615 P.2d 1228
Supreme Court of Utah.
1230*1230 Jonathon H. King of Utah Legal Services, Salt Lake City, for defendant and appellant.
Robert B. Hansen, Atty. Gen., R. Paul VanDam, Salt Lake County Atty., Gerald Conder, Deputy Salt Lake County Atty., Salt Lake City, for plaintiffs and respondents.
Plaintiffs initiated this lawsuit to establish defendant's paternity of a child born out of wedlock to plaintiff Phillips and to compel defendant to support the child. The case was tried to a court sitting without a jury. The court found the defendant to be the father and ordered support payments to be paid. The central issue on this appeal is whether the trial court erred in admitting the results of a relatively new scientific test known as the HLA (Human Leucocyte Antigen) test which purportedly proved the defendant to be the father of the child in this case to a 97% degree of probability. Defendant in addition contends that the trial court's finding of paternity was contrary to the weight of the evidence and that the cumulative effect of the trial court's evidentiary rulings constituted reversible error.
We reverse and remand for further proceedings because it was prejudicial error for the trial court to admit the HLA test results without a proper foundation as to the reliability of both HLA tests in general and the particular test in this case.
The testimony of plaintiff Phillips at trial was self-contradictory and also controverted by defendant. Phillips testified that she and the defendant had had sexual intercourse with one another three to four times a week from the middle of January 1975 to March 15, 1975. She first testified that she had not had intercourse with anyone else during that period, but later admitted to having had sexual relations with another man about January 15. Her child was born full term October 14, 1975. Phillips testified that she disclosed her pregnancy to the defendant in February 1975 and that she telephoned him on Thanksgiving Day of the same year to inform him of the birth. On both occasions, she claimed, he made admissions to her concerning his paternity. At trial defendant denied paternity and testified that he had not engaged in sexual intercourse with the plaintiff Phillips at any time. He also testified that Phillips had not informed him of her pregnancy until after the child was born when she telephoned him on Thanksgiving.
Prior to trial plaintiff Phillips, the child, and defendant had blood samples taken and submitted to an HLA tissue-typing test. The test indicated that the defendant was the father of the child in question.
As we understand the HLA test, it is based on the identification and typing of antigen markers found in white blood cells and other tissues of the body.In recent years a number of different tests or systems — by one account as many as fifty — have been developed to resolve questions of disputed parentage. Wiener and Socha, 1231*1231 Methods Available For Solving Medico-Legal Problems of Disputed Parentage, 21 J.For.Sci. 42, 61 (1975). The tissue-typing test is a genetic test based upon the chromosomal makeup of the test subject. Human body cells have 23 pairs of chromosomes which carry genetic markers called HLA antigens. An antigen is a substance which can stimulate antibody production when introduced into another individual. Antigens, which are produced under genetic control by genes, have been scientifically identified and classified. The basic theory is that by identifying the antigen markers of a child and of the mother, the child's antigen genetic markers which could only be inherited from the father can generally be determined, thereby identifying the father to a high degree of certainty.This is so because, it is claimed, most people are "rare" types in the sense that only about one out of a thousand people have a similar HLA type. Therefore, a rare type that occurs in a putative father and that also occurs in a child produces a high degree of probability that the putative father is in fact the father. See Terasaki, Resolution by HLA Testing of 1000 Paternity Cases Not Excluded by ABO Testing, 16 J.Fam.L. 543, 544-45 (1977-78).
In the instant case, plaintiffs called two witnesses to establish the admissibility of the HLA test. Paula Simenson, a medical technologist with a B.S. degree in bacteriology, a chemistry minor, and 2 1/2 years' work experience, testified that she had witnessed the taking of the blood sample from the defendant. She traced the chain of custody of the blood sample. Over defendant's objection to the admissibility of the test evidence, she also explained how the test works and described the testing procedure in this particular case. Ms. Simenson conducted the laboratory work for the HLA test and prepared a work sheet which represented her findings. The work sheet was admitted into evidence over defendant's objection.
Plaintiff's second expert witness, Dr. Charles DeWitt, a pathologist, based his opinion as to the paternity of the father on the work sheet and on tables of percentages published by another person. Dr. DeWitt testified that the test has been used for approximately 15 years for "medical purposes." He did not specify for which medical purposes, although it appears that the use referred to by Dr. DeWitt was primarily for determination of tissue compatibility in organ transplantation procedures. Dr. DeWitt also testified without elaboration that the HLA test, when performed under certain conditions, is highly accurate and widely accepted.
