar claims. See State v. Harding, 137 Ariz. 278, 670 P.2d 383 (1983), cert. denied, ___ U.S. ___, 104 S.Ct. 1017, 79 L.Ed.2d 246 (1984); State v. Lambright, 138 Ariz. 63, 673 P.2d 1 (1983). We continue to reject these arguments.

Defendant argues that the denial of the right to jury trial during the sentencing portion of a capital case violates article 2, § 23 of the Arizona Constitution. Defendant points out that at the time the constitution was adopted Arizona followed a procedure whereby the decision on imposition of capital punishment was left to the jury. He argues that, therefore, the right to have the jury decide upon the question of capital punishment attached when the constitutional provision was originally adopted and that the subsequent legislative change in procedure was invalid. We note, however, that the procedure which left the question of life or death to the jury at the time our state constitution was adopted was statutory in nature. See Penal Code of 1901, § 174. At common law, the right of trial by jury was limited to the question of guilt or innocence. The decision on punishment, within the statutory limits, was left to the judge. We believe that the right of jury trial protected by article 2, § 23 refers to the common law right of jury trial on the question of guilt or innocence and that no defendant has a constitutional right to be tried according to any particular procedure being followed at the time the constitution was adopted. See Rothweiler v. Superior Court, 1 Ariz. App. 334, 339, 402 P.2d 1010, 1015 (1965) aff'd, 100 Ariz. 37, 410 P.2d 479 (1966); State v. Cousins, 97 Ariz. 105, 106, 397 P.2d 217, 219 (1964). See also Spaziano v. Florida, ___ U.S. ___, ___, 104 S.Ct. 3154, 3165, 82 L.Ed.2d 340 (1984) (sentencing by jury not required by the federal Constitution). To hold otherwise would be to freeze procedures to those recognized and followed in 1910 and to embed them in the constitution. We do not believe that this was the intent of the framers of the constitution.

Propriety Of Imposing Death

Defendant argues that the death penalty was improperly imposed. We have a duty independently to review the existence of aggravating or mitigating circumstances and to determine whether the death penalty was improperly imposed and should be reduced to life. State v. Richmond, 114 Ariz. 186, 560 P.2d 41 (1976), cert. denied, 433 U.S. 915, 97 S.Ct. 2988, 53 L.Ed.2d 1101 (1977).

By special verdict [A.R.S. § 13-703(D)] the trial court found as an aggravating circumstance that the murder was committed in an especially heinous, cruel or depraved manner. It found no mitigating circumstances sufficiently substantial to outweigh this aggravating circumstance. We agree that the killing was especially cruel because the facts show infliction of both mental distress and physical pain upon the victim. State v. Lambright, supra. Cruelty is established beyond any doubt.

The words "heinous" and "depraved" refer to the nature of the crime and the state of defendant's mind. Abduction, violent sexual penetration and strangulation of a helpless seven year old child are circumstances that lead to only one conclusion. The senseless killing and the entire nature of the attack are repugnant to civilized society. The elements of a heinous crime and a depraved state of mind are present. See State v. Gillies, 135 Ariz. 500, 662 P.2d 1007 (1983); State v. Zaragoza, 135 Ariz. 63, 659 P.2d 22, cert. denied, 462 U.S. 1124, 103 S.Ct. 3097, 77 L.Ed.2d 1356 (1983); State v. Ortiz, 131 Ariz. 195, 639 P.2d 1020 (1981), cert. denied, 456 U.S. 984, 102 S.Ct. 2259, 72 L.Ed.2d 863 (1982).

In light of defendant's past history, his prior crime and the nature of the present offense, we do not believe defendant's age or relationship with family or friends are sufficient mitigating circumstances to outweigh the cruel, heinous and depraved nature of this crime. The fact that a felony murder instruction was given is not a relevant circumstance in this case. Defendant is not being held accountable for the acts of an accomplice. Cf. State v. Tison, 142 Ariz. 446, 690 P.2d 747 (1984). There is no doubt that it was defendant and not an accomplice who killed Laura. It is difficult to infer from the facts on this record that defendant had any intent other than to kill the victim. The trial court specifically found this in its special verdict. We believe that finding is supported by overwhelming evidence. Our review does not lead us to the opposite conclusion.

Proportionality Review

We must also examine the concept of proportionality to determine whether imposition of death violates the Eighth Amendment. The question is whether the death penalty imposed upon this defendant is excessive or disproportionate to the penalty imposed in similar cases. See State v. Richmond, 114 Ariz. 186, 196, 560 P.2d 41, 51 (1976).

We have reviewed our other cases and find that the imposition of death here is not disproportionate. See State v. Gillies, supra; State v. Mata, 125 Ariz. 233, 609 P.2d 48, cert. denied, 449 U.S. 938, 101 S.Ct. 338, 66 L.Ed.2d 161 (1980).

We have searched the record for fundamental error according to the mandate of A.R.S. 13-4035 and have found none. The judgment of conviction is affirmed. The sentence of death is approved.

HOLOHAN, C.J., GORDON, V.C.J., and HAYS and CAMERON, JJ., concur.