Appellant, Kenneth S. Cobey, was convicted by a jury in the Circuit Court for Montgomery County of rape, several counts of sexual offense, robbery, and theft of over $300. All of the charges stem from an incident which occurred on 4 September 1985. On that evening, a young woman drove her 1985 blue Subaru automobile to Northwest Branch Park. She parked her car and went for a walk on a jogging trail which leads through the woods. While she was on the trail, she heard a man coming at her from behind, so she stepped aside to let him pass. Instead, the man grabbed her, knocking her down and off of the trail. He threatened to kill her if she screamed. He then pushed her into the woods, away from the trail, where he forced her to have oral sex with him, raped her, and had anal sex with her, all against her will. After the ordeal, he took the keys to her car and disappeared, leaving her in the woods. Later that night, the victim returned to the park with the police and her car was gone.
On 27 September 1985, the Washington, D.C. рolice set up a traffic observation checkpoint on Kennedy Street in the District of Columbia. Appellant Cobey was driving a 1985 blue Subaru that night and was ordered by the police to pull over and stop at the checkpoint. The officer who ordered appellant to stop testified that he did so because the car *236 appellant was driving failed to display an inspection sticker and a front license plate, both of which are required by Washington D.C. law. The officer аsked to see appellant’s driver’s license and a registration for the car. Appellant produced a valid Maryland license, but no registration. According to the officer, appellant said that the car belonged to a friend, but was unable to tell the officer his friend’s last name or address. An initial computer check of the number of a Washington, D.C. license plate on the rear of the car failed to produce any information, so the police impounded the car for further investigation by a police auto theft unit. Appellant Cobey was given two traffic tickets and allowed to leave. Subsequent investigation revealed that the car appellant had been driving was the victim’s 1985 blue Subaru which had been taken from Northwest Branch Park 23 days earlier. Appellant was arrested by Montgomery County Police on 30 September 1985.
During the first week of October 1985, the victim learned that she was pregnant. She testified that the only possible source of the pregnаncy was the rape of 4 September 1985. She procured an abortion on 21 October 1985. With her permission, the police took possession of the aborted fetus. The fetus and blood samples from the victim and appellant were flown to Portland, Oregon for testing to determine whether appellant might be the man who fathered the fetus during the attack. Dr. Susan Olson, a cytogeneticist at the Oregon Health Sciences University, performed a technique known as Chromosome Variant Anаlysis (“C.V.A.”) on cell tissues from the fetus and on both of the blood samples. Like blood tests, C.V.A. cannot prove that a man is the father of a child. It can only determine whether he is possibly the father. According to Dr. Olson, the C.V.A. performed in this case did not exclude the possibility that appellant fathered the victim’s fetus.
In July of 1986, appellant was brought to trial for the September 1985 attack, but this resulted in a mistrial. Nevertheless, at this first trial, Judge Irma Raker had denied appellant’s motion to exclude Dr. Olson’s testimony *237 before the mistrial was declared. At appellant’s second trial, in December of 1986, Judge Richard Latham declined to relitigate the issue and permitted Dr. Olson to testify over appellant’s objection. We know of no other Maryland case which has admitted testimony based on C.V.A.
Appellant presents three issues on appeal:
I. Did the lower court err in admitting expert testimony of chromosome variant analysis where the state failed to establish that the analysis was reliable and generally accepted?
II. Did the lower court err in admitting evidence obtained through seizure of an automobile driven by appellant where the seizure resulted from an unreasonable traffic roadblock?
III. Did the lower court err in permitting the introduction of mug shot photographs whose prejudice outweighed any minimal relevance?
I.
Appellant contends that C.V.A. has not been generally accepted as reliable in the relevant scientific community and therefore that testimony derived from the results of C.V.A. is inadmissible under
Reed v. State,
*238
Under the
Fry e-Reed
test, the proponent of a new scientific test bears the burden of producing evidence to establish the technique’s general acceptance.