Dr. DeWitt stated that the statistical probability that a particular man could be correctly identified as the father of a child ranged from 70% to over 90%, depending on the number of men with whom the mother had sexual intercourse at the time of ovulation. That is, assuming the mother had had sexual intercourse with 15 different men near the time of her ovulation, there would be a 70% likelihood that a person identified as the father was in fact the father. If the mother had had sexual intercourse with only two men during the same period of time, there would be a 97% likelihood that the man identified as the father by the test was in fact the father. Dr. DeWitt was not able to recall the title of the publication from which he obtained these percentages, nor did he give any information as to how widely accepted the tables were for determining paternity, what limitations or variables the tables were subject to, or the extent or nature of verification studies that had been done with respect to the tables. Although he stated that the "literature [was] full of reports" regarding the HLA test, he did not refer to any specific authority for his statements regarding the reliability of the HLA test or its alleged widespread use for determining parentage. Nor does it appear that he himself had done any 1232*1232 research in developing the test or compiling and verifying the tables showing probabilities of parentage.
Dr. DeWitt concluded that the HLA blood test in the instant case did not exclude the defendant as the father, and that, based on calculated statistical probabilities taken from tables published in a book, there was a 97% degree of probability that defendant was in fact the father of plaintiff's child.
HLA tissue typing is a comparatively new form of test insofar as its use in the courtroom is concerned, and, according to our research, its admissibility has been dealt with by only a few appellate courts. In Cramer v. Morrison, 88 Cal. App.3d 873, 153 Cal. Rptr. 865 (1979), a California court of appeals reversed a trial court's refusal to admit the results of HLA testing in a paternity action. The trial court had ruled (1) that California statutory law allowed only evidence of an alleged father's nonpaternity and not evidence affirmatively showing paternity, and (2) that statistical evidence of this nature would have a prejudicial effect on the jury which would outweigh its probative value.In an evidentiary hearing before the trial judge, evidence was adduced that the HLA test indicated a 98.3% degree of probability that the defendant was the father. The trial court found that available data indicated the test was reliable but nevertheless held the test inadmissible for the reasons stated.
The court of appeals held that California law did not require "that the admissibility of scientific-test evidence must be predicated on a 100 percent degree of accuracy."(153 Cal. Rptr. at 872.) The court also held that California statutory law did not prohibit the admission of a test affirmatively tending to prove paternity. That law is based in part on the Uniform Act on Blood Tests to Determine Paternity which provides for the admission of tests such as the Landsteiner classification of red blood cell groups into evidence to exclude paternity. The Uniform Act also permits the admission of such tests, in the discretion of the trial court, to prove probability of paternity. However, in adopting the Uniform Act, California refused to adopt the latter provision. The court of appeals in Cramer held that the omission of the latter provision did not indicate a legislative intent to bar the admissibility of all tests which affirmatively identify a father. The court also noted that at the time the California Legislature adopted the Uniform Act the HLA test was not in use for paternity testing.
Finally, the court of appeals declined to address the issue as to whether the test had received general acceptance in the scientific community and therefore met the foundational requirements for admissibility. Accordingly, the court remanded for a determination of that issue. The court stated that the issue, being one of mixed fact and law, should be determined by the trial court on the basis of expert testimony, legal and scientific publications, and judicial opinions.
We have found only two lower appellate court cases which have held that HLA tests are admissible. The Superior Court of New Jersey in Malvasi v. Malvasi, 167 N.J. Super. 513, 401 A.2d 279 (1979), held that the HLA test had received general scientific acceptance and could be used along with other evidence to determine parentage. Commissioner of Social Services v. Lardeo, 100 Misc.2d 220, 417 N.Y.S.2d 665 (1979), also held the test admissible.
The Wisconsin court of appeals in J.B. v. A.F., 92 Wis.2d 696, 285 N.W.2d 880 (1979), refused to admit the results of an HLA tissue-typing test because of Wisconsin's highly restrictive statutory approach to the use of medical evidence in paternity disputes. The court, however, suggested that a review of the limiting nature of its statute might be in order in light of "medical advances and changed social conditions." In Simons v. Jorg, Fla., 375 So.2d 288 (1979), the court refused to admit the test on grounds not having to do with its reliability, and the court did not address that issue.
