Brickley, J.
On December 7, 1978, defendant was convicted by a jury of first-degree murder, as a result of a homicide committed during the perpetration of a burglary.1 Evidence linking defendant to the crime included inculpatory statements by the defendant made to a prosecution witness, fingerprints of the defendant obtained at the scene, and the results of blood analyses which included the defendant in the class of possible perpetrators. There were no eyewitnesses to the crime. There was, however, testimony at trial which established that the decedent was seen alive at 8 a.m. on May 16, 1978. The victim’s body was discovered in his apartment at approximately 11:30 a.m. that same day. The apartment had been broken into and ransacked.
The Court of Appeals affirmed defendant’s conviction, 106 Mich App 323; 308 NW2d 194 (1981). We granted leave to appeal, 414 Mich 865 (1982).
Defendant raises three issues on appeal. First, whether burglary in the former first-degree murder statute2 referred to common-law burglary and required proof of all of the elements of the com[5]*5mon-law offense including breaking and entering. in the nighttime; second, whether the results of the blood analyses were admissible at trial without a prior showing that the technique of serological electrophoresis enjoys general scientific acceptance among impartial and disinterested experts; and, third, whether the results of blood analyses are admissible to include an accused within the class of possible perpetrators.
We answer the first question in the affirmative, and the second question in the negative. The answer to the third question must await development of a record by the trial court at the hearing which we order to determine if serological electrophoretic analysis has achieved general scientific acceptance for reliability, Frye v United States, 54 US App DC 46; 293 F 1013 (1923); People v Davis, 343 Mich 348; 72 NW2d 269 (1955), by disinterested and impartial experts, People v Barbara, 400 Mich 352; 255 NW2d 171 (1977); People v Tobey, 401 Mich 141; 257 NW2d 537 (1977).
We retain jurisdiction.
I
Defendant begins his argument by declaring that the Legislature has never enacted a burglary statute. It follows, defendant then argues, that the term "burglary” in the first-degree murder statute in effect when this crime was committed must refer to the common-law crime. Citing Cole v People, 37 Mich 544 (1877), defendant avers that the elements of this offense are the breaking and entering of a dwelling house in the nighttime with the intent to commit a felony. Since proofs at trial established that the act occurred during the daytime, defendant concludes that his first-degree murder conviction cannot stand because the people [6]*6failed to establish all requisite elements of the underlying felony. Conceding that the jury was properly instructed on and necessarily found him guilty of second-degree murder, defendant asks us to reduce his conviction to that of second-degree murder and remand for resentencing.
The people respond by arguing that burglary has been codified in Michigan since ,1837. In addition, it is argued, the courts have looked to statutory burglary to determine the sufficiency of allegations in informations charging burglary. The people further argue that statutory burglary contained all the elements of the common-law crime until 1964, when the Legislature deleted the nighttime element from the "burglary-breaking and entering” statute, MCL 750.110; MSA 28.305. This, it is alleged, impliedly changed the common-law definition of burglary for purposes of the first-degree murder statute. The people acknowledge that a 1969 amendment to the first-degree murder statute did not alter the requirement for proof of a burglary. Nevertheless, it is contended, the amendment was enacted with full knowledge of the posited 1964 change in the elements of common-law burglary, thus amounting to a form of legislative ratification of a new definition of burglary. As a result, the people claim that the failure to prove the nighttime element in the case at bar should not be fatal to this defendant’s conviction of first-degree murder.
Both the people and the defendant rely on People v McDonald, 409 Mich 110; 293 NW2d 588 (1980). In McDonald, the defendant was convicted of first-degree murder during the perpetration or attempted perpetration of a rape. On appeal, the defendant questioned whether a felony-murder conviction could be obtained on the basis of the [7]*7underlying felony of rape once the carnal knowledge statute had been repealed and replaced by the criminal sexual conduct law, criminal sexual conduct not having been added to the felony-murder statute at that time. This Court held that rape survived for purposes of prosecution under the first-degree murder statute:
"We find the Legislature intended that the repealed carnal knowledge statute define rape for purposes of the first-degree murder statute. The conduct proscribed by the former carnal knowledge statute upon which a first-degree murder conviction was based is also presently prohibited under the criminal sexual conduct act. Consequently, the Legislature did not intend to abrogate such conduct as an aggravating circumstance required for first-degree murder. Rape, as formerly defined under the carnal knowledge law, survives for purposes of prosecution under the first-degree murder statute.” 409 Mich 116.
