United States v. Bush, 44 M.J. 646

44 M.J. 646 | A.F.C.C.A. | 1996

OPINION OF THE COURT

MORGAN, Judge:

This appears to be a case of first impression for federal criminal jurisprudence, where we are asked to decide whether the trial judge erred in admitting chemical hair analysis to support appellant’s conviction for the unlawful use of cocaine. We hold that he did not.

On November 15, 1998, appellant was selected to provide a sample for a random drug urinalysis. He showed up at the base theater, as he was directed to do. Thereafter, accompanied by Technical Sergeant (TSgt) Robichaud, the observer, appellant took the sample bottle to the men’s room to provide a specimen. A number of irregularities, the significance of which was not then appreciated, ensued. Appellant elected to use a stall, rather than a urinal, complicating Robichaud’s observation. He continued to wear his field jacket, and had to be directed by Robichaud to turn slightly so that Robichaud could see him urinate. Even then, as Robichaud later admitted, appellant was positioned in such a way that he did not actually see if appellant urinated, but only that liquid appeared to be filling the bottle.¹ Robichaud, and subsequently the urinalysis monitor, remarked on the clarity of the specimen. However, when asked, appellant explained that he had been drinking a great deal of fluid, and the matter passed. The specimen bottle, which had not left appellant’s possession from the time he was given it until he gave it back to the urinalysis monitor, was duly logged, initialed, signed, taped, secured, and mailed to the Armstrong Laboratory at Brooks Air Force Base (AFB) for testing. Once there, one of the laboratory technicians observed that the specimen was colorless, odorless, and did not foam when shaken. She suspected that a false or adulterated sample had been provided.. When a field test indicated that the specimen was not urine, she sent it to Wilford Hall Medical Center, which confirmed that the specimen was not urine, but some sort of saline solution.

Over defense objection, the government introduced evidence that appellant, a medical technician assigned to the Malcolm Grow Medical Center on Andrews Air Force Base (AFB), had access to intravenous bags containing saline solution, along with surgical tubing and a thumbscrew to control flow from the bags. As well, testimony indicated that appellant was capable of reverse self-eatheterizing, replacing the urine in his bladder with a saline solution. Regardless of the specific mechanism employed, it is clear that appellant did not provide a genuine urine specimen as he was required to do.

On learning of the discrepancy in early December, the base looked into the possibility of testing appellant’s hair for the presence of drugs. Special Agent (SA)' Toni, of the Air Force Office of Special Investigations (AFOSI), contacted the FBI’s forensics laboratory, and was advised that the technology existed to test hair based upon the same biomedical and scientific principles as urinalysis. The advantage, he learned, was that hair potentially would continue to show the presence of cocaine for a period of months after ingestion. The FBI agreed to perform the tests. Using an example borrowed from the AFOSI at Langley AFB, Virginia, SA Toni then prepared an affidavit, stating in pertinent part as follows:

4____ As a result of your affiant’s training and information gathered from the Federal Bureau of Investigation (FBI) forensics laboratory, and the Brooks AFB forensics laboratory, your affiant believes trace amounts of drugs may be trapped in the cortex of BUSH’s hail (sic) follicles and in his urine. This is based on the following:

a. As blood circulates through the body, it nourishes the hair follicle. If there are drugs in the blood, trace amounts of the drug become entrapped in the core of the hair in amounts roughly proportional to those ingested. These cannot be washed or flushed out, and do not diminish with time. Urine tests can only determine if drugs have been used within the few days prior to providing a sample, however, hair analysis can detect the use of drugs for months, depending on the length of the hair sample.

b. Hair analysis is not subject to false negatives due to temporary abstention or excessive fluid intake. Hair records drug use in a chronological manner and in proportion to the amount consumed. The FBI laboratory can distinguish between heavy, medium, and light drug users.

5. If drug metabolites are present in BUSH’s hairs, at a level in excess of 3 ng/mg of hair, it would indicate repeated use of drugs.

6. Based on all the information provided above, your affiant requests authorization to seize approximately 100 hairs and a urine sample from the body of SSgt MICHAEL W. BUSH.