1233*1233 In sum, no state court of last resort has held that the HLA test meets all the necessary foundational requirements for admission in evidence.
In this case, the threshold issue is whether the test is inadmissible under the Uniform Act on Paternity, adopted in Utah as §§ 78-45a-1 et seq. This Act expressly authorizes the use of blood tests for the purpose of excluding paternity. Section 78-45a-10 of the Act states that "[i]f the court finds that the conclusions of all experts, as disclosed by the evidence based on the tests, are that the alleged father is not the father of the child, the question of paternity shall be resolved accordingly."
Under this statute blood tests may also be used to show a probability of paternity. The above-cited section provides further: "If the experts conclude that the blood tests show the possibility of the alleged father's paternity, admission of this evidence is within the discretion of the court, depending on the infrequency of the blood type."
There are two reasons why the Utah Act constitutes no bar to the admissibility of HLA tests if they otherwise meet the appropriate criteria for establishing reliability. First, the statute was enacted with reference to blood tests based on red blood cell groupings and was not intended to apply to HLA tests, which are of a different nature.Cramer v. Morrison, supra. HLA tests are not necessarily properly characterized as blood tests. Antigens may be found in most tissues of the body, including the liver and the kidneys, as well as component parts of the blood. J.B. v. A.E., supra, 285 N.W.2d at 882. Second, even if the statute is deemed applicable, admissibility is left in the discretion of the court. Since red blood cell group tests produce relatively lower probabilities in affirmatively identifying paternity than the probabilities claimed for HLA tests, the latter, if otherwise admissible, should also be admissible. We conclude that § 78-45(a)-10 does not preclude the admissibility of HLA tests if they otherwise meet the relevant legal standards for the admission of scientific evidence.
We turn next to the issue of the legal standards which determine the admissibility of scientific evidence. The most widely used standard for making that determination was formulated in Frye v. United States, 293 F. 1013 (D.C. Cir.1923). The Frye test has been adopted by a majority of those jurisdictions in this country which have established standards to be applied in admitting scientific evidence which is new to the courtroom. Frye held that scientific tests still in the experimental stages should not be admitted in evidence, but that scientific testimony deduced from a "well-recognized scientific principle or discovery" is admissible if the scientific principle from which the deduction is made is "sufficiently established to have gained general acceptance in the particular field in which it belongs."(293 F. at 1014.)
General acceptance in the scientific community, or more specifically the particular discipline or disciplines of the scientific community which deal with the principles involved, assures the validity of the basic principle. Verification of the basic principle and its application through widespread replication and practical usage is an appropriate indicium of reliability.People v. Kelly, 17 Cal.3d 24, 130 Cal. Rptr. 144, 549 P.2d 1240 (1976). The Frye standard, however, does not demand infallibility as a condition to admitting scientific evidence.United States v. Franks, 511 F.2d 25 (6th Cir.1975); United States v. Stifel, 433 F.2d 431 (6th Cir.1970); United States v. Alexander, 526 F.2d 161 (8th Cir.1975).
Although a computation of probabilities not based on scientifically established data is inadmissible, People v. Collins, 68 Cal.2d 319, 66 Cal. Rptr. 497, 436 P.2d 33 (1968), it 1234*1234 generally is the case that "[t]here is a probability factor in even the most carefully structured scientific inquiry; seldom is it possible to exclude all possible chance for error in human endeavor.But there is no requirement in our law that the admissibility of scientific-test evidence must be predicated on a 100 percent degree of accuracy."People v. Slone, 76 Cal. App.3d 611, 625, 143 Cal. Rptr. 61, 70 (1978). Indeed, nonscientific evidence often falls far short of such accuracy, especially in the area of paternity identification.
The courts in admitting new scientific evidence have frequently relied on the practical application of a principle in a given discipline or area of endeavor as a sufficient indication of reliability. The widespread use of x-rays and radar prior to their judicial acceptance was an important factor in achieving test acceptance. Strong, Questions Affecting the Admissibility of Scientific Evidence, 1970 U. of Ill.L.F. 1, 12. However, the rule requiring general acceptance should not be too restrictively applied. "[N]either newness nor lack of absolute certainty in a test suffices to render it inadmissible in court.Every useful new development must have its first day in court."United States v. Stifel, supra, at 438.