It was concluded that the rape instructions objected to were properly given by the trial court.
Our analysis of the sexual conduct proscribed, characterized by the term rape, as it existed at the time the first-degree murder statute was last recodified in 1931 PA 328 revealed that the prohibited sexual conduct had been codified from the common-law crime and had remained unchanged through various statutes including the present criminal sexual conduct law. The defendant’s conduct always had been punishable. We looked to the 1931 recodification of the first-degree murder statute because no express legislative action to change the sexual conduct required as an aggravating circumstance to support a first-degree murder conviction had occurred in that statute in the interim. It is important to note that McDonald does not suggest that the Legislature amended the [8]*8meaning of the term rape by adopting the criminal sexual conduct law.
Restating the arguments in terms of our analysis in McDonald, defendant would have us look to the last recodification of the first-degree murder statute in 1931 to determine the Legislature’s understanding of the aggravating conduct required to support a first-degree murder conviction. Defendant’s argument focuses on the failure of the Legislature to expressly amend that statute as to burglary in the interim between enactment in 1931 PA 328 and the date of this crime. The people, in contrast, would have us view a 1969 amendment to the first-degree murder statute, which retained a requirement for burglary, in the light of a 1964 amendment to MCL 750.110; MSA 28.305 which deleted the "in the nighttime” requirement for a statutory violation. In short, the people would have us expand the McDonald analysis in this case to include an implied amendment of the conduct required to support a first-degree murder conviction.
In order to better understand our analysis, we first present a brief history of the statutes involved. That will be followed by a more detailed examination of the statutes which prohibited unauthorized entries.
The Revised Statutes of 1838 deemed murder committed during a burglary to be murder in the first degree. RS 1838, pt 4, tit 1, ch 3, § 1. Until 1980, after the commission of the crime in this case, the term burglary appeared in the first-degree murder statute as aggravating conduct which would support a first-degree murder conviction. In 1980 PA 28, the Legislature deleted the reference [9]*9to burglary and substituted the phrase "breaking and entering of a dwelling”.3
The Revised Statutes of 1838 also prohibited the breaking and entering of a dwelling house in the nighttime with the intent to commit a felony. RS 1838, pt 4, tit 1, ch 4, § 10.4 As will be discussed, this statute was treated by this Court as the statutory codification of the common-law crime of burglary. Over time, the Legislature greatly expanded the sweep of this statute by increasing the classes of enumerated structures. Nevertheless, until 1964, the conduct prohibited always included the breaking and entering of a dwelling house in the nighttime with the intent to commit a felony. In 1964 PA 133, the nighttime element was removed.
In cases which construed these statutes, this Court recognized that the Legislature had created statutory burglary. In Pitcher v People, 16 Mich [10]*10142 (1867), Pitcher was charged with statutory-burglary. He argued for a limited meaning for the words "dwelling-house”. Justice Cooley, writing for the Court, noted:
"The statutory definition of burglary in a dwelling-house, is the same as that of the common law; and we must infer that the statute designs simply to provide for the punishment of the common law offense, unless we discover some reason for believing that the legislature employed the definition in some new and restricted sense. No other reason is suggested, except that the words 'dwelling-house’ are used with restricted meaning in prescribing penalties for other offenses — a reason the force of which must depend upon other considerations.
"If the statute had provided distinct punishments for burglary in the dwelling-house proper, and in the outhouses, etc., used in connection therewith a legislative intent to employ the words 'dwelling-house’ in a sense not embracing such out-houses, would be very apparent; but it will be perceived, by an examination of the statute, that no such division of the common law offense has been made, and that unless the word 'dwelling-house’ is used in the sections referred to in their common law sense, many burglaries will not be covered by the statute, and their punishment will consequently not be provided for at all.” 16 Mich 146-147.
Justice Cooley found that the livestock barn into which the defendant broke and entered in the nighttime with the intent to commit a larceny fell within the meaning of the common-law term "curtilage”. He concluded that the defendant was properly charged with burglary under the statute.
In Cole v People, supra, Cole was charged with statutory breaking and entering of a store. Justice Cooley, again writing for the Court, compared the charged offense with burglary. This case has been cited by both the people and the defendant for its definition of burglary. It is noteworthy that the [11]*11definition employed by Justice Cooley is consistent with both the common-law definition and the statutory definition.
