sults, the court credited testimony that an error in the testing process would mean that there would be no result, rather than a false-positive or false-negative result. The court also credited the efficacy of the laboratory's control runs and approved the use of statistical data to determine the probability of a match.
In Spencer v. Commonwealth,26 the Supreme Court of Virginia affirmed a trial court's finding that evidence derived from RFLP analysis was sufficiently reliable to be admitted. The trial court heard testimony from three experts for the prosecution in molecular biology and genetics. The defense called no expert witnesses. The trial court credited testimony that there is no risk of false positives, that the testing techniques are reliable and generally accepted in the scientific community, and that the particular test was conducted in a reliable manner.
At a later proceeding involving the same defendant, the Supreme Court of Virginia held that evidence based on a sample analysis that used a PCR technology was admissible. In discussing the standard for admitting novel scientific evidence, it rejected the Frye test, asserting instead that the court should make a ''threshold finding of fact with respect to the reliability of the scientific method offered." Without discussing the details of the experts' testimony, the court concluded that the evidence supporting admissibility was credible.27
A Delaware trial court held in State v. Pennell28 that DNA evidence was admissible under a state statute similar to the federal rules, but refused to admit probability statistics. There was no dispute about the underlying theory of DNA typing or its general application in the particular case. The defendant challenged the laboratory's claims that the population databank it used was in Hardy-Weinberg equilibrium and that its "binning process" was valid. The defense held that the state's experts' assessment of the probability of declaring a match was overstated. The court accepted some of the defense contentions and faulted the laboratory for its procedure. The state later introduced new evidence based on the laboratory's revised procedure and a new databank. The court agreed to allow the new evidence if the state would provide the raw data to the defendant, but the state did not do so. The court expressed concern over testimony that the measurements of allele size can depend on who is doing the measuring, and it concluded that the state's evidence did not sufficiently support the probability calculation.
As of February 1991, one federal and 10 state appeals from decisions to admit DNA evidence had been decided. Eight of the state appellate courts upheld trial courts' decisions to admit; the other two approved the scientific theory underlying DNA typing, but one excluded the work of a particular laboratory because of process unreliability, and one found that there was