Moreover, admissibility is not governed solely by the general acceptance test, although a showing of general acceptance would generally be sufficient.The Frye test has been criticized as being overly rigorous, and some jurisdictions have held that a conflict in expert opinion affects weight rather than admissibility, and consequently have modified the rule.
Various legal scholars have proposed other foundational standards by which to determine admissibility of new scientific evidence. The paramount concern is, of course, whether the evidence is sufficiently reliable. Different types of scientific evidence may pose varying and sometimes difficult problems for the integrity of the fact-finding process, but in an age when one scientific advancement tumbles in rapid succession upon another and may be known only among a limited circle of scientists, we are not inclined to adopt a standard that would deprive the judicial process of relevant scientific evidence simply because it is of recent vintage or because knowledge of the principles, or the process for applying a principal, is limited to a small but highly specialized group of experts.Tests that have passed from the experimental stage may be admissible if their reliability is reasonably demonstrable. 3 Jones on Evidence § 15.9 (6th ed. 1972).
An analysis of the admissibility of scientific evidence, while taking into account general scientific acceptance and widespread practical application, must focus in all events on proof of inherent reliability.A scientific test designed specifically for 1235*1235 the purpose of a lawsuit may pass muster with sufficient proof of reliability and an adequate explanation of the pertinent variables and potential inaccuracies so that a trier of fact may make a rational appraisal. We do not intend, however, that a courtroom should be a forum for scientific experimentation. Adjudication means fact-finding, and while speculation is not legitimate in that process, a trier of fact should not be deprived of scientific data because some controversy attaches to it.Management of doubt is a major aspect of our rules of procedure and evidence, and that which reasonably leads to resolution of doubt and ascertainment of truth should be adducible.
In this light it is appropriate in determining reliability to give some consideration to the nature and the reliability of the evidence that must be relied upon in the absence of the scientific evidence. In any event, when the underlying scientific principle, and the means for applying that principle to resolution of legal issues, have received widespread acceptance, there will usually be no reason to reject the test.
In the instant case the following elements must be addressed to provide a sufficient foundation for the admissibility of HLA tests: (1) the correctness of the genetic principles underlying the test for determining paternity; (2) the accuracy and reliability of the methods utilized in application of the principle to determine paternity; (3) the effect of variables such as occur in persons of different nationalities or ethnic origins that would influence the accuracy of the test; (4) other factors that might tend to invalidate the test or significantly change the probability of accuracy; (5) establishing that the actual method employed and the particular test used in a given case were performed in accordance with proper procedures and with proper materials and equipment; and (6) the qualifications of the necessary witnesses.
We recognize that it has been asserted in some literature that the test is highly accurate when performed under the right conditions and is widely accepted, even though it is of recent vintage, at least in this country. A number of articles in medical and 1236*1236 legal periodicals assert that the HLA test is an improved and reliable method for determining paternity. Joint AMA-ABA Guidelines: Present Status of Serologic Testing in Problems of Disputed Parentage, 10 Fam.L.Q. 247 (1976);Polesky & Krause, Blood Typing in Disputed Paternity Cases — Capabilities of American Laboratories, 10 Fam.L.Q. 287 (1976);Terasaki, Resolution by HLA Testing of 1000 Paternity Cases Not Excluded By ABO Testing, 16 J.Fam.L. 543 (1978).
Although these articles are helpful in ascertaining the extent to which the HLA test, or that test in conjunction with others, is demonstrably reliable and has achieved acceptance in the scientific community for paternity identification, the articles are not sufficient, absent expert testimony, for this Court to determine as a matter of law the issue of general admissibility, especially in view of the paucity of legal opinions on this point.The articles require expert interpretation and elaboration. It is not clear, for example, that they all define the HLA test in the same manner, or require that the same procedures be followed to achieve the degree of reliability claimed. Nor is it clear what other tests, if any, should be used in conjunction with the HLA test to achieve the highest degree of accuracy. In short, there are numerous unanswered questions which should be addressed by expert testimony to lay the necessary foundation, if indeed it can be laid.