This Court has also looked to the statutory crime to determine the sufficiency of allegations in informations charging burglary. In Harris v People, 44 Mich 305; 6 NW 677 (1880), Harris and Williams were charged with and convicted of the statutory crime of attempted burglary. In affirming their convictions, Justice Graves, writing for the Court, noted that burglary is a common-law offense, not a crime created by the Legislature. He further noted that the statutory crime distinguished between simple burglary and aggravated burglary. He cautioned against the conclusion that the addition of aggravating factors to the common-law crime created a new crime:
"The statute does not carve [burglary] into two. It exposes it to different grades of punishment, according as it may or may not be accompanied by the incidents specified in the statute. It may be laid according to the common law, and without referring to the facts on which the imposition of the higher penalty depends; but in such case the punishment cannot exceed the lesser penalty. The accusation will support nothing more. Where the facts are supposed to warrant it, and the higher penalty is contemplated, the crime must be described with the attending facts which justify that penalty.”
Turning to statutory break-ins, Justice Graves continued:
"The various breakings resembling burglary which have been declared criminal by the legislature, are distinguishable from the ancient offense of the common law. They owe their definition to the statute, and the statute must be consulted to ascertain their ingredients. [12]*12When they are charged they must be set forth in substance, as in the statute, with all descriptive incidents, whether negative or otherwise. Koster v People, 8 Mich 431 (1860); Byrnes v People, 37 Mich 515 (1877).” 44 Mich 307.
See also Hall v People, 43 Mich 417; 5 NW 499 (1880); People v Shaver, 107 Mich 562; 65 NW 538 (1895).
In sum, this Court recognized that the Legislature had codified the common-law crime of burglary in these early statutes. We construed this statutory crime by ascribing to it common-law content. The validity of informations charging burglary was tested against the requirements of the statute. Finally, our recognition of statutory burglary was frequently couched in language which compared the offense (and its common-law analogue) with statutory breakings and enterings.
1931 PA 328, § 110, prohibited, inter alia, the breaking and entering of any dwelling house in the nighttime with the intent to commit a felony. Section 111 prohibited such conduct during the daytime. This distinction with respect to the time of the offense persisted until it was abolished in 1964 PA 133.
At all times prior to 1980, the first-degree murder statute required, as an aggravating circumstance, conduct termed burglary. As we have seen, the common-law elements of this crime were adopted as the statutory standard by the Legislature and were consistently construed in their common-law sense by this Court. Thus, at the time the first-degree murder statute was last recodified in 1931 PA 328, the conduct historically known as burglary was prohibited by statute. That conduct included, as an element, that the act be done in the nighttime.
[13]*13The people do not dispute this conclusion. Rather, as previously stated, the people argue that we should look to the 1969 amendment of the first-degree murder statute. The people contend that the Legislature must be presumed to have been aware of the change effected by 1964 PA 133 and that that change "must reflect the intention by the Legislature of repealing the common-law requirement for an 'in the nighttime’ element of burglary”.5 We disagree with this argument in several respects.
It is an elementary principle of construction that we will assign to common-law terms their common-law meaning unless the Legislature directs otherwise. We have done so with burglary. Pitcher v People, supra; Cole v People, supra; Harris v People, supra. In effect, the people ask us to disregard the plain meaning of the word that the Legislature chose to use and find, instead, that the Legislature meant breaking and entering when it said burglary in 1969 PA 331. The rationale for such an unusual request is apparently that burglary and breaking and entering have become so entwined over time by virtue of their placement in the same statute that the Legislature failed to see the difference between the two crimes and regarded them as interchangeable.
1969 PA 331 retained the reference to burglary. Notwithstanding the people’s conjecture about the Legislature’s inability to distinguish between the crimes of breaking and entering and burglary or the Legislature’s treatment of the two crimes as interchangeable, we treat the retention of the term burglary as dispositive on this point. Pitcher v People, supra.
