(dissenting). The instant appeal presents us with a question of apparent first impression, to wit, whether one party to a matrimonial action in which custody of children is contested may compel the other party to submit to a novel and unproven method of testing of hair samples that allegedly can chart a chronology of the latter’s cocaine abuse. Inasmuch as I cannot conclude from the scant record before us that the threshold requirements of reliability and validity of the testing procedure proposed herein have been satisfactorily established, I respectfully dissent and vote to reverse and remit the matter for a hearing on this issue.
As accurately noted by my colleagues, this matrimonial action commenced by the plaintiff wife seeks, inter alia, a judicial determination as to the proper custody of the parties’ two infant children. The defendant husband has opposed the plaintiffs custody request on the ground, among others, that the plaintiff since 1982 has engaged in a continuing, habitual and uncontrollable pattern of cocaine abuse which renders her fitness as a custodial parent highly questionable. In this vein, the defendant moved, inter alia, for an order compelling the plaintiff to submit to the removal of two sets of hair samples from her head so that the samples could be tested for the presence of cocaine and cocaine metabolites. In his supporting affidavit, the defendant stated that he had removed hair specimens from the drains of a sink and bathtub used by the plaintiff and had forwarded them to an expert for testing. According to the affidavit submitted by the expert retained by the defendant to conduct the testing, these specimens were subjected to radioimmunoassay (hereinafter RIA) analysis, which allegedly revealed "high, off-scale readings of cocaine or cocaine-related substances from in the hair itself’. However, the expert claimed that confirmatory testing by means of gas chromatography or gas chromatography/mass spectrometry could not be performed due to the inadequate size of the samples submitted. He therefore requested two larger samples, each "equivalent to the diameter of a pencil”.
*220The plaintiff opposed the motion, conceding that she had engaged in occasional, social use of cocaine in the past with the defendant, but maintaining that she had ceased all use of the narcotic several months prior to the bringing of the defendant’s motion. The plaintiff’s affidavit was accompanied by the sworn statements of two scientific experts, each of whom expressed the view that RIA analysis of hair samples for the detection of cocaine was empirically unproven and not generally accepted as reliable by experts in the field. Additionally, each expert opined that such testing could not supply a time frame for the ingestion of the drug with any reasonable degree of scientific accuracy.
The defendant thereupon submitted reply papers in further support of the motion, contending that RIA testing of various sections of hair samples, when analyzed, in conjunction with the average growth rate of hair, could be employed to provide a "calendar” or "timetable” of the plaintiff’s cocaine abuse, thereby revealing whether she had in fact discontinued her drug use at the time she claimed to have done so. Numerous scientific articles concerning the field of hair analysis were appended to the defendant’s reply papers.
In obvious reliance upon the general policy favoring broad disclosure in civil actions, the Supreme Court, Westchester County, in an order entered December 16, 1987, granted that branch of the defendant’s motion which was for an order directing the plaintiff to submit to the removal of hair samples for testing, observing that "[t]he admissibility of any test results or the weight to be accorded them are questions which the Court will take up at the trial of the action”. By order entered January 7, 1988, the defendant was permitted to have a female representative attend the hair removal procedure and take custody of the samples collected. These appeals by the plaintiff followed. I now vote to reverse the order entered December 16, 1987, insofar as appealed from, and the order entered January 7, 1988, and to remit the matter to the Supreme Court for an immediate hearing on the question of whether the proposed testing procedure will yield valid and reliable results.
Initially, I note my agreement with the general observations of the majority regarding the broad parameters of discovery in civil actions embodied in CPLR article 31 and the similarly broad discretion of the trial court in resolving discovery requests. Likewise, I recognize that liberal discovery is essential in custody disputes, for the overriding concern of the *221court is the best interests of the children, and each party’s physical and mental condition is perforce subject to challenge. Furthermore, I agree with the majority’s conclusion that neither the primary question of whether the results from RIA testing of hair samples for evidence of substance abuse are legally admissible in general, i.e., whether the testing procedure has gained general scientific acceptance (see, Frye v United States, 293 F 1013; People v Tarsia, 50 NY2d 1; People v Leone, 25 NY2d 511), nor the secondary question of whether the specific procedures proposed herein are subject to successful attack by the plaintiff’s experts as defective or empirically unacceptable, need be determined at this juncture.
However, notwithstanding the foregoing, I cannot join in the majority’s view that the plaintiff may be compelled to submit to the physically intrusive procedure outlined above upon the mere principle that policy considerations favor liberal discovery of potentially relevant evidence in actions such as the one at bar. These broad policy considerations are not, in my view, without limits, especially where they compel persons to submit to physical examinations or provide corporeal samples for testing. Similarly, these considerations do not and should not sanction court-ordered examination or testing by bizarre and unrecognized methods. In short, our courts cannot compel litigants to become unwilling participants in wholly experimental exercises. Rather, I find that before the plaintiff may be so compelled, there must be a reasonably adequate threshold demonstration that the testing procedure proposed by the defendant’s expert will yield information which is not only relevant, but scientifically valid and reliable as well (see generally, Matter of Abe A., 56 NY2d 288; Matter of Barber v Rubin, 72 AD2d 347).
