*595OPINION OE THE COURT BY
RICHARDSON, C.J., AND LEVINSON, J.The petitioner-appellee, pursuant to HRS § 579-1, petitioned the family court (1) for an adjudication of paternity against the defendant-appellant, (2) for an order of reimbursement of expenses incurred because of the pregnancy of the petitioner and the birth of her alleged illegitimate son, and (3) for an order to pay for the support, maintenance and education of the child until he reaches the age of 20 years. The defendant appeals from an adverse judgment following a jury trial.
The petitioner was married at the time the child was conceived but she was living separately and apart from her husband. The child was born January 13,1967 between the dates of the petitioner’s interlocutory and final decrees of divorce.
Prior to the close of the petitioner’s case-in-chief and over the defendant’s objection, the court permitted the mother to carry the baby, then aged nine months, through the courtroom and in front of the jury for approximately 30 seconds. The defendant objected to the exhibition for the following reasons: (1) that the presence of the child *596would create sympathy in favor of the petitioner, (2) that the exhibition of the child for the purpose of showing some physical resemblance to the defendant is “contrary to the facts of life,” and (3) that the exhibition “does not and would not add to the probative evidence in the case.” The court ruled that the reasons for permitting the exhibition were twofold: “to let the jurors see that there is a live baby and to generally appraise the physical characteristics of the baby.” (Emphasis added.) The defendant specifies the exhibition of the child as error.
The petitioner’s final decree of divorce was admitted into evidence during the trial of the present case. The petitioner was allowed to read to the jury the specific declaration in the decree finding that the husband “had no responsibility to support or maintain” the child. The defendant objected to the selected reading on the ground that “it adds nothing to this case.” The objection was overruled and the defendant specifies that ruling as error in his appeal.
The defendant also specifies as error an instruction given to the jury that the average duration of pregnancy is about 280 days. The basis for this contention is that there was no medical evidence offered at the trial to support this instruction.
We reverse.
I. EXHIBITION OF THE CHILD.
It is often said that “like is apt to beget like”; yet this literary epigram cannot serve as the basis for a wholesale or limited exhibition of the child to a jury in a paternity case in light of contemporary scientific knowledge. The law must rely upon hard facts rather than flaccid assumptions which have outlived their usefulness. A rule grounded in the popular beliefs of many generations may become so entrenched that its claim to authority appears absolute *597when in fact the reason for the rule has long been undercut by scientific knowledge. The present case requires that we question the assumption that laymen, as jurors who must determine the paternity of a child, can profit from an exhibition of the child. This necessarily compels consideration of relevant scientific facts with respect to physical resemblances, especially in the fields of genetics and physical anthropology. After exhaustive research we can discern no good reason either in law or in science to warrant the exhibition of a child to a jury for the purpose of proving paternity.1
In determining whether the exhibition of a child to a jury for the purpose of proving paternity is proper, we must first narrow the inquiry by asking whether the resemblance of child to parent is ever evidence of paternity. The unstated and untested assumption that resemblance is a relevant issue in a paternity case has resulted in a great deal of analytical confusion and division of authority in *598the case law.2 Consequently, it is necessary to ask: (1) what kind of resemblance evidence, if any, is relevant to the issue of paternity and, (2) if some kind of resemblance evidence is relevant, who is qualified to determine from an exhibition whether the child resembles the alleged father?
A. General Resemblance as Proof of Paternity.
The introduction of resemblance evidence as proof of paternity is justified only if general or specific resemblance of a child to his parent has a basis in scientific fact. We address ourselves first to the relevancy of general resemblance to the ultimate issue of paternity.
The assumption that inferences concerning paternity may be drawn by the fact finder, as a matter of common knowledge, from the general resemblance of a child to the alleged parent has long been accepted and generally unquestioned.3 Indeed, in the present case the trial court *599stated that one reason for allowing the exhibition was to permit the jury “to generally appraise the physical characteristics of the baby.” But if general resemblance evidence is to be used to prove paternity, then a child’s face must be inherited as a unit from the parent.4 This is simply contrary to well settled scientific evidence.5
If a child’s face is not inherited as a unit, then any evidence such as an exhibition of the child or testimony by an expert for the purpose of showing general resemblance is irrelevant in determining paternity and can work only to the prejudice of the defendant. Wigmore himself rejected general facial resemblance as proof of paternity, while arguing that specific resemblance may be proven by an exhibition to the jury. 1 Wigmore, Evidence § 166 at 626 (3d ed. 1940).