The search authority, Colonel Moore, swore SA Toni to the affidavit and granted authority to seize “approximately, 100 hairs,” but did not authorize seizure of appellant’s urine. Pursuant to that authority, approximately 100 hairs were cut from the crown of appellant’s head. Although never precisely measured, there was a general consensus that appellant’s hair was “quite short,” and that the hairs measured approximately 1/2 inch in length. Observing the same chain of custody procedures employed in urinalysis drug testing, the hairs were placed into a bottle, sealed, and sent to the FBI laboratory. By letter of February 28, 1994, the FBI reported that the specimens contained “cocaine and its metabolite, benzoylecgonine at concentrations of 17 and 2.7 nanograms per milligram of hair, respectively.”²

Based upon this evidence, a general court-martial consisting of members convicted appellant, contrary to his pleas, of dereliction of duty for failure to provide a urine specimen on November 15, 1993, and use of cocaine between on or about November 15,1993, and January 12, 1994. Appellant now contends that the military judge erred in admitting the hair analysis for two reasons. First, he argues that the seizure of the hairs itself was unlawful because the probable cause (appellant’s November 15 substituted urine specimen) preceded by nearly two months the taking of the hair, and second, he argues that hair analysis testing does not pass muster under Mil. R. Evid. 702. Finally, he argues that the evidence was legally and factually insufficient to support his conviction for both charges. We are unpersuaded by appellant’s arguments, and affirm.

Seizure of Appellant’s Hair

The first prong of appellant’s attack on the hair analysis stems from the seizure of the hair itself. It was conceded by both sides at trial that hair grows at approximately the rate of 1/2 inch per month. Hence, appellant continues, hair which was only 1/2 inch long when it was seized nearly two months after the probable cause triggering incident, could not possibly reveal whether he had drugs in his system on November 15, 1993. Where the information which informs a search authorization is stale, appellant concludes, the search authorization is fatally infirm. See, e.g., United States v. Poole, 30 M.J. 271, 274 (C.M.A.1990).

Appellant’s analysis erroneously presumes that an individual facing a valid, random urine inspection, governed by Mil. R. Evid. 313(b), may by his own misconduct frustrate that inspection and require the government to produce probable cause for any subsequent search or seizure, governed by Mil. R. Evid. 315 and 316. Baldly put, appellant argues that he should profit by the delayed discovery of his subterfuge. That defies common sense, and it is not the law. It is well-settled that the refusal to submit a urine specimen, or submission of a substituted specimen, justifies a subsequent order to submit a valid specimen, and that the subsequent order stands on the same legal footing as the original. United States v. Streetman, 43 M.J. 752 (A.F.Ct.Crim.App.1995) (initial refusal to provide random urinalysis specimen, followed by direct order to do so, did not change requirement into an inadmissible, “command directed” urinalysis); United States v. Nand, 17 M.J. 986 (A.F.C.M.R.1984) (submission of tap water for medically required urinalysis, followed by order to provide another specimen; second specimen still qualified as intrusion for valid medical purpose notwithstanding it was also used to screen for drugs); cf., United States v. Moeller, 30 M.J. 676 (A.F.C.M.R.), pet. denied, 32 M.J. 14 (C.M.A.1990) (entire random urinalysis collection lost in mail, subsequent order to reconduct test not “command directed”).

Even if we take the analytical path appellant prefers, his argument fails. Appellant does not deny that his substitution of saline solution for urine provided a reasonable basis from which a search authority could conclude that he might be using drugs, nor does he argue that Colonel Moore’s search authorization was itself defective. Instead, he takes aim at the affidavit informing that authorization, contending that the failure of SA Toni to advise the search authority that appellant’s hair might not be long enough to provide assurance that drugs taken on or before November 15 would still be found in it was tantamount to a “false statement knowingly or intentionally made or with reckless disregard for the truth” thereby vitiating the good-faith exception of Mil. R. Evid. 311(b)(3). See also United States v. Figueroa, 35 M.J. 54, 56 (C.M.A.1992) (omissions of information should be analyzed in the same manner as misrepresentations), cert. denied, 507 U.S. 910, 113 S.Ct. 1257, 122 L.Ed.2d 655 (1993).

At worst, the omission of the growth rate of hair in SA Toni’s affidavit was merely negligent.³ SA Toni was a new AFOSI agent, still undergoing his first year of probationary training. Not unreasonably, he elected to use a “proven” form of affidavit from another AFOSI office. Further, Colonel Moore authorized him to seize any hair; he was not limited to that on appellant’s skull. That he elected the least intrusive sample, ignoring the possibility of longer pubic or body hairs, if anything, redounded to appellant’s benefit. In any event, a mere negligent omission will not serve to undermine probable cause. Figueroa, 35 M.J. at 56-7. See also United States v. Lopez, 35 M.J. 35 (C.M.A.1992) (failure to mention fact that information provided to search authority was five weeks old not bad faith).