In this case the plaintiff failed to establish an adequate foundation at trial for the admissibility of the HLA tests. This conclusion is required for several reasons. First, the laboratory technician who did the basic workup on the blood samples for the test was clearly not qualified to testify with respect to the basic validity of the test. Her testimony indicated that most of her work with HLA tissue typing was used in connection with organ transplantation. It is not possible to discern from the record whether the reliability claimed for HLA tests in determining tissue compatibility in organ transplants is transferable to paternity identification. She did, however, testify to the necessary chain of custody of the blood samples and the actual use of the blood samples in performing the test.
Dr. DeWitt, a pathologist, was relied on to establish the necessary scientific foundation. Counsel stipulated that he is an expert, a practice wholly appropriate in many cases, but one that leaves this record devoid of evidence of his qualifications — evidence that is essential in this particular case. In a case dealing with the proposed admissibility of a new scientific test which presumably will be relied on innumerable times in the future, the stipulation leaves a hiatus in the necessary foundation.
Furthermore, his testimony does not supply the necessary information as to the general acceptance of the test, the existence of verification studies, if any, and the particular tests that were in fact performed in this case. There is no evidence in the record which establishes his expertise either in the theory or in the use of HLA testing for paternity purposes. In addition, there is no evidence indicating whether special training in pathology or some other field is a necessary prerequisite to qualify a witness to testify concerning the test.
Dr. DeWitt did state that the test is highly accurate and has been in use for some fifteen years, and that "the figures that we used to deduce the possibilities are based on the analysis of a large number of families." He further testified that the test was widely used "for medical purposes." The difficulty with this testimony is that it is too general, too vague, and too unrelated to the specific requirements for establishing a foundation for the test as a means for determining reliability. Since his testimony did not focus specifically on paternity identification, it may and, as best as can be determined from the record, in fact does refer to other medical uses such as tissue compatibility for purposes of organ transplantation. Furthermore, Dr. DeWitt 1237*1237 did not indicate how the table of percentages used to establish paternity probabilities was arrived at, although he did testify generally that the probabilities "were widely accepted" and "supported by similar work elsewhere done in public by other people." But he did not explain what he meant by "widely accepted," or by whom, and he did not supply any detail as to the work done by others. Nor does it appear that he had particular knowledge obtained from a technical background and training in the area, or from familiarity with the scientific literature on the subject. The general statement that the method is used widely and has wide scientific acceptance is not sufficient, especially in view of the fact that the test applications apparently were unrelated to paternity identification.
Furthermore, in order to make a proper determination of the advisability of admitting HLA test results in any given case, the foundational information before the court should include the number and type of other blood and tissue tests which have been administered to the persons involved in the litigation and the cumulative effect of the additional tests on the predictive accuracy of the HLA test. As stated in J.B. v. A.F., supra, 285 N.W.2d at 883:
The mean probability of excluding a male who in fact is not the father of a child through HLA testing, alone, is between 78% and 80% for blacks, whites and Japanese. If six systems (ABO, Rh, MNSs, Kell, Duffy and Kidd) plus HLA are used, the cumulative probability of excluding a male who in fact is not the father of a child rises to 91.21% for blacks, 93.34% for whites and 91.42% for Japanese. [Footnotes omitted.]
In the instant case there is no evidence at all that the ABO, Rh, MNSs, Kell, Duffy or Kidd tests were employed, yet the percentage Dr. Dewitt testified to seems to assume that those tests were administered. It may be that there was no necessity for administering these tests, but if so, the record must so demonstrate.
Also, evidence should be adduced showing the effect, if any, of the particular racial or ethnic origin of the subject on the calculated probability of exclusion or inclusion of paternity. In addition, qualified witnesses should address the significance of the particular genetic markers relied upon, whether they were inherited from only one parent or both, and the frequency with which they may appear in the population at large. As stated in Lee, Current Status of Paternity Testing, 9 Fam.L.Q. 615, at 628:
Each genetic marker or system of genetic markers provides different chances of exclusion ... . The white blood cell isoantigen system alone provides a 76% chance of exclusion. The next 13 systems provide from 32% to 13.8% chance of exclusion. By using the first 4 systems, a cumulative chance of over 90% is reached; by the first 7 systems, a 95% chance; and by all systems, a chance of 99.27%. In practice, only a limited number of laboratories presently have the capability of testing nearly all these genetic markers. The amount of involvement may not be justified by the small increase in chance of exclusion... . In the United States, tests with a chance of 70% exclusion can be carried out by a number of laboratories. If demand and interest increase, the capability of conducting tests with a 90% or higher chance of exclusion could be reached in a short time.