A review of the legislative history of the 1969 [14]*14amendment to the first-degree murder statute convinces us that the Legislature intended that the underlying felony be burglary. House Bill No. 2431 was a bill to amend the Michigan Penal Code. It would have added to the first-degree murder statute, as aggravating circumstances, the crimes of larceny of any kind, breaking and entering, extortion, and kidnapping. It did not remove the reference to the crime of burglary. See 2 Michigan House J (1969), p 1188. This bill passed the House by a vote of 94 to 3 on June 3, 1969. See 2 Michigan House J (1969), pp 1859-1860. On July 1, 1969, a message was received from the Senate, returning the bill with an amendment which deleted the crimes of larceny of any kind and breaking and entering. See 3 Michigan House J (1969), p 2593; 2 Michigan Senate J (1969), pp 1738-1739. On July 2, 1969, the House voted against concurring in the Senate amendment by a vote of 86 to 9. See 3 Michigan House J (1969), pp 2601-2602. Later that day, the Senate again returned HB 2431, together with its amendment, to the House with, the message that the Senate "insisted” on its amendment and requested a conference. See 3 Michigan House J (1969), p 2653; 2 Michigan Senate J (1969), pp 1758-1759. At conference, compromise was achieved. In final form, the amendment to the first-degree murder statute added the crimes of larceny of any kind, extortion, and kidnapping. It is important to note that the reference to the crime of breaking and entering was stricken specifically. See 4 Michigan House J (1969), p 3225; 3 Michigan Senate J (1969), p 2266.
This legislative action must be viewed against the historical backdrop of almost 150 years of legislative action distinguishing between the crimes of breaking and entering and burglary. [15]*15Commencing with the enactment of the territorial laws of 1816, the crimes of burglary and breaking and entering existed side by side until 1964. It is unlikely that the Legislature did not know that the crimes were different and did not act on that knowledge. This point is underscored by the Legislature’s action in 1980 PA 28 where "breaking and entering of a dwelling” was substituted for "burglary”.
The argument that the crimes were viewed by the Legislature as interchangeable is similarly suspect. Approximately 150 years of legislative and judicial history suggest the contrary. Even if the people’s assertion is accepted as true, one might inquire why the Legislature would choose to perpetuate the inconsistency.
We can only conclude that in 1969 the Legislature intended that the crime of burglary, in its common-law sense, be the aggravating circumstance for purposes of the first-degree murder statute.
The people argue that the Legislature has the power to change the common law. Const 1963, art 3, § 7. We do not disagree. The Legislature has changed the common law numerous times and knows how to do so. See 1935 PA 127, now MCL 600.2901; MSA 27A.2901 (prohibition of actions for alienation of affection); 1915 PA 314, now MCL 600.2902; MSA 27A.2902 (abolition of certain real actions known to the common law); 1958 PA 182, now MCL 600.2917; MSA 27A.2917 (limitation on liability of a merchant for conduct involving persons suspected of larceny of goods); 1941 PA 303, now found in MCL 691.561-691.564; MSA 27.1683(1)-27.1683(4) (limitation on the common-law bar to contribution among joint tortfeasors); 1972 PA 294; MCL 500.3101 et seq.; MSA 24.13101 [16]*16et seq. (limitation on tort liability arising out of the ownership, maintenance, or use of a motor vehicle). This is not such an instance.
In People v McDonald, supra, we looked to the first-degree murder statute at the time of its enactment to determine the Legislature’s understanding of the aggravating conduct it deemed necessary to support a first-degree murder conviction. In that case, as here, the Legislature had not expressly amended the first-degree murder statute in the interim between enactment and the commission of the crime in issue. Fundamental to our view is the premise that an express change is necessary for certainty in the conduct proscribed. Such a view precludes our expansion of the holding in McDonald to include implied changes.
It is noteworthy that the people’s argument assumes, without authority, that the 1964 amendment to MCL 750.110; MSA 28.305 which eliminated the "in the nighttime” requirement did, sub silentio, amend the meaning of the common-law term burglary and thus the conduct required to be proved to support a conviction of another crime in a separate unnamed statute, the first-degree murder statute, MCL 750.316; MSA 28.548. Such a result would violate at least the spirit of the title-object clause of Const 1963, art 4, § 24.
We hold that the term burglary in the first-degree murder statute required that the people prove the historic common-law elements of that offense. It was that conduct, the breaking and entering of a dwelling house in the nighttime with an intent to commit a felony, that the Legislature proscribed as an aggravating circumstance in the first-degree murder statute until it expressly acted to amend the first-degree murder statute in 1980 PA 28. Although the defendant, conceding that the [17]*17jury was properly instructed on and necessarily found him guilty of second-degree murder, properly asks us to reduce his conviction to that of second-degree murder, we decline to do so pending resolution of the remaining issues in this case following remand.
II
During the defendant’s trial, identification evidence obtained by blood analyses using the novel technique of serological electrophoresis was admitted in evidence.6 Defense counsel’s timely objection to the scientific acceptability of the technique was overruled. The results of the analyses were offered to demonstrate that the defendant was included in the class of persons who could have committed the charged offense. The defendant argues that these results were inadmissible at trial without a prior showing that the technique of serological electrophoresis enjoys general scientific acceptance among impartial and disinterested experts. We agree.