Admissibility of Hair Analysis

We review a military judge’s decision to admit or not admit expert testimony on an abuse of discretion standard. United States v. Nimmer, 43 M.J. 252 (1995); United States v. Houser, 36 M.J. 392 (C.M.A.), cert. denied, 510 U.S. 864, 114 S.Ct. 182, 126 L.Ed.2d 141 (1993); United States v. Thomas, 43 M.J. 626, 632 (A.F.Ct.Crim.App.1995). On that basis, we hold that the military judge, who made extensive findings of fact and conclusions of law, did not abuse his discretion in admitting the results of appellant’s hair analysis.

The admissibility of scientific evidence has, since the watershed case of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), provided fecund soil for a proliferation of law review articles, evidentiary analyses, and scholarly comment. We need not add substantially to it here. Daubert did no more than apply Federal Rule of Evidence 702, which is identical in all material respects to Mil. R. Evid. 702. With United States v. Gipson, 24 M.J. 246 (C.M.A.1987), the Court of Military Appeals, now the Court of Appeals for the Armed Forces, anticipated the Daubert holding by six years, and established the legal syllogism by which we approach the question of the admissibility of scientific evidence.

Both Daubert and Gipson sensibly began their analysis by looking to the philosophic undergirding of the appropriate rules of evidence. Rule 402, like its federal counterpart, announces "with disarming simplicity that all relevant evidence is admissible except as otherwise declared. This presumption of admissibility reflects the broader policy choice by the drafters that relevant evidence ought to be provided to the trier of fact unless there is good reason not to. This theme carries over to Mil. R. Evid. 702 which, read along with Mil. R. Evid. 401-403, describes a “comprehensive scheme for processing expert testimony.” Nimmer, 43 M.J. at 255. Rule 702 requires only that “scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue____” This relatively liberal rule of admissibility supplants, if it does not eliminate altogether, the “general acceptability” litmus of Frye v. United States, 293 F. 1013 (D.C.Cir.1923).

Our most recent guidance from the Court of Appeals for the Armed Forces on scientific or expert testimony, coincidentally, concerns hair analysis. In Nimmer II the Court reversed the Navy-Marine Corps Court of Criminal Appeals’ published affirmance (United States v. Nimmer, 39 M.J. 924 (N.M.C.M.R.1994))⁴ of a trial judge’s decision- to exclude exculpatory hair analysis evidence. The Nimmer II majority reasoned that Daubert, which had not been decided at the time of Nimmer’s trial in 1992, sufficiently enlarged upon the rationale in Gipson as to justify remand of the ease for further consideration. Tempting as it is, we cannot on that basis conclude (although we may suspect) that the Court of Appeals for the Armed Forces is persuaded of the legal vitality of hair analysis. Substantive differences between the factual, legal, and even scientific contexts in Nimmer and the case sub judice necessitate further technical and logical discourse.

Nimmer, who pursuant to urinalysis “barely” tested positive for the metabolite of cocaine, benzoylecgonine,⁵ sought to introduce evidence of a gas chromatography/mass spectrometry (GC/MS) analysis of a 1 1/2 inch length of his hair which failed to produce any evidence of benzoylecgonine. Technical differences aside, we face an issue which is the logical obverse of that in Nimmer. That is, the question of whether the absence of benzoylecgonine in a GC/MS analysis of hair reliably tends to prove that the subject didn’t use cocaine, is a logically and scientifically discrete question from whether the presence of benzoylecgonine or cocaine itself in a tandem stage quadrapole mass spectrometer (MS/MS) analysis of hair reliably tends to prove that the subject did use cocaine. Put another way, it is necessary to separately evaluate the reliability and the validity of hair analysis with reference to the dangers of false positives and false negatives. We are concerned with the former; the Nimmer court is concerned with the latter.