B. Specific Resemblance as Evidence of Paternity.
Just as a child’s face is not inherited as a unit, neither is an individual feature inherited as a unit from any one parent. See authorities cited note 5 supra. Rather, inheritance has been traced to physical traits even more specific *600than a single feature. Consider, for example, the effect of heredity on the nose.
Some studies might indicate that there is one key gene producing the general shape of the nose, but most authorities agree that quite a number of genes are at work, each on a different part. That is, there may be separate genes for the bridge (its shape, height and length); the nostrils (breadth, shape and size of apertures) ; the root of the nose and its juncture with the upper lip; and the bulb, or point of the nose.
Often, it is true, the nose as an entire unit appears to be “inherited” from one parent. Where such resemblance occurs it could be assumed that the different genes involved were passed over in combination and were almost all of them dominant over those of the other parent.
Equally often, on the other hand, a child has a nose which seems to be a cross between that of both parents. This would bear out the theory that several or many unit factors are involved. At any rate, it is clear that distinctive genes are at work, and that they sort out independently; otherwise, the nose of every child would be a blend of its parents’ noses. Even in the most inbred peoples, however, noses of every shape and size appear, proving the Mendelian segregation and sorting out of the “nose” genes. A. Sheinfeld, Your Heredity and Environment 96-8 (1965).
In short, the individual does not “inherit” characteristics but only the potentialities to develop in a certain way through the transmitted genes.6 Moreover, since genes themselves interact, all genes probably affect all characteristics to a certain degree.
*601Since only potentialities are “inherited,” other factors such as environment and age may also shape the visible manifestations of the “inherited” characteristics. See E. Hooton, Up from, the Ape 443 (2d ed. 1946). Indeed, the number of factors involved in shaping any normal trait is so large that resemblance evidence has been regarded as “highly unreliable.”7 Thus even where the inquiry is focused on the precise characteristics susceptible of establishing any scientifically ascertainable resemblance, other factors must be taken into account in the final analysis.
Because the link between parent and child can be discerned only in these very specific instances and not by evidence of general resemblance or by a comparison of individual features, we now reach the question presented in this case: whether an exhibition of a child to a jury is probative and admissible demonstrative evidence. In light of the specialized knowledge required to judge the relevancy of corporeal conditions to the issue of paternity and the vast number of other non-hereditary factors involved in shaping a trait, we think any evidence concerning resemblance or non-resemblance must be given by the testimony of a qualified expert and not by an exhibition of the child to the jury.8 The identification of a physical *602characteristic, whether that characteristic is in fact hereditary, what other factors may have helped shape it, and how the characteristic in question is linked to a similar characteristic possessed by the alleged parent are all questions for experts, not laymen. Even color9 and form10 may contain far more variations, than we as laymen may discern, when viewed by the experienced eye of the expert.
If only an expert can inform the jury of any resemblance between parent and child, it follows that the exhibition of a child to a jury would be useless in determining paternity. The only way an exhibition might possibly be *603justified would be as au aid to the expert in pointing out the reasons for his findings or opinions. Certainly other visual aids are available to perform this function which would further the inquiry without prejudicing the jury against the defendant.11 Were we to allow such an exhibition we would be subjecting the parties to a fanciful legal process which could only serve to prejudice the accused. The rule of evidence which governs resemblance in paternity cases must rest on demonstrable findings of science; the law must change accordingly.
In sum, we agree that the specific resemblance between a child and the person alleged to be the father is a relevant issue in a paternity case but we cannot find any rule of reason, any policy of the law of evidence, or any fact of science which provides a basis for allowing the exhibition of a child to show resemblance. As we have stated, a jury gains nothing from an exhibition even when their attention is focused upon the relevant inherited traits since independent expert interpretation is required. An exhibition can only serve to expose the defendant to proven dangers. Therefore, we hold that the exhibition of a child to the finder of fact in a paternity case is not to be permitted. However, expert testimony concerning the resemblance of a child to the person alleged to be the father is admissible to prove or disprove the paternity of the child.