Indeed, in Matter of Abe A. (supra), the Court of Appeals was confronted with the question of whether a homicide suspect, not under arrest, could be compelled, within the restrictions of the Fourth Amendment, to submit to the drawing of a blood sample for the purpose of conducting an investigatory blood-grouping analysis. The court determined that a judicial mandate requiring such a submission could not properly issue unless the prosecution first established (1) probable cause to believe that the suspect committed the offense, (2) a clear indication that relevant material evidence would be elicited by the challenged procedure, and (3) that the method to be employed would be both safe and reliable. From the evidence presented therein, the court reasoned that the *222suspect could be compelled to provide a sample for the test because the probable cause and relevancy criteria had been satisfied and the validity, reliability and safety of the blood-grouping procedure "are well recognized in both the medical and legal communities” (Matter of Abe A., supra, at 299).
Our own earlier decision in Matter of Barber v Rubin (72 AD2d 347, supra) bears an even greater factual resemblance to the case at bar. There, the Supreme Court initially granted an application by the prosecution to require a criminal defendant to submit to the extraction of hairs from his head for comparison with specimens of hair recovered from the clenched fists of a homicide victim. The accused brought a proceeding pursuant to CPLR article 78 in the nature of a writ of prohibition, which proceeding was held in abeyance pending an evidentiary hearing to determine the necessity, degree of invasion, potential harm and probative value of the procedure (Matter of Barber v Rubin, 65 AD2d 811). At the hearing, a police chemist, a physician, an agent of the Federal Bureau of Investigation, and a police detective gave testimony with respect to these factors, thereby providing a record from which we could reasonably conclude, as did the hearing court, that an expert in hair comparison equipped with proper hair samples could determine with a reasonable degree of certainty from a microscopic examination whether the hairs taken from a known source matched those removed from an unknown source, and that the close similarity of the hair specimens indicates with a high degree of probability that they originated from the same source. Based upon these and other findings, we concluded that the accused could be directed to submit to the procedure because, among other things, the reliability and validity of hair comparison analysis had been sufficiently demonstrated. However, we expressly left open the question of whether such evidence was admissible, instead confining our discussion only to those factors which were relevant to a determination of whether there was an adequate scientific basis for requiring the accused to submit to the test.
Unlike the majority, I believe that the application of a similar standard is appropriate herein. While it is true that the Fourth Amendment considerations voiced in the two aforementioned decisions are not implicated in the matter before us due to the civil nature of this proceeding and the placing of the parties’ physical and mental conditions in issue, there are no less compelling reasons which warrant a determination as to the relevancy, reliability, and validity of the *223procedure challenged herein before the plaintiff can be required to submit to it. Indeed, the use of such a new and experimental testing procedure should be viewed with particular caution in matrimonial actions, for while liberal discovery, including physical examinations pursuant to CPLR 3121, is encouraged in such actions, the potential for abuse is substantial (see, e.g., Lohmiller v Lohmiller, 118 AD2d 760; Rosenblitt v Rosenblitt, 107 AD2d 292). Hence, parties to matrimonial actions must rely upon the vigilance of the courts to safeguard them from abusive discovery tactics. As the Court of Appeals has observed: "CPLR 3121 does not prohibit such [physical] examinations in matrimonial actions, and although we recognize the potential for abuse in these cases, the court’s broad discretionary power to grant a protective order 'to prevent unreasonable annoyance, expense, embarrassment, disadvantage, or other prejudice to any person or the courts’ (CPLR 3103) should provide adequate safeguards” (Wegman v Wegman, 37 NY2d 940, 941). In my view, the instant plaintiff is entitled to the minimal safeguard of a preliminary hearing and judicial determination regarding the reliability and validity of the proposed hair analysis, as the record before us falls woefully short of providing an adequate foundation upon which to base such a decision.
The defendant contends that a mere showing of relevancy alone is sufficient to justify the order directing the plaintiff to comply with his novel discovery request, and that questions of reliability and validity must be left for a subsequent hearing on the issue of admissibility. I cannot agree where, as here, the scientific analysis to be employed is as yet unproven and experimental (see, Matter of Barber v Rubin, 72 AD2d 347, supra). The defendant’s relevancy argument is twofold in nature. He first claims that the challenged test is effective in detecting the presence of cocaine and cocaine metabolites in human hair, such that a positive result in this case would indicate that the plaintiff has ingested cocaine at some point in the past. While there is some scientific evidence in the record to support this claim, I fail to perceive the relevance of such a finding, as the plaintiff has already admitted her previous use of cocaine. Insofar as the challenged procedure detects the mere presence of the narcotic, a positive result would constitute nothing more than cumulative proof of the plaintiff’s conceded past cocaine abuse. Therefore, if the defendant seeks this test to confirm the plaintiff’s past use of cocaine, it is not relevant. Relevant evidence has been defined *224as that which "logically renders the existence of a material fact more likely or probable than it would be without the evidence” (Richardson, Evidence § 4 [Prince 10th ed]). Material evidence is "evidence of a fact in issue or evidence probative of a fact in issue. If the evidence is relevant to establish or disprove a proposition, but the proposition is neither in issue nor probative of a fact in issue, it is immaterial” (Richardson, Evidence § 4 [Prince 10th ed]). The definition of relevant evidence has been synthesized by modern scholars to mean "evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence” (Proposed NY Code of Evidence § 401 [1982]). Clearly, the plaintiffs past use of cocaine, allegedly with the defendant, is conceded and hence, is not in issue. Thus, the defendant’s first claim of relevance must fall.