Thompson v. Thompson,
In Reed, supra, the Court of Appeals did not address the issue of whether appellate review should be limited to consideration of the materials in the record. But, in holding that the trial court’s decision to admit expert testimony based on spectrography (voiceprint analysis) was erroneous, the Court seems to have gone beyond the record:
Thus, based on our examination of the record in the instant case, the judicial opinions which have considered this question, and the available legal and scientific commentaries, we do not believe that “voiceprint” аnalysis has achieved the general acceptance in the scientific community, at this time, which is required under Frye. [Emphasis added.]
Reed,
We also believe that the standard of review applicable to the trial court’s finding of general acceptance is whether it is against the weight of thе evidence rather than whether it is clearly erroneous. In Reed, supra, the Court of Appeals recognized that the issue of whether a scientific technique was sufficiently reliable for use at trial is not a traditional question of fact:
The question of the reliability of a scientific technique or process is unlike the question, for example, of the helpfulness of particular expert testimony to the trier of facts in a specific case. The answer to the question about the reliability of a scientific technique or process does not vary according to the circumstances of each case. It is therefore inappropriate to view this threshold question of reliability as a matter within each trial judge’s individual discretion. Instead, considerations of uniformity and consistency of decisionmaking require that a legal standard or test be articulated by which the reliability of a process may be established.
Wе also note that a standard of review which created a presumption of correctness in the trial court’s finding of general acceptance would be inconsistent with our understanding that the appellate court may consider evidence which was not presented to the trial court.
*240
The need for an independent appellate determination of general acceptance was illustrated in
Harker v. State, 55
Md.App. 460,
In the case at bar, Judge Raker held a hearing out of the jury’s presence to determine whether Dr. Olson would be allowed to testify about the results of the C.Y.A. performed on the victim, appellant, and the fetus. At the hearing, Dr. Olson testified that medical genetics is the study of diseases associated with inherited characteristics and that cytogenetics is a s.ubspecialty concentrating on chromosome structures and abnormalities. She described chromosomes as composed of DNA and other associated proteins. Their function is to transmit units of genetic information from cell to cell (within a person), and from one generation of people to the next. For every person, the sets of chromosomes in each cell are identical: they do not vary from cell to cell within a person. But they can vary from one person to another. The chromosome sets consist of 46 сhromosomes existing in 23 distinctive pairs. For each pair, one chromosome is donated by the subject’s mother and the other by the father.
*241 Dr. Olson further explained that each chromosome contains an area known as the heterochromatin, which is a part of the DNA protein structure. Scientists are not aware of any genetic function that heterochromatin serves. Therefore, the amount and shape of the heterochromatin of a particular chromosоme may vary from person to person without affecting any noticeable genetic trait. The different shapes of heterochromatin are called heteromorphisms, or “variants.” Barring mutations, which Dr. Olson claimed are rare in heterochromatin, a person inherits his variants from his parents. In other words, a particular chromosome will exhibit the same variant in the child as it does in the parent who contributed it.
C.V.A. is a process through which cytogeneticists look at chromosome variants. Dr. Olson explained that a cell culture is taken from a person and set aside. The cells are chemically treated so that they will stop developing at an optimum point for viewing. The cells are then dropped onto a slide, where they burst open and the chromosomes spill out onto the slide. A stain is then applied to the chromosomes which Dr. Olson claimed allows cytogeneticists to identify not only different chromosomes, but also regions of those chromosomes. She uses a stain called quinacrine, which she says enables her to clearly view the variants of chromosomes 3, 4,13, 14, 15, 21, 22, and the Y chromosome. The stained chromosomes are photographed using a microscope. The chromosomes are three-dimensional and land in various positions on the slide. Therefore, for each person, pictures of several of each of the chromosomes (i.e., several of the person’s chromosome 3’s) are рhotographed, so that cytogeneticists can see what those particular variants look like from different angles. The photographs are enlarged. Individual chromosome pairs are cut out of the pictures and arranged into karyotypes, which contain slots for each of a person’s 46 chromosomes. Cytogeneticists then can compare the chromosomes of different people. They can, for example, look at the chromosomе 3’s of a child and compare them to the chromosome 3’s of the mother. The variants of *242 one of the child’s chromosome 3’s will match those of one of the mother’s, since it was contributed by her. That will mean that the other of the child’s chromosome 3’s came from his father, and its variant will match the variant of one of the father’s pair of chromosome 3’s.