There was no dispute between the parties on the major premise of scientific hair analysis, viz., that mass spectrometry analysis of hair could detect the presence of cocaine. Nimmer II, 43 M.J. at 257. At trial, the government produced Dr. Brian Donnelly, whom the defense conceded to be an expert in the field of forensic toxicology and hair analysis, and who personally performed the analysis using a state of the art, relatively rare, tandem stage quadrapole mass spectrometry machine.⁶ The defense produced Dr. Bruce Goldberger, a similarly distinguished forensic toxicologist. Both had published papers in the professional literature, and both were recognized as among the leading exponents of hair analysis in the Society of Forensic Toxicologists (SOFT). In this sense, it is instructive to note that both agreed on the scientific validity and the reliability of the MS/MS analysis. Nor was there any disagreement about the foundational principle of hair analysis itself. Hair has been analyzed for 50 or 60 years to determine if somebody has been poisoned with, for example, a heavy metal or arsenic. The experts also agreed that cocaine, once ingested, appears in the hair, although uncertainty persists as to how it gets there. The reason that analysis for cocaine in human hair is relatively new is that improved analytical technology, epitomized by the MS/MS, enabled detection, and even quantification, of substances in extraordinarily minute quantities. Appellant’s expert did not take issue with the validity of the test. Dr. Donnelly testified that in periodic round robin proficiency tests, the FBI laboratory had done extremely well. As' applied to appellant’s case, both experts agreed that there was, in fact, cocaine in appellant’s hair. Both sides also agreed that a one-time use of cocaine in moderate quantity might not show up,' even on a machine as sophisticated, and sensitive, as a tandem quadrapole mass spectrometer.

In fact, although Dr. Goldberger differed at the margins with Dr. Donnelly respecting the internal chain of custody and procedural nuances, and was “uncomfortable” with the application of hair analysis in a criminal context, there was general harmony between the two on the relevant science and the methodology employed in appellant’s case. Such disagreement as arose turned chiefly on subtle issues of policy, not science. Nothing in the literature submitted by both parties, or in the testimony of the experts, contradicted that, with proper controls, chain of custody, scientific methodology, and instruments of sufficient sensitivity, cocaine found in hair is strongly indicative that cocaine was at some point ingested by the subject, and may properly be considered evidence of wrongful use of that drug. United States v. Thompson, 34 M.J. 287 (C.M.A.1992). Thus, for all intents and purposes appellant conceded the two principal threshold scientific hypotheses: (a) that cocaine appears in the hair of users; and (b) that scientific analysis using MS/MS (or even GC/MS) instruments can reliably and validly detect that cocaine.

Perhaps the most important source of disagreement between the two lay in the ques-' tion of passive exposure to the drug, that cocaine found in hair might be explained by other than knowing ingestion. Thus, where the two sides part company is over the final conclusion the government wished the court to draw from Dr. Donnelly’s testimony — that appellant wrongfully used cocaine.

To eliminate the passive exposure/contamination hypothesis, Dr. Donnelly first washed the hair samples in a methanol solution, then analyzed the wash for the presence of cocaine. It proved negative. This was a significant datum to both experts, who agreed that if appellant had merely been exposed to cocaine, but had not ingested it, or if his hair sample had been adulterated, that it should have been revealed in the wash. But Dr. Donnelly went farther. According to him, cocaine manifests itself differently in hair than in urine. Because the metabolite benzoylecgonine is a result of the body’s hydrolysis of cocaine, one looks primarily for it, as opposed to cocaine itself, in urine. By contrast, unmetabolized cocaine is the dominant substance found in hair analysis. In fact, Dr. Donnelly testified that a ratio of approximately 5-1 of cocaine to benzoylecgonine was strongly indicative of actual ingestion of the drug.⁷

In short, the system worked as it was intended. If we may put Daubert, Gipson, and Mil. R. Evid. 702 in the vernacular, the judge’s first responsibility is to filter from the triers of fact “junk science.” Mass spectrometer analysis of hair samples is accepted as scientifically reliable in the relevant community of forensic chemistry, has been subjected to peer review, is the subject of a growing body of professional publications, studies, and monographs, and, most important, can be both probative and helpful to the trier of fact. In short, MS/MS hair analysis is a far cry from palmistry, phrenology or chicken guts, and clears the pseudo-science hurdle easily.

There was more than ample evidence to conclude that the particular tests performed on appellant’s sample were scientifically reliable and valid, certainly enough for the military judge to permit testimony to that effect to be given to the triers of fact. According to Dr. Donnelly, the FBI laboratory had never reported a false positive. Having thus passed the threshold of admissibility, the expert testimony was tested, as it should be, in the crucible of trial advocacy, through the presentation of contrary evidence, energetic cross-examination, and careful instruction on the burden of proof. Daubert, 509 U.S. at 596, 113 S.Ct. at 2798. Appellant’s expert witness conceded, ultimately, that the tests were valid in this case, or at least that there was cocaine in appellant’s hair. He also agreed with Dr. Donnelly that, if cocaine were present as the result of either passive exposure or adulteration of the hair sample, he would have expected it to have shown up in the methanol wash. The chain of custody of the hair samples was sound, leaving little or no doubt that the hair samples tested were appellant’s.