In the case of In re Ah Sam, 24 Haw. 591 (1918) this court held that it was not error for the trial court to per*604mit the alleged illegitimate child to remain in the courtroom in the presence of the jury and be carried by the mother to the witness stand while testifying. That portion of Ah 8am which approves of such exposure of the child to the jury is hereby overruled.
II. THE ADMISSIBILITY OF THE DIVORCE DECREE.
The defendant argues that the final decree of divorce stating that the husband was not responsible for the support or maintenance of the child was not admissible for the purpose of proving the defendant’s paternity. Although the defendant waived any error committed by the trial judge on this point, a comment upon this apparent error may prevent its repetition upon retrial.
The thrust of the defendant’s case was that he was not the father of the child and that the former husband was the father. The record shows no basis for the determination in the final decree of divorce of the husband’s non-responsibility for the support or maintenance of the child. It is possible that the jury drew an inference that the issue of non-paternity on the part of the husband had already been judicially decided. A determination of the husband’s non-paternity in the divorce proceedings was certainly not binding upon the defendant, who was not a party thereto-, nor was it admissible as evidence by the petitioner in the present case. In re Estate of Cunha, 49 Haw. 273, 290, 300, 414 P.2d 925, 934, 939 (1966). The divorce decree, if it was relevant and material at all, was admissible only for the purpose of establishing the mere fact of its own rendition and those legal consequences which result from that fact. It should not have been used as affirmative evidence against a stranger to the suit in which it was rendered to prove the existence of any fact underlying the divorce decree and support order, such as the issue of *605paternity. Id. at 300, 414 P.2d at 940. The court should take great pains to assure that the jury is not made aware of any prior judgment which might reasonably be construed to indicate an underlying determination of the paternity question in another action not binding on the party against whom it was offered.
III. JURY INSTRUCTION RELATING TO DURATION OF PREGNANCY.
Over defendant’s objection, the following instruction to the jury was given by the trial court although no evidence had been introduced relating to the duration of pregnancy:
The average duration of pregnancy is about 280 days, that is ten lunar months, or approximately nine calendar months. This includes the period from the first day of the last menstrual flow until the birth of the child. As conception, however, presumably occurs around the middle of the menstrual month, the actual duration of pregnancy is 270 days — that is between the act of intercourse and the birth. There are, of course, many individual variations, and normal pregnancies have been known to last considerably less or more than the average time stated. Because of these uncertainties, it is not always possible to predict accurately the date of an expected childbirth.
We hold that this instruction properly covered matters that were appropriate for judicial notice. See Territory v. Duvauchelle, 40 Haw. 534, 536, 540 (1954), Barretto v. Akau, 51 Haw. 383, 390, 463 P.2d 917, 922 (1969). A fact is a proper subject for judicial notice if it is common knowledge or easily verifiable. See 9 Wigmore, Evidence §§ 2571, 2583 (3d ed. 1940). The facts stated in the instruction are common knowledge.
Other specifications of error are without merit.
*606Helen B. Ryan (Bruce M. Clark with her on the briefs) for defendant-appellant. Wayne Luke, Deputy Corporation Counsel (Paul Devens, Corporation Counsel, City and County of Honolulu, of counsel) for petitioner-appellee.Reversed and remanded for a new trial.
In our research we have reviewed the leading authorities in the fields of genetics and physical anthropology. Some of them are cited in our opinion today. We view the use of such authorities to be relevant in making decisions of general applicability, not necessarily limited to the immediate parties. The process of adjudication would be severely limited and appellate opinions deprived of the most persuasive support if judges were not permitted to resort to such authorities.