Finally, and in addition, the proponent must establish that the sera used in the test and the sophistication of the laboratory are of the quality necessary to obtain the degree of reliability claimed:
Were blood specimens drawn from the right parties? Were the tests done properly with reliable reagents, suitable instruments, appropriate techniques and by experienced technologists? Were results of the tests correctly interpreted? Has the validity of an indirect exclusion been seriously and carefully examined? Have all the known genetic variations, ethnic differences, as well as physiologic and 1238*1238 pathologic conditions been taken into consideration? If any of these aspects are neglected, a true father may be relieved from supporting his child, a true parent may be denied his child, or an immigrant child may be barred from reunion with its true parents. These considerations will become even more pertinent as soon as a variety of genetic markers not yet customarily used in many laboratories are included. [Id. at 625-26.]
See also Footnotes 9 and 10.
For the foregoing reasons, we hold that admission of the HLA test was without proper foundation and was clearly prejudicial error. In view of this conclusion, it is unnecessary to address other assignments of error by the defendant.
Reversed and remanded for a new trial. No costs awarded.
MAUGHAN and WILKINS, JJ., concur.
CROCKETT, Chief Justice (dissenting).
It is my belief that the majority opinion itself demonstrates that the parties have had their entitlement to a fair trial in which the rulings on evidence complained of were well within the latitude of discretion of the trial court and that it is therefore the duty of this Court to affirm the judgment.
There are several propositions which should be considered and which support that conclusion. The first is that this was a trial to the court, and not to a jury. For that reason, the rulings on evidence need not be as restrictive, because the court should be more knowledgeable than a jury in analyzing and determining the weight and effect to be given the evidence.
From the admirably informative and lucid exposition in the main opinion, it appears that the HLA test provides proof to a very high degree of probability on the question of paternity. As the opinion states, Dr. DeWitt was relied on to establish the necessary foundation for its admission. His qualifications were sufficient to satisfy counsel for both sides and the trial court. I see no reason for this Court to doubt either their knowledge or integrity, or the propriety of entering into such a stipulation; and it seems to me quite anomalous for this Court to do so. It being so agreed by the parties, Dr. DeWitt's qualifications should be taken as unquestioned.
As the main opinion states, he testified that the test is highly accurate and has been in use for some fifteen years, and that the figures that are used to deduce the possibilities are based on the analysis of a large number of families. He further testified that the test was widely used for medical purposes.
The opinion also correctly points out that Sec. 78-45a-10 provides that the admissibility of blood tests showing the possibility of paternity is within the discretion of the court; and by sound reasoning points out that the HLA test should be considered as included within that statute.
I heartily approve and subscribe to the statement from United States v. Stifel that every new and useful acquisition of knowledge must sometime have its use for the first time; and that neither newness nor lack of absolute certainty in such a test should prevent its results from being received and considered as evidence.
To whatever degree the evidence in question may be lacking in certainty, that should be considered as going to the weight to be given it, rather than to its admissibility. This would have the advantage of allowing the court to receive evidence which appears to have substantial probative value, to be considered along with all of the other evidence in the case, rather than to forego entirely the use of such evidence.
1239*1239 It is upon the basis of what is said in the main opinion and what has been said herein that it is my judgment that there was no prejudicial error, because the receipt of such evidence was well within the latitude of discretion which should be allowed the trial court, and that, consequently the judgment should be affirmed.
HALL, J., concurs in the dissenting opinion of CROCKETT, C.J.
 The trial court's findings of fact indicate the parties voluntarily submitted to the blood tests. There is no court order or stipulation of the parties regarding blood testing in the record. Appellant claims that counsel agreed only to the performance of the HLA test, not to the admissibility of its results. In passing we note that the trial court has authority pursuant to § 78-45a-7 to order blood tests of "the mother, child and alleged father ..." in a paternity action and that a refusal to submit to such a test may be used as a basis for resolving the question of paternity against a party. The court may also order blood tests pursuant to § 78-25-18. It should be noted, however, that the Uniform Act on Paternity, which constitutes Chapter 45a of Title 78, does not specify what types of blood tests may be ordered or are admissible in evidence.