That experts might dispute some particularities of the testing protocol or suggest ways that it could have been improved, or that different controls might be used, or that SOFT might harbor policy concerns about the feasibility of hair analysis for workplace testing, or deem it prudent to have independent corroboration of hair analysis,⁸ even considered in the aggregate, are insufficient bases upon which to exclude the results. A vigorous forensic dialogue between both experts was aptly engaged before the triers of fact, who ultimately decided that Dr. Goldberger’s reservations about and disagreements with Dr. Donnelly’s conclusions were insufficient to raise a reasonable doubt that appellant had used cocaine. Thomas, 43 M.J. at 633. Thus, we hold the military judge did not abuse his discretion in denying appellant’s motion in limine and permitting qualitative and quantitative analysis of appellant’s hair to go before the court members.

Legal, Factual Sufficiency of the Evidence

Last, appellant challenges the legal and factual sufficiency of the evidence supporting his conviction of both charges. Legal sufficiency is measured by whether, considering the evidence in a manner most favorable to the prosecution, a reasonable fact finder could have found all the essential elements of the offenses beyond a reasonable doubt. Jackson v. Virginia, 443 U.S. 307, 309, 99 S.Ct. 2781, 2784, 61 L.Ed.2d 560 (1979). Factual sufficiency requires that we be personally satisfied of the appellant’s guilt beyond a reasonable doubt, allowing for the fact that we are handicapped by the sterility of a written record. United States v. Turner, 25 M.J. 324 (C.M.A.1987). Even using the least generous view of the government’s evidence, we are persuaded of appellant’s guilt beyond a reasonable doubt.

After exhaustive effort, appellant’s trial defense counsel was unable to shake the chain of custody for the false urine sample. And we need not speculate long as to the possible motivation for appellant to submit a false sample. The government was not obliged to prove how appellant did it, nor did it have to physically observe the manner of its doing. Somebody switched the urine samples, and the chain of custody leads inescapably to the one person who had the motive to do so— appellant.

Turning to the question of cocaine use, the issue of quantification also deserves comment. At trial, appellant’s defense counsel challenged the absence of “cut-offs” in Dr. Donnelly’s protocol. We must understand Department of Defense “cut-offs” for what they are — scientifically derived, policy-driven figures aimed at minimizing the danger of conviction based upon passive inhalation or some otherwise innocent exposure to a given substance.⁹ They are not aimed, nor should they be employed, to frustrate the truthfinding process of trial, nor do they substitute for a reasoned application of the Military Rules of Evidence. See United States v. Johnston, 41 M.J. 13, 16 (C.M.A.1994) (overruling United States v. Arguello, 29 M.J. 198 (C.M.A.1989)). The court in this case benefitted from the learned testimony of experts on both sides of the question. Significantly, despite Dr. Goldberger’s criticism of Dr. Donnelly’s quantification protocol, he agreed that it was not so defective as to impeach the ultimate conclusion that cocaine appeared in appellant’s hair in such fashion and quantity as to impart great confidence that it originated with ingestion.¹⁰ Article 112a of the UCMJ does not distinguish between frequent use, nor even one-time use of cocaine. The use of cocaine is prohibited, period.

Like the court members, we are impressed with the rigor, precision, and science of the tandem stage quadrapole mass spectrometry analysis of appellant’s hair. There was expert testimony that the presence of cocaine in the hair shafts was metabolically explained by ingestion, and that it did not occur as a natural phenomenon. Appellant’s own witness conceded that there was cocaine in the hair sample tested, and Dr. Donnelly testified, in response to a question from the court-members, that he was unequivocally certain that the hair he tested was the sample given him by the AFOSI. In turn, the chain of custody established that sample to be appellant’s. Nor, as appellant seems to suggest, should we ignore the transparent purpose behind appellant’s substitution of his urine specimen on November 15, 1993. Accordingly, the findings and sentence are correct in law and in fact, the sentence is appropriate, and the same are hereby

AFFIRMED.

Senior Judges PEARSON and SCHREIER concur.