The admissibility of scientific evidence in this state is governed by the so-called Davis-Frye rule. In Frye v United States, 54 US App DC 46, 47; 293 [18]*18F 1013 (1923), a standard for admissibility of scientific evidence was articulated:
"[W]hile courts will go a long way in admitting expert testimony deduced from a 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.”
In People v Becker, 300 Mich 562; 2 NW2d 503 (1942), Justice Sharpe, writing for a unanimous Court, concluded that the results of a lie-detector test offered by defendant to prove his innocence were properly ruled inadmissible by the trial court. The opinion noted that no testimony had been offered at trial which revealed "general scientific recognition” of such tests at that time. The opinion concluded:
"Until it is established that reasonable certainty follows from such tests, it would be error to admit in evidence the result thereof.” 300 Mich 566.
In People v Davis, 343 Mich 348; 72 NW2d 269 (1955), the admission of allegedly favorable lie-detector results was again sought by a defendant following an unfavorable trial court ruling. Mindful of the .rule in Becker, a trial record was made on the issue, and authorities were cited for the reliability of the results. After restating the rule in Becker, the Court evaluated the material before it in the context of proof of general reliability and acceptability. In a unanimous opinion by Justice Butzel, the Court declined to overrule its earlier holding in Becker. The Court identified its concern about the admission at trial of such results before [19]*19general reliability and acceptance were established:
"The tremendous weight which such tests would necessarily carry in the minds of a jury requires us to be most careful regarding their admission into evidence and we should not do so before its accuracy and general scientific acceptance and standardization are clearly shown.” 343 Mich 372.
In People v Barbara, 400 Mich 352; 255 NW2d 171 (1977), this Court once more considered the issue of the admissibility of lie-detector results. As before, the issue was presented by a defendant who sought the admission of allegedly favorable test results in support of a motion for a new trial following his conviction of extortion. On this issue, a unanimous Court declined the invitation to modify our traditional Davis-Frye rule. Chief Justice Williams, in an exhaustive and scholarly opinion, interpreted the phrase "general scientific recognition”. He noted that the record was devoid of testimony by "disinterested and impartial experts”. 400 Mich 358. Such experts were later characterized in the following way:
“[T]he witnesses were not disinterested scientists. While one would not want an expert witness without experience or background in the technical field, one would want, where the task was to demonstrate the general scientific acceptability, an acknowledgment of the value of the device and the techniques by disinterested scientists whose livelihood was not intimately connected with it. In addition, the Davis-Frye test requires acceptability by those in certain established scientific disciplines.” 400 Mich 376.
In People v Tobey, 401 Mich 141; 257 NW2d 537 (1977), this Court was confronted with the admissi[20]*20bility of voiceprint identification evidence. The trial court had ruled that such evidence was admissible. The Court of Appeals held that an inadequate foundation had beén established and reversed. We found it unnecessary to reach the foundation issue presented until a more basic question was resolved — whether voiceprint analysis had achieved general scientific acceptance as a reliable identification technique. After noting that the expert witnesses offered as proponents of voice-print identification could scarcely be considered disinterested and impartial, and after examining the relevant literature on the subject of voiceprint analysis, Justice Levin, writing for a unanimous Court, concluded that the people, as the offering party, had failed to meet the burden of demonstrating that this novel identification evidence had gained general scientific acceptance as reliable. The standard announced in Becker and Davis, and expanded in Barbara, was specifically reaffirmed.
In People v Salvadore Gonzales, 415 Mich 615; 329 NW2d 743 (1982), we last responded to an attack on the Davis-Frye rule. The trial court ruled that hypnotically refreshed testimony was admissible. Applying the Davis-Frye standard, the Court of Appeals reversed. The people appealed. In an opinion written by Justice Kavanagh, we unanimously concluded that the Court of Appeals was correct in applying the Davis-Frye test. In response to the prosecutor’s argument that the Davis-Frye rule is applicable only to lie-detector tests, truth serums, drunkenness tests, and narcotics tests, Justice Kavanagh stated:
"The prosecutor’s argument that the Frye rule is inapplicable proceeds from an unduly narrow reading of the opinions invoking the Frye rule. The purpose of this rule is to prevent the jury from relying on unproven [21]*21and ultimately unsound scientific methods.” 415 Mich 623.