Where paternity is in dispute, as it is here, cytogeneticists use C.V.A. to compare the variants of the child to the mothér, determining which сhromosomes the mother contributed. The child’s remaining chromosomes are then compared to an alleged father. If each of the child’s variants matches a variant of one of the man’s chromosomes from that pair, then the man could have contributed all of the chromosomes. If any one of the child’s non-maternal variants does not have a match among the father’s variants for that chromosome pair, then the man is excluded from paternity. Dr. Olson testified that she supervised a study in which the variants of 57 Caucasian people were all compared with each other. The study used quinacrine staining and concentrated on chromosomes 3, 4, 13, 14, 15, 21, 22, and the Y chromosome. No two people were found to have the same variants on all of the chromosomes. Dr. Olson concluded, based in part on this study, that it is very unlikely that two people (other than identical twins) will have the same set of variants. She estimated the probability to be .0003 (.03%). Dr. Olson summarizеd and discussed her findings in her article, “Human Chromosome Variation: The Discriminatory Power of Q-band Heteromorphism (Variant) Analysis in Distinguishing between Individuals, with Specific Application to Cases of Questionable Paternity,” 38 AmJ.Hum. Genetics 235 (1986). This article was presented into evidence by appellant. The article also described a C.V.A. study conducted on twelve mother-father-child trios. The mother-child karyotypes were kept together. The karyotypes of the fathers were then compared to each mother-child combination. (The fathers’ karyotypes had been “coded” so that the cytogeneticist making the comparisons would not know which man actually was the father of the child being compared.) Dr. Olson reported that this study *243 resulted in 120 cases of alleged paternity. All 108 instances where the “alleged father” was not the natural father were correctly excluded. All 12 cases of actual paternity were correctly interpreted as such. The article cоncludes that the probability of excluding a wrongfully accused man using quinacrine C.V.A. in paternity disputes is 1.000 (100%). Olsen, et al., 88 AmJ.Hum. Genetics at 249. The article then admits that “[o]ur figure is in contrast to Niebuhr and Gurtler’s figures of .72 for females and .74 for males.” Id. This is a reference to a 1981 study by two scientists which concluded that the probability of excluding a wrongfully accused man would only be .72 (72%) where the child is female and .74 (74%) where the child is male. At the hearing, Dr. Olson was critical of the Niebuhr-Gurtler study, but was not aware of any published commentary оn her article which would have shown support for her results. We have also been unable to find any, and were told of none at oral argument. The Niebuhr-Gurtler study indicates that there may be some dispute as to the reliability of C.V.A. in paternity disputes. Dr. Olson’s criticism of that study may be justified, and she may have perfected C.V.A. to a point near 100% reliability. But that is not for the courts to decide. Reed requires that the scientific community make that judgment.
Dr. Olson did testify that her beliefs about the reliability of quinacrine C.V.A. were based not only on her study, but on her experience аnd “the experience of the people in the laboratory and the experience of people throughout the world who have come out with such statements regarding the uniqueness of the individual.” She also answered “[y]es” to a question of whether “the way in which you arrive at your conclusion [in the case at bar] and the methodology by which you would arrive at the conclusion [are] based upon principles and techniques generally accepted as valid within the scientific community[.]” She asserted that her belief in the uniqueness of the individual’s variants was “shared by a number of different [cytogeneticists] that are familiar with this particular testing.” She also said that *244 there “have been many cases throughout the world” in which C.V.A. results have been admitted by courts to establish paternity. Two of these cases were in Oregon, but Dr. Olson gave no citations to American court opinions on the subject.
Dr. Olson did not refer to any published articles supporting her theories or name any of the other cytogeneticists she claims share her views. Her statements are vague assertions of general acceptance, but nothing more. The State did not produce any journal articles or textbooks which show that C.V.A. is believed to be as reliable in paternity cases as Dr. Olson asserts. The Niebuhr-Gurtler study found it less reliable than Dr. Olson’s study, and less reliable than required to be used to establish paternity in a civil case in Maryland under Md.Fam.Law Code Ann. § 5-1029 (1984) (test must exclude at least 97.3% of wrongfully accused fаthers). We have not been able to find any scientific commentary in the record or outside of it, which discusses Dr. Olson’s 1986 article or her belief in the reliability of C.V.A. in establishing paternity. 1 We are therefore not satisfied that the State has met its burden under the Frye-Reed test.
