BUTZNER, Circuit Judge:

Carl J. Baller, Jr., was convicted by a jury of four counts of telephoning bomb threats. The principal issue on appeal is the admissibility of expert testimony identifying his voice by spectrographic analysis. We hold that the evidence was properly admitted and affirm the judgment.

Baller was an employee of the Shoemaker Coal Mine in Benwood, West Virginia. During 1972 and early 1973, the mine received anonymous, telephoned bomb threats. In an effort to detect the caller, mine clerks tape-recorded four bomb threats, and the telephone company traced the fourth call to Baller's residence. Baller acknowledged using the telephone at the time the fourth call was made but said he had been talking to a girlfriend. Later he provided tape-recorded voice exemplars for comparison with the recorded calls.

Tapes of the voice exemplars and the four bomb threats were submitted to Detective Lieutenant Ernest W. Nash of the Michigan State Police for voice spectrograph analysis. Lt. Nash testified that Baller's voice positively matched that of the person who made the first three of the recorded threats. He could not, however, identify the speaker on the tape of the call that was traced to Baller's home because the mine clerk, following instructions to keep the phone open so the trace could be completed, had constantly interrupted the caller and dominated the conversation. Baller objected to the admission of the testimony primarily on the grounds that the spectrographic analysis was not of sufficient scientific reliability and that Lt. Nash was not properly qualified as an expert.

I

Before discussing the principal issues, we must consider two preliminary claims. Baller's first contention, that the mine clerks who made the recordings were not qualified to operate the machine, is without merit. The man who installed the recording device testified that all an operator had to do was turn it on when the phone rang. The clerks who recorded the threats testified that they properly turned on the machine, and they verified the conversations when the tapes were played in open court. No other proof of their accuracy was needed.

Baller also contends there was no proof that the technical quality of the tapes was sufficient for accurate spectrographic analysis. However, Lt. Nash distinguished the three usable tapes from the one he found unsuitable. His testimony is not impeached or contradicted. Once the recordings were verified for accuracy, their technical quality for spectrographic analysis was properly established by one learned in that technique.

II

The admissibility of voice spectrogram analysis, sometimes called voiceprints, as evidence of a speaker's identity has been contested frequently in recent years in both state and federal courts. The scientific principles of the technique have been so exhaustively chronicled that we need only summarize them. See, e. g., Commonwealth v. Lykus, Mass., 327 N.E.2d 621 (1975).

The type of spectrograph used in this case reduces the spoken word to a pattern of light and dark lines representing frequency, duration, and amplitude. Lt. Nash testified that identifications are made by visually comparing the spectrograms of known and unknown voices speaking the same word and then listening to both voices. He can identify positively, he said, when ten pairs of sounds from known and unknown voices match visually and aurally. He explained that although successive spectrograms of the same speaker saying the same thing are not identical, the same pattern appears in both. He demonstrated as an example of this continuity the spectrograms of the word "bomb" from the exemplar and three of the threats, which he had identified as Baller's.

Spectrographic identification rests on the proposition that the organic and acquired characteristics of an individual voice, despite its variations, depict visual representations of the same sounds from the same speaker that are so different from those of all other speakers that a positive identification or exclusion can be made and objectively demonstrated. See People v. Law, 40 Cal.App.3d 69, 114 Cal.Rptr. 708, 712 (1974). The admissibility of spectrographic identification turns primarily on whether this theory has been sufficiently proved to allow a jury to give the evidence whatever weight it sees fit. Most of the earlier cases excluded the evidence on the ground that the technique had not been adequately tested under field conditions. See, e. g., State v. Cary, 99 N.J.Super. 323, 239 A.2d 680 (1968), aff'd per curiam, 56 N.J. 16, 264 A.2d 209 (1970); People v. King, 266 Cal.App.2d 437, 72 Cal.Rptr. 478 (1968); contra, United States v. Wright, 17 U.S.C. M.A. 183, 37 C.M.R. 447 (1967). Since these decisions, however, extensive experiments on voice spectrography has been conducted at Michigan State University by Dr. Oscar Tosi, with the participation of Lt. Nash. It appears from the results, published in 1971, that an experienced operator would incorrectly identify a speaker approximately six percent of the time. Subsequently, Dr. Tosi suggested refinements, employed by Lt. Nash, which reduce the percentage of mistaken identifications to about two percent. See Commonwealth v. Lykus, Mass., 327 N.E.2d 671, 676 (1975); Tosi, "Michigan State University Voice Identification Project," Voice Identification Research 35, 57-58 (L.E.A.A. 1972). One prominent early critic of spectrographic analysis, Dr. Peter Ladefoged, has since testified in support of admitting this means of identification of an undisguised male voice.