Courts have used similar materials in determining that children of racially mixed marriages are not inferior to their parents, Perez v. Lippold, 32 Cal. 2d 711, 720, 198 P.2d 17, 22 (1948) (Traynor, J.); in deciding whether regulating the weight of bread is reasonably necessary, Jay Burns Baking Co. v. Bryan, 264 U.S. 504, 517 (1924) (a weUdocumented dissent by Brandéis, J. with Holmes, J. joining); and in ruling on the detrimental effect of segregation of white and black children in public schools, Brown v. Bd. of Education, 347 U.S. 483, 494 (1954). Further, the United States Supreme Court has said in an opinion by Mr. Justice Holmes that “the Court may ascertain as it sees fit any fact that is merely a ground for laying down a rule of law. . . .” Chastleton Corp. v. Sinclair, 264 U.S. 543, 548 (1924). See also Davis, Judicial Notice, 55 Colum. L. Rev. 945, 952-66 (1955); Currie, Appellate Courts Use of Facts Outside of the Record by Resort to Judicial Notice and Independent Investigation, 1960 Wis. L. Rev. 39, 50 (written by a justice of the Wisconsin Supreme Court); Proposed Rules of Evidence for the United States District Courts and Magistrates, Advisory Committee’s Note to Rule %-01 (1969).
Those jurisdictions ruling on the question have taken positions which are sometimes diametrically opposed. For example, the New York rule excludes exhibition of the child and all testimony relating to resemblance as such evidence is “neither accurate nor reliable.” Bilkovic v. Loeb, 156 App. Div. 719, 723, 141 N.Y.S. 279, 281 (1913). At the other extreme are jurisdictions which permit the exhibition of the child regardless of the child’s age, reasoning that the age and immaturity of the child affect only the weight to be attributed such evidence by the jury. See, e.g., People v. Haab, 260 Mich. 673, 676, 245 N.W. 545, 546 (1932) (36-day-old child). Some courts take the ostensibly “middle” position that as a matter of law there should be no exhibition of a child under a certain minimum age. See, e.g., State v. Harvey, 112 Iowa 416, 84 N.W. 535 (1900) (no exhibition of child under two years unless race is relevant).
That assumption was introduced to the common law by the House of Lords by way of dictum in the Douglas Peerage Case, 2 Hargr. Collect. Jurid. 402 (1769) and had its counterpart in the literature and science of that day. From an appeal in Douglas, Lord Mansfield said: “ ‘It is proved that the elder child, the appellant, was the exact picture of his father; and the younger child, Sholto, as like Lady Jane as ever a child was like a mother. I have always considered likeness as an argument of a child’s being the son of a parent; and the rather as this distinction between individuals is more discernible in the human species than in other animals. A man may survey ten thousand people before he sees two faces perfectly alike. . . .”’ Quoted in Kenny, Physical Resemblance as Evidence of Consanguinity, 39 Law. Q. Rev. 297-98 (1923).
The theory of “preformation” was that “each egg contained a very small man or woman, complete in all characteristic details, which would grow by expansion into an adult.” H. Papazian, Modern Genetics 11 (1967). That theory was widely accepted in the literature and science of earlier times. See the reference to preformation in Shakespeare’s plays, 1 Wigmore Evidence, § 166 at 623 (3d ed. 1940). See also H. Papazian, supra at 10-15; A. Montagu, Human Heredity 77 (2d ed. 1963); T. Dobzhansky, Genetics and Origin of Species 177 (3d ed. 1951).
E.g., A. Scheinfeld, Tour Heredity and Environment 94-96 (1965); T. Dobzhansky, supra, note 4; E. Hooton, Op From the Ape, 423-661, especially 507 (3d ed. 1946). 9 World Book Encyclopedia, Heredity 192 (1969); 7 Chamber’s Encyclopedia, Heredity 58, 63 (1968). Since the rule governing resemblance evidence was in its inception a judicial creation and the courts have rendered every conceivable variation of the rule in the cases, the experience of the law in this area has provided guidance in fashioning the rule. The emphasis in the judge-made law has since turned to inherited characteristics of more conclusive certainty such as blood types. See e.g., Note, Domestic Relations — Mass. Joins States Holding Blood Grouping Tests Conclusive in Paternity Suits, 9 De Paul L. Rev. 281 (1960).