1 . Robichaud's decorousness and misplaced trust in appellant earned him a letter of reprimand. Commendably, he was honest about his failure throughout the investigation and the trial. It is heartening, therefore, that he recovered and was subsequently promoted to master sergeant.

2 . Unlike urinalysis, where principally metabolized cocaine (benzoylecgonine) is excreted in urine, unmetabolized cocaine is typically found in hair in five times the amount of its metabolite. According to the testimony of Dr. Donnelly, the government’s expert, the 5-1 ratio is typical of actual ingestion, and indeed, "almost precludes any possibility of external contamination.” External contamination would yield a much higher ratio. This datum proved significant in the trial itself, as appellant repeatedly suggested that the hair sample might have become contaminated through some kind of passive exposure.

3 . We note, in this respect, that at the time SA Toni was completing the affidavit, he did not know how long appellant’s hair was. Indeed, he was unsure as to its length even after it was cut. A precise measurement is simply not in the record. Appellant is an African-American, and everyone who handled the hair had some difficulty getting an accurate measurement of its length because of its tendency to curl tightly back on itself. We accept therefore, the estimate of Dr. Donnelly that the hair was 1/2 inch long.

4 . For ease in reference, we will refer to the Court of Appeals for the Armed Forces’ opinion as Nimmer II.

5 . 151 nanograms per milliliter (ng/ml), only 1 ng/ml above the Department of Defense cut-off level for reported positive using the radioimmunoassay (screening) test (RIA) and 51 ng/ml above the cut-off for the gas chromatography/mass spectrometry (GC/MS). Nimmer II, 43 M.J. at 253.

6 . MS/MS represents a quantum advance over GC/MS, the system used in most urinalysis cases, and is quite expensive, costing nearly $500,000 each, in comparison to $50,000 to $75,000 for a GC/MS. Very few laboratories in the world have such a machine. The defense expert acknowledged that "MS/MS is a very acceptable tool and it’s a fascinating tool for drug testing;” that the FBI was "lucky” to have such a machine.

7 . The ratio for appellant was approximately 6 to 1, a figure Dr. Donnelly concluded was consistent with actual ingestion and not passive exposure. The comparatively minute concentration of benzoylecgonine (2.7 ng/mg), detected only by the most-sophisticated machine available to science, may explain the experts’ reluctance to assert that hair analysis, particularly where only the metabolite is screened, is sufficiently sensitive to rule out a one-time use. Dr. Goldberger was less confident in the actual quantification of Dr. Donnelly’s results and did not go along with Dr. Donnelly’s conclusion that cocaine in the amounts found in appellant’s hair was more consistent with regular abuse as opposed to a single instance.

8 . According to Dr. Goldberger, the consensus of SOFT was that hair analysis drug testing was most prudentially used as corroborative of some other evidence of illicit drug use. Without accepting that policy admonition for anything more than what it is, we observe that appellant provided ample corroboration in his submission of the phony urinalysis specimen.

9 . A good analogy would be the legislative decision of various states as to what constitutes legal intoxication. That an individual might have .09% blood alcohol content, and thereby escape a DUI conviction in a state which establishes that .10% blood alcohol content is presumptive for legal intoxication, does not mean that he or she is not "drunk,” much less does it mean that he or she did not have anything to drink. Rather, the figure represents a policy judgment, accommodating the variant ability of humans to tolerate alcohol, that .10% represents too much alcohol to drive. The armed forces' zero-tolerance of use of controlled substances means that evidence of knowing ingestion of any amount of such a substance has evidentiary value. As Chief Judge Cox put it, "It is a recognized scientific principle that a sample is positive if it contains any trace of the tested substance, whether that be one kilogram (1000 grams), one nanogram (one billionth of a gram), or a picogram (one trillionth of a gram). It is negative if it contains zero or no trace whatsoever of the tested substance.” Johnston, 41 M.J. at 17 (Cox, J., concurring).

10 . It should be recalled that the quantification of cocaine in hair analysis uses a relationship established by a ratio of mass of cocaine (nanograms) to mass of hair (milligrams). Urinalysis results are expressed by a ratio of mass of cocaine metabolite (nanograms) to volume (milliliters) of urine. It would be unreasonable to expect that these figures would have much relationship to each other. It is also important to note that use of ratios, in conjunction with the use of deuterated isotopes of cocaine used in Dr. Donnelly’s protocol as a positive control, gives a reasonable mathematical assurance of quantitative accuracy.