In the instant case, and despite our invariant and unanimous application of the Davis-Frye rule to the admissibility of novel scientific evidence, the trial court did not conduct a Davis-Frye hearing before the results of serological electrophoresis testing were admitted into evidence. That was error.7
Before this Court, the people argue variously that the technique of serological electrophoresis is accurate and reliable in scientific literature, has the support of disinterested and impartial experts, and has been accepted in other states. These assertions would be persuasive of general scientific acceptance of the technique’s reliability for identification if they were accompanied by a record that addressed that point. They do not. The limited record information we have before us on which to form an opinion about the reliability of the technique comes from the testimony of the people’s expert witness, Mark Stolorow, a Michigan State Police employee and co-developer of the technique, who testified that he devoted approximately 90% of his work time to using the technique on bloodstain samples received at the State Police Crime Laboratory. We think this to be insufficient to [22]*22determine whether the procedure enjoys general scientific acceptance among impartial and disinterested experts.8 Although we do not doubt that the technique of electrophoresis enjoys general acceptance as a diagnostic and a research tool, the record before us is devoid of impartial and disinterested expert opinion that serological electrophoresis is sensitive and specific in measuring what it purports to measure. People v Barbara, supra. People v Tobey, supra. In addition, the literature and case law cited by the people do not aid us at this juncture because they refer to specific procedures or methods. The record in this case does not. Stolorow’s slide show at trial did go into considerable detail on a method using starch gel as the support medium. He also disclosed during cross-examination that he uses two other methods which employ different support media: an agarosestarch gel mixture and polyacrylamide gel. He testified that he used these three methods to test for five different proteins, the presence or absence of which formed the basis of his trial testimony. We will not speculate, on a matter this grave, that the methods used by Stolorow are those described in the literature and the cases.9
[23]*23The people’s other arguments fall generally into two classes: an attack on the Davis-Frye rule and the assertion that any Davis-Frye error in this case is harmless beyond a reasonable doubt. We treat these arguments in turn.
The people argue that the Davis-Frye rule should be abolished because it hinders early courtroom use of technological advances in the rapidly developing discipline of forensic science. In addition, it is argued, the adoption of MRE 702 by this Court substantially undercuts the necessity for the Davis-Frye rule because the party opposing admissibility will be able to challenge general scientific acceptance during the proponent’s qualification of its expert witness.
We answered the people’s technological advance argument in People v Salvadore Gonzales, supra, where we stated that the purpose of the Davis-Frye rule "is to prevent the jury from relying on unproven and ultimately unsound scientific methods”. We approve the reasoning of the court in United States v Brown, 557 F2d 541, 556 (CA 6, 1977):
"A courtroom is not a research laboratory. The fate of a defendant in a criminal prosecution should not hang on his ability to successfully rebut scientific evidence which bears an 'aura of special reliability and trustworthiness,’ although, in reality the witness is testifying on the basis of an unproved hypothesis in an isolated experiment which has yet to gain general acceptance in its field.”
We stress that the issue is not whether this test, or any test, is an appropriate scientific undertaking. It is whether the inferences to be drawn from the test results are admissible. Inferences are admissible if they are generally accepted by impar[24]*24tial and disinterested experts of the relevant scientific community.
The argument of the superfluity of the Davis-Frye rule in the light of MRE 702 simply misses the point. That rule states:
"If the court determines that recognized scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.” (Emphasis supplied.)
At issue in this case is whether identification evidence derived from serological electrophoretic analysis of blood has achieved general scientific acceptance as being reliable. The Davis-Frye standard is the means by which the court can determine that the novel evidence offered for admission here enjoys such recognition.
Finally, the prosecutor argues that a Davis-Frye error here, if any, was harmless beyond a reasonable doubt because of the other evidence of defendant’s guilt adduced at trial. In the light of the hearing which we order, we conclude that this argument is premature and need not be addressed at this time.
We hold that the admissibility of novel scientific evidence is governed by the Davis-Frye standard. Such evidence must have achieved general scientific acceptance among impartial and disinterested experts. In this case, we are unable to determine from the record before us whether blood analysis identification evidence from the technique of serological electrophoresis is competent evidence. A Davis-Frye hearing should have been held. We remand this case to the trial court for an eviden[25]*25tiary hearing to determine whether the results of serological electrophoresis have achieved general scientific acceptance for reliability among impartial and disinterested experts. We retain jurisdiction.
Williams, C.J., and Kavanagh, Levin, and Ryan, JJ., concurred with Brickley, J.