At appellant’s second trial, Dr. Olson was allowed to testify that C.V.A. showed that the non-maternal variants of the fetus all matched variants exhibited by appellant’s chromosomes. She then testified that it was highly unlikely that anyone else had the same variants as appellant. She stated that probability to be “.003.” (As noted earlier, Dr. Olsen’s published study estimated a probability of .0003
*245
that two people would have identical sets of variants.) The only issue at trial was the identity of the assailant. Dr. Olson’s testimony, if believed, could have led the jury to infer that appellant was possibly the father of the fetus and that there was only a .003 probability that anyone else could possibly be the father. Because we are not satisfied that the basis of this testimony was shоwn to be generally accepted as reliable in the relevant scientific community, appellant’s convictions must be reversed. On remand the State will not be permitted to introduce evidence which is based upon C.V.A. As the Court of Appeals indicated in
Reed, supra,
Because the other issues appellant raises here are likely to appear again in a retrial, we will address them.
II.
Appellant contends that the Washington, D.C. Police stopped him illegally on 27 September 1985 and that all evidence derived from the stop should have been excluded as “fruit of the poisonous tree.” We disagree with the premise and thus the conclusion.
Appellant’s assertion that the stop was illegal is grounded on the requirements for a valid roadblock set out in
Little v. State,
III.
Appellant contends that the trial court abused its discretion by allowing the jury to view “mug shots” of him from six different arrests. Although the pictures were “sanitized” by cutting off the chest plates appellant was holding in them, they “still showed the front and profile view commonly associаted with police ‘mug shots.’ ”
Area v. State,
In the case at bar, the “mug shots” did have probative value. The assailant’s identity was at issue. The victim testified that she did not notice any facial hair on the assailant. Appellant’s mother, sister, and a young friend of his named Serena Locust ¿11 testified that he had facial hair throughout the summer of 1985. Appellant’s mother further stated that he has had facial hair at all times since at least the 1970’s. The “mug shots” had been taken at various times from 26 September 1979 through 12 November 1985. In all of them, appellant exhibits some facial hair, though in varying amounts. In some it is barely visible. The three witnesses were shown the photos and agreed that they showed varying amounts of facial hair. Serena Locust observed that “in some of them he doesn’t have nothing on his face.” This shows that, at times, it is *247 difficult to notice appellant’s facial hair even upon close examination. An investigator from the Public Defender’s office testified that the victim had said she did not get a good look at her assailant. The “mug shots” dated 30 September 1985 and 12 November 1985 show that appellant’s facial hair was darker in early November than in late September. As such it could have been very light in early September, 1985.
When appellant produced witnesses to state that he had facial hair in September, 1985, he was implying that he could not have been the assailant, who had no noticeable facial hair at that time. The “mug shots” thus became relevant to show that even though appеllant did have facial hair at the time of the attack, it may have been so light that one in the victim’s circumstances would not have noticed it. Seeing the photographs allowed the jurors to decide for themselves whether appellant’s facial hair varied and was, at times, difficult to see. They were probative of the question of identity, a question the jurors had to answer. We do not believe the decision to allow the jurors to view the pictures was a clear abuse of discretion.
JUDGMENT REVERSED AND REMANDED FOR A NEW TRIAL. COSTS TO BE PAID BY MONTGOMERY COUNTY.
Notes
. The issue of using "DNA prints” as evidence in rape trials and paternity cases was recently addressed by Sharon Begley in “Leaving Holmes in the Dust,” Newsweek, 26 October 1987 at 81. The article concerns tests using "DNA fingerprints" which are professed to be unique to the individual. It does not mention C.V.A. specifically, or Dr. Olson. The tests have been admitted at trials in England, but the author predicts that it is “too soon” for them to meet the "reliable and trustworthy" standard required for admission into evidence by most American courts. Id. One of the U.S. companies developing the tests recently opened for business in Maryland.