 Lee, Current Status of Paternity Testing, 9 Fam.L.Q. 615, 621 (1975). See also Joint AMA-ABA Guidelines: Present Status of Serologic Testing in Problems of Disputed Parentage, 10 Fam.L.Q. 247, 272-78 (1976), for an explanation of HLA blood typing.
 Comment, The Psychologist as Expert Witness: Science in the Courtroom, 38 Md.L.Rev. 559, 557-78 (1979).
 In United States v. Franks the court stated:
Though United States v. Stifel, 433 F.2d 431, 438, 441 (6th Cir.1970), cert. denied, 401 U.S. 994, 91 S.Ct. 1232, 28 L.Ed.2d 531 (1971), applied the Frye v. United States, 54 App. D.C. 46, 293 F. 1013 (1923), standard governing admissibility of scientific evidence as [to] whether the scientific process has gained "general acceptance in the particular field in which it belongs," we deem general acceptance as being nearly synonymous with reliability.If a scientific process is reliable, or sufficiently accurate, courts may also deem it "generally accepted."Accord, United States v. Brown, 13 Crim.L.Rep. 2203, 2204 (D.C.Super. Ct. May 1, 1973). [511 F.2d at 33, n. 12.]
Unless an exaggerated popular opinion of the accuracy of a particular technique makes it use prejudicial or likely to mislead the jury, it is better to admit relevant scientific evidence in the same manner as other expert testimony and allow its weight to be attacked by cross-examination and refutation.[519 F.2d at 466.]
 Boyce, Judicial Recognition of Scientific Evidence in Criminal Cases, 8 Utah L.Rev. 313 (1962-64);Latin, Tannehill and White, Remote Sensing Evidence and Environmental Law, 64 Cal.L.Rev. 1300 (1976); Strong, Questions Affecting the Admissibility of Scientific Evidence, 1970 U. of Ill.L.F. 1; McCormick, Law of Evidence, § 203 (1972).
 An illustration of the need for flexibility is Coppolino v. State, Fla.App., 223 So.2d 68, cert. denied, 399 U.S. 927, 90 S.Ct. 2242, 26 L.Ed.2d 794 (1970), in which a test which was developed specifically for that trial and its results were admitted in evidence. Obviously, in this case general acceptance in the scientific community was impossible.
 Published articles and books may also be used as evidence supporting points (1) and (2) above.
 The accuracy of the HLA test is clearly dependent on, among other things, the quality of the sera used in a given test and the sophistication of the laboratory involved. Weiner and Socha, Methods Available for Solving Medico-Legal Problems of Disputed Parentage, 21 J.For.Sci. 42, 61 (1975), in discussing the claims that HLA testing can exclude the chance of parentage to a 99% certainty, stated the following:
It would seem from these considerations that the virtual solution of problems of disputed parentage is now at hand. Unfortunately, this solution is beset with numerous pitfalls, and few laboratories, if any, are equipped to carry out all the necessary tests. The performance of all the tests mentioned in this report would be a laborious task indeed, and too costly in time and material for routine use. Furthermore, the HLA tests are reputed to have the reproducibility of only about 90%, so that the possibility of errors is a real one indeed.
 In fact, there may be some question as to the extent of the use of the HLA test. In a survey of representative American laboratories conducted in 1974, only 16% of the American Association of Blood Banks (AABB) indicated they had the capacity for HLA typing, and only 2% of the non-AABB laboratories had the capacity for HLA typing; and of all the laboratories surveyed, none indicated they routinely used HLA typing in paternity testing. Poleski and Krause, Blood Typing in Disputed Paternity Cases — Capabilities of American Laboratories, 10 Fam.L.Q. 287, 289-92 (1976).
 HLA tissue typing was originally developed to match donor and recipient pairs for organ transplantation. The inheritance pattern for purposes of paternity testing has only been recently studied. Shaw and Kass, Illegitimacy, Child Support, and Paternity Testing, 132 Hous.L.Rev. 41, 56 (1975).
 See Del Porto v. Nicolo, 27 Utah 2d 286, 495 P.2d 811 (1972) and authorities therein cited, including 5A C.J.S. Appeal and Error § 1715.
 See main opinion and Cramer v. Morrison, 88 Cal. App.3d 873, 153 Cal. Rptr. 865 (1979).