T. Dobzhansky, Evolution Genetics and Man 72-90 (1961); T. Dobzhansky, Heredity and the Nature of Man 49-51 (1964); Carson, Heredity and Human Life 75 (1963); A. Montagu, Human Heredity 83 (2d ed. 1963); L. Snyder, Principles of Heredity 3 (1935).
Wiener, Determining Parentage, 40 The Scientific Monthly 323 (1935); see also Wiener, Blood Tests for Paternity, 95 J.A.M.A. 681; Wiener, Parentage and Blood Groups, 191 Scientific American 78 (July 1954); Holz, The Trial of a Paternity Case, 50 Marquette L. Rev. 450, 489, 491 (1967); Note, Scientific Gadgets in the Law of Evidence, 53 Harv. L. Rev. 285, n. 8 (1939). But see Keiter, Advances in Anthropological Paternity Testing, 21 Am. J. of Physical Anthropology [n.s.] 81 (1963). In England, where the rule originated, the trend is to place little weight on resemblance evidence. S. Schatkin, Disputed Paternity Proceedings 159 (3d ed. 1953).
New York has excluded resemblance evidence altogether. Bilkovic v. Loeb, 156 App. Div. 719, 723, 141 N.Y.S. 279, 281 (1913). Until more conclusive evidence is discovered, we believe the difficulties in measuring resemblances for the purpose of proving paternity require not only expert guidance, but a focused examination of each relevant, specific characteristic.
Contrary to widely accepted thinking, the identification of “racial characteristics” and their relevancy to paternity are not matters of common knowledge but subjects of expert testimony. See Holz, supra note 7, at 491 n.271. Generally, races, as selected populations differing in the frequency of occurrence of particular inherited characteristics, possess measurable physical traits which overlap considerably between groups of people.
Broadly speaking, individuals belonging to different major groups of mankind are distinguishable by virtue of their physical characters, but individual members, or small groups, belonging to different races within the same major group are usually not so distinguishable. Even the major groups grade into each other, and the physical traits by which they and the races within them are characterized overlap considerably. With respect to most, if not all, measurable characters, the differences among individuals belonging to the same race are greater than the differences that occur between the observed averages for two or more races within the same major group. Statement on the Nature of Race and Race Differences by Physical Anthropologists and Geneticists, Sept. 1952 (UNESCO) quoted in A. Montagu, Man’s Most Dangerous Myth: the Fallacy of Race 368 (4th ed. 1964).
The fact that Hawaii represents a unique population, where interracial unions are common, only underscores the need for a focused examination of “racial characteristics” by the fact finder under expert guidance. See A. Montagu, Human Heredity 57 (2d ed. 1963) ; Morton, Chung, Mi, Genetics of Interracial Crosses in Hawaii (Monographs in Human Genetics 1967). Most anthropologists agree that the Mongoloid, European and Negroid features are all “blended” in the Polynesian. See E. Hooton, Up From the Ape 616 (2d ed. 1946).
While some courts permit the exhibition of a child to show resemblance of hereditary abnormalities, we cannot ascertain the benefit of an exhibition to show that a child merely possesses the alleged hereditary abnormality; such a fact is easily admitted. What is material to the issue of paternity is not the existence of the abnormality, but the precise nature and the connection, if any, it may have with the defendant. The nature of the abnormality can only be ascertained by a focused study, and the connection found only by expert testimony.
Thus we need not balance probative weight against jury sympathy where there is no probative weight in an exhibition to a layman. The prevalent emotional factors which hinder a proper perspective of the issues in paternity cases have been recounted many times. Britt. Blood-Grouping Tests and the Law: The Problem, of “Cultural Lag,” 21 Minn. L. Rev. 671, 699 (1937). Note, Admissibility of Blood Tests to Prove Non-Paternity, 29 N.D. L. Rev. 156, 169 (1953); Note, Issues as to Paternity, 17 Mod. L. Rev. 152 (1954); Schatkin, Paternity Proceedings —A Changing Concept, 42 J. Crim. L. C.P.S. 821 (1952); Feirich, Blessing or Blight: The Paternity Act of 1957, 47 Ill. B.J. 824, 827 (1959); Note, Scientific Gadgets in the Law of Evidence, 53 Harv. L. Rev. 285, 286 (1939).