A more interesting issue is raised by the second prong of the defendant’s relevancy contention, i.e., that RIA analysis of human hair may actually be used as a "calendar” to provide a chronology of the plaintiffs cocaine ingestion. Significantly, the defendant does not seek to establish any alleged current use of cocaine by the plaintiff by means of this testing procedure. Indeed, it is clear that blood and/or urine tests, scientifically established procedures to which the plaintiff has suggested she would willingly submit, would provide far more accurate data regarding cocaine use within the 48-to-72-hour period immediately preceding their administration. Rather, the defendant wishes to demonstrate that the plaintiff ingested cocaine at some point subsequent to the time at which she claims to have ceased doing so and to thereby attack her credibility. While this aspect of the defendant’s discovery request might have some impact on the question of the plaintiffs truthfulness, there is little scientific support in the present record for the reliability and validity of such a timetable analysis. The threshold question still remains: would this test and its interpretation logically render the existence of past cocaine ingestion at a particular time more likely or probable? The defendant has submitted only a small handful of experimentation reports suggesting that the "calendaring” of substance abuse is possible through RIA testing of human hair samples, and other scientific literature supplied by the parties emphasizes the "experimental” and uncertain nature of hair analysis in general. Moreover, the plaintiffs experts hotly dispute any claims that the proposed testing can reliably *225and accurately measure the time period of drug ingestion or the quantities of the substance ingested. Additionally, the experts for both sides agree that the field of scientific testing of human hair samples for evidence of substance abuse is still in its infancy, and my own research has failed to disclose a single reported decision of any court which has even discussed, much less accepted, the reliability and validity of the purported calendaring aspects of RIA testing of human hair. It is also noteworthy that there is no evidence that the confirmatory tests proposed by the defendant’s expert, gas chromatography/mass spectrometry, can yield this "calendaring” information. Given the strong disagreement among experts and the paucity of proof in this record to support the reliability and validity of the proposed procedure, I cannot in good faith or good conscience acquiesce in compelling the plaintiff to submit to what can only be characterized as a new and experimental testing procedure. Instead, I conclude that the defendant must first be required to make a threshold showing of the scientific reliability and validity of the proposed test at a hearing in which expert evidence from both sides may be adduced. Only then should the defendant’s discovery request be granted or denied in accordance with the hearing evidence.
I emphasize that I am not presently concerned with the question of the potential admissibility of any test results, an inquiry which explores the degree of general acceptance of the challenged testing procedure within the scientific community. I merely conclude that the present record provides inadequate evidence to establish the reliability and validity of the proposed test, characteristics which in my view constitute essential prerequisites to any order compelling the plaintiff herein to submit to the new and experimental hair analysis procedure heretofore discussed. Hence, I would remit the matter for a prompt and thorough hearing on these issues (see, Matter of Barber v Rubin, 65 AD2d 811, supra).
Finally, I conclude that the court compounded its error by improperly providing that a female representative of the defendant is to take custody of the plaintiff’s hair samples in its order entered January 7, 1988. The basis for that order was the court’s desire "to ensure an uninterrupted chain of custody”. However, even if I were to agree that the order directing the plaintiff to submit the samples was appropriate, the record is devoid of any evidence to support the conclusion that the appointment of a representative of the defendant is *226necessary or appropriate to preserve the chain of custody of the samples. Absent such a showing, I conclude that the order entered January 7, 1988, constituted an improvident exercise of discretion and that the chain of custody will be adequately safeguarded by permitting the disinterested physician who removes the hair specimens to take custody of them.
Accordingly, the order entered December 16, 1987, should be reversed insofar as appealed from, and the order entered January 7, 1988, should be reversed, and the matter should be remitted to the Supreme Court, Westchester County, for a hearing and new determination as to that branch of the defendant husband’s motion which was for an order directing the plaintiff to appear at the offices of a physician and submit samples of her hair for testing.
Kunzeman, J. P., and Rubin, J., concur with Balletta, J.; Sullivan and Kooper, JJ., dissent in a separate opinion by Sullivan, J., who dissents and votes to reverse the order entered December 16, 1987, insofar as appealed from, to reverse the order entered January 7, 1988, and to remit the matter to the Supreme Court, Westchester County, for a hearing and new determination as to that branch of the defendant husband’s motion which was for an order directing the plaintiff to appear at the offices of a physician and submit samples of her hair for testing.
Ordered that the order entered December 16, 1987 is affirmed insofar as appealed from, and the order entered January 7, 1988, is affirmed, with one bill of costs.