Two United States Courts of Appeals have recently expressed opposite views on the admissibility of spectrograms. In United States v. Addison, 162 U.S.App.D.C. 199, 498 F.2d 741, 743-45 (1974), the court concluded that the technique had not gained "the general acceptance in the particular field to which it belongs" required by Frye v. United States, 54 App.D.C. 46, 293 F. 1013-14 (1923). The Sixth Circuit, on the other hand, differed with this conclusion and allowed the evidence of spectrographic identifications to be admitted. United States v. Franks, 511 F.2d 25 (6th Cir. 1975); cf. United States v. Sample, 378 F. Supp. 44, 51-54 (E.D.Pa. 1974). A majority of state courts which have considered the question since the Tosi study favor admissibility. Significantly, this group includes New Jersey, which previously had excluded such evidence. See Hodo v. Superior Court, 30 Cal.App.3d 780, 106 Cal.Rptr. 547 (1973); Alea v. State, 265 So.2d 96 (Fla.App. 1972); Worley v. State, 263 So.2d 613 (Fla.App. 1972); Commonwealth v. Lykus, Mass., 327 N.E.2d 671 (1975); State ex rel. Trimble v. Hedman, 291 Minn. 442, 192 N.W.2d 432 (1972); State v. Andretta, 61 N.J. 544, 296 A.2d 644 (1972). Voice spectrographic identification was excluded in People v. Law, 40 Cal.App.3d 69, 114 Cal.Rptr. 708 (1974), which, however, involved the identification of disguised voices, a problem on which there has been insufficient research.

There are good reasons why not every ostensibly scientific technique should be recognized as the basis for expert testimony. Because of its apparent objectivity, an opinion that claims a scientific basis is apt to carry undue weight with the trier of fact. In addition, it is difficult to rebut such an opinion except by other experts or by cross-examination based on a thorough acquaintance with the underlying principles. In order to prevent deception or mistake and to allow the possibility of effective response, there must be a demonstrable, objective procedure for reaching the opinion and qualified persons who can either duplicate the result or criticize the means by which it was reached, drawing their own conclusions from the underlying facts. See United States v. Addison, 162 U.S. App.D.C. 199, 498 F.2d 741 (1974); United States v. Amaral, 488 F.2d 1148 (9th Cir. 1973); United States v. Stifel, 433 F.2d 431 (6th Cir. 1970).

Deciding whether these conditions have been met is normally within the discretion of the trial judge. United States v. Brumley, 466 F.2d 911 (10th Cir. 1972); Fineberg v. United States, 393 F.2d 417 (9th Cir. 1968). Absolute certainty of result or unanimity of scientific opinion is not required for admissibility. "Every useful new development must have its first day in court. And court records are full of the conflicting opinions of doctors, engineers, and accountants, to name just a few of the legions of expert witnesses." United States v. Stifel, 433 F.2d 431, 438 (6th Cir. 1970). Unless an exaggerated popular opinion of the accuracy of a particular technique makes its 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. United States v. Stifel, supra; Coppolino v. State, 223 So.2d 68 (Fla.App. 1968); see McCormick, Evidence § 203 at 490-91 (2d ed. 1972).

In this case, the district court adequately guarded against dangers inherent in the use of newly developed scientific tests. Lt. Nash has as much experience in forensic voice spectrography as anyone; he is one of the pioneers of the technique and is a qualified practitioner of it. The evidence presented in an extensive voir dire demonstrated spectrography's probative value, despite doubts within the scientific community about its absolute accuracy. Competent witnesses were available to expose its limitations, and the defense was furnished with the names of other experts who could conduct their own analyses of the tapes. Although the defense did not call any expert witnesses, Baller's attorney demonstrated thorough knowledge of the subject in a detailed cross-examination that developed the possibility of error in both the general technique and the specific identification of the defendant. Moreover, the tapes of Baller's voice exemplars and all of the bomb threats were played so that the jury could make its own aural comparisons. Finally, the court instructed the jury that the spectrograms were only a basis for Lt. Nash's opinion and that they could disregard his testimony if they decided that his opinion was not based on adequate education or experience or that his "professed science of voiceprint identification" was not sufficiently reliable, accurate, and dependable. The court also cautioned the jury that they need not accept his opinion if they believed the reasons supporting it were unsound or if contradictory evidence cast doubt on it. Cf. 1 Devitt and Blackmar, Federal Jury Practice and Instructions § 11.27 at 241 (2d ed. 1970). The reliability and credibility of Lt. Nash's opinion were opened to refutation, and the question of its weight was fairly presented to the jury. We find no abuse of discretion in the admission of the testimony.

The judgment is affirmed.