(dissenting):
I would deny the government’s appeal. On appeal, we should reverse a military judge’s ruling on the production of evidence only if the ruling is clearly erroneous. That is the standard for a review of findings of fact.1 In my view, the question as to whether any particular evidence is necessary for an adequate defense rests upon a factual determination. In this case, I not only find his decision far short of being clearly erroneous, I find it to be *1062correct. In reviewing the findings of fact made by the military judge, I find his explanation of the necessity for the additional tests clear and convincing. He ordered the tests based on the following rationale:
(1) The defense in both cases is, essentially, “I never used cocaine, and I don’t know how this cocaine metabolite (BE) got in my urine sample.”
(2) Testing for BE, EME, and cocaine is a relatively simple procedure.
(3) The accused were financially unable to pay for the testing.
(4) The tests under the right circumstances could show whether the cocaine in the urine was ingested by the accused or was placed in the urine from an outside source.
The military judge was obviously concerned about the limited means available to an accused in urinalysis cases to contest guilt. He was no doubt troubled, in these cases, by the expert testimony which explained that a positive reading for the metabolite BE does not conclusively show that cocaine was ingested by the person providing the sample. United States v. Mack, 33 M.J. 251 (C.M.A.1991). Yet the government was contending that the additional tests were not necessary since the urinalysis results and the uncontested chain of custody were alone sufficient to support a conviction. United States v. Harper, 22 M.J. 157 (C.M.A.1986).
The military judge was correct in concluding that the evidence which the defense was requesting was both relevant and necessary. R.C.M. 703(a) provides that “the prosecution and defense shall have equal opportunity to obtain witnesses and evidence.” In my view, the majority opinion overlooks the military judge’s responsibility in determining what evidence the court should have before it. R.C.M. 801(c). There can be little dispute that if the evidence produced by the retesting is as the defense hopes, it would completely undermine the government’s case.2 As a matter of fundamental due process, an indigent accused must be provided “the basic tools of an adequate defense.” United States v. Van Horn, 26 M.J. 434 (C.M.A.1988).
A search for truth lies at the heart of our criminal justice system. When not inconsistent with the rules of evidence, judicial economy or other compelling reason, the interests of justice weigh in favor of insuring that factfinders have access to all available evidence in their search for truth. The military judge is in the best position to determine the need to have this evidence before the court. The military judge here not only heard the expert testimony but is also, depending upon the forum chosen by the defense, potentially the factfinder.
I am unable to find what standard the majority uses in determining when requests of this nature are necessary. It’s clear they conclude that the information sought is not necessary in order for the government to present a prima facie case. But it’s not clear why, as a matter of law, it is error for the judge to require the government to perform the retests. They seem to say that the defense has other means of ensuring an adequate defense, for example, by attacking the reliability of the test procedure and by cross-examination of testing personnel. However, these means would not necessarily reveal the one thing that these accused are hanging their hats on, namely, that somehow their samples were tampered with.
The majority opinion suggests several things that the retests would not show and concludes that the only possible benefit to be gained by the defense from the requests would happen only if raw cocaine had somehow found its way into the urine samples. However, the experts who testified at the Article 39(a) session pointed out that the additional tests would disclose additional information about when, how, and how much cocaine entered the urine. The following colloquy took place during the questioning of Dr. Donald Lee Frederick, Chief *1063of Quality Assurance, Air Force Drug Testing Laboratory:
Q. And do you also agree with Dr. Vasiliades that if you perform the EME test and the test for cocaine that the results of those tests would tell you a lot more things than just the BE test? For instance, would it narrow the window of when the cocaine was used and how much was used?
A. Yes. They may narrow the window. They may not answer the questions specifically and most likely would not because of the variability between individuals and sets of circumstances. They would give you more information than you have with the result of one test.
When an accused is saying “I never used cocaine, and I don’t know how this cocaine metabolite (BE) got in my urine sample,” the information provided by the retests would be essential for the preparation of a defense. The tests would definitely establish whether the sample had been “spiked” with raw cocaine and whether or not the cocaine had passed through the body. They may also allow the experts to testify concerning the amount of cocaine consumed and when the cocaine was used in relation to when the sample was provided. Certainly, if unknowing ingestion is a defense, this information would be exceedingly useful. It is information that the finders of fact should have and there is no other way for an innocent, indigent accused to obtain this data.
Analogizing these requests to requests for investigative or expert assistance, the majority opinion concludes that the defense must demonstrate that the retests are necessary for an adequate defense. I find the majority’s analogy particularly troubling since these are urinalysis cases. The ability of an accused to prepare a defense in urinalysis cases is uniquely different than in other criminal cases. The evidence can be taken from the accused’s body without the requirement of a probable cause search. The chain of custody is maintained by personnel who are not trained in law enforcement. The evidence is solely scientific. The results of the testing are normally unavailable for several weeks after the sample is provided. All of these factors weigh in favor of requiring the government to provide an accused who professes innocence with the information needed to prepare a reasonable defense.
Presumably, the majority would require the defense to first attack the chain of custody in order to show any necessity for the retests. They note the chain of custody documentation is unblemished in these cases. I do not see the fairness in forcing an accused to attack the chain of custody when the test relied upon by the government is itself insufficient to scientifically show that the drug was ingested.
Accordingly, I disagree with the legal gymnastics engaged in by the majority opinion in concluding the military judge erred as a matter of law. The majority criticizes the trial judge’s analysis of R.C.M. 703(f)(1), which indicates relevant evidence is necessary “when it would contribute to a party’s presentation of the case in some positive way on a matter in issue.” They suggest his analysis is flawed because chain of custody was never put “in issue.” I submit the military judge correctly sees the “matter in issue” in these cases as whether the accused used cocaine and not whether the chain of custody is proper. His decision to compel evidence which might contribute to the resolution of the issue of whether cocaine was consumed should not be overturned.
Finally, I dissent from the majority’s conclusion that the military judge erred by not treating the defense requests as ones for the government’s investigative assistance rather than as requests for additional evidence. This distinction lacks any legal significance. In either case, the judge’s decision will rest on a finding of facts of the necessity for the requests. Moreover, I do not agree that these requests seek governmental assistance in “finding” a defense. The accused are not seeking to find a defense but rather seek evidence which would be consistent with their denials of cocaine *1064use. Based on the expert testimony given at the Article 89(a) session, the probability that the test results will match the hopes of the accused appear somewhat remote. Nevertheless, regardless of the outcome, the results would be relevant and admissible. As a matter of due process to insure a fair trial, I would uphold the military judge’s decision ordering the retests.3
. United States v. Burris, 21 M.J. 140 (C.M.A. 1985).
. If the results of the retests were positive for cocaine, negative for the metabolite EME and positive for the metabolite BE, expert witnesses would conclude that the person who provided the sample had not used cocaine. The logical conclusion would be that the sample had been contaminated by an outside source.
. I further think this case shows why the dissent in United States v. True, 28 M.J. 1 (C.M.A.1989) is the better view. To permit the government to appeal rulings by the military judge which do not terminate the proceedings significantly diminishes the authority of the military judge. It's significant to note that Congress specifically authorized an appeal from a ruling which excludes evidence, yet we are now faced with an appeal from a ruling that seeks additional evidence. The ruling did not terminate the proceedings; rather, it was the government's refusal to comply with the order to produce additional evidence which forced the abatement. I find it difficult to imagine that Congress intended appellate courts to become involved with this circumstance.