dissenting.
I respectfully dissent from that portion of the majority opinion which holds admissible the results of the blood-alcohol test. I would reverse and remand for a new trial.
Orr's challenge to the adequacy of the foundation relative to the accuracy of the blood-aleohol test concerns the proper testing of the control sample. At the suppression hearing and at trial, Orr presented his expert witness, Dr. Frajola, who testified that accurate results could not be obtained unless an independent test was first conducted upon the control sample to determine that it was devoid of the substance for which the other samples were to be tested-in this instance, ethyl alcohol, and that this independent test must be conducted before the known quantity of alcohol is added to the control sample; otherwise, the tester cannot be certain that the control sample is pure and contains only the substances added for purposes of the testing procedure. According to Frajola, the test cannot be presumed to be accurate unless this preliminary step is performed. Frajola stated that because Naomi Carey did not perform a preliminary test on the control sample to determine that it contained no aleohol, the test results could not be accu*639rate. In arguing this point, Orr places great importance on Carey's "sniff test" and the inherently unscientific nature thereof. The significance of this issue, however, is not that Carey smelled the blood for alcohol, but whether she omitted the preliminary test on the control sample to determine that it was devoid of alcohol, which, in turn, may control the reliability of the test results. The issue which confronts us, therefore, is whether the evidence of Carey's failure to test the control sample for alcohol is fatal to the admissibility of the test results, or whether it more properly goes to the weight to be accorded the blood-aleohol analysis.
The general rule for the admission of blood-alcohol test results is that a proper foundation must first be established to show the reliability of the analysis. An examination of Indiana cases addressing the foundational requirements discloses a number of essentials upon which admissibility must be predicated. Among these, the party offering the blood-alcohol evidence must establish (1) the integrity of the blood sample by showing that its purity was preserved. Conrad v. State (1974) 262 Ind. 446, 317 N.E.2d 789, (2) the identity of the blood sample by showing a sufficient chain of custody, Fendley v. Ford (2d Dist. 1984) Ind.App., 458 N.E.2d 1167; Orr v. Econo-Car of Indianapolis, Inc. (1971) 150 Ind.App. 411, 276 N.E.2d 524, (8) the qualifications of the technician who performed the test, Pollard v. State (1st Dist.1982) Ind.App., 489 N.E.2d 177; Shultz v. State (2d Dist.1981) Ind.App., 417 N.E.2d 1127, and (4) the accuracy of the test by showing that it was conducted properly. See Shultz v. State, supra; Thomphkins v. State (1978) 270 Ind. 163, 383 N.E.2d 347, and State v. Alderson (3d Dist.1982) Ind.App., 485 N.E.2d 614. With the exception of the last requirement, there is ample case law addressing and elaborating upon each of the above propositions. However, for purposes of admissibility, it is not altogether certain what constitutes the proper performance of a blood-alcohol test upon a blood specimen. Cases which hold such tests admissible, eg., Shultz v. State, supra, 417 N.E.2d at 1187; Thompkins v. State, supra, 383 N.E.2d at 351, assume the existence of some evidence that the test in question was properly conducted. See also Jones v. State (1981) Ind., 425 N.E.2d 128. But see State v. Alderson, supra, 485 N.E.2d at 615-616 (in which the exclusion of test results was affirmed for failure by the State to establish that appropriate tests were conducted and thus a failure to lay a proper foundation for the admission.) Although there are no reported Indiana cases addressing the admissibility of blood test results for alcohol content in which the technician's test procedures have been challenged as improper, commentators and authorities are in agreement that as a foundational prerequisite for admission of medical or scientific test results, including blood-alcohol test results, it must be shown that the test was conducted properly so as to obtain accurate results. See, Richardson, J., Modern Scientific Evidence, 24 Ed. Ch. 13, § 18.10 et seq., p. 897; Conrad, 1 Modern Trial Evidence, 1966 Supp., § 713, p. 218-214; Donigan, Chemical Tests and the Law, (1950) p. 71-80; Erwin, 2 Defense of Drunk Driving Cases, 8d Ed., § 22.04, p. 22-41.18; McCormick, Evidence, 2d Ed., § 209, p. 511-518; 29 Am.Jur.2d, Evidence, 1983 Supp., §§ 829, 880. See also, Johnson, D., Blood Test Results-Their Admissibility to Show a Decedent's Intoxication, 38 Ind.L.J. 603, 614-617.
Many states have enacted statutes setting forth test procedure standards upon which admission of blood-alcohol tests are conditioned. Frequently, these statutes require the blood-alcohol analysis to be performed in compliance with guidelines established by a state health agency. See, eg., State v. Pickering (1983) Me., 462 A.2d 1151; State v. Miller (1988) 218 Neb. 274, 828 N.W.2d 769; State v. Green (1982) La., 418 So.2d 609; State v. Erickson (1976) N.D., 241 N.W.2d 854. Our legislature has enacted similar statutes containing specific foundational requirements for admission of breathalyzer test results. See I.C. 9-11-4-5 (Burns Code Ed., 1984 Supp.). However, there are no Indiana statutory re*640quirements for the admission of blood or urine analyses which determine the alcohol content. Because no Indiana case addresses the immediate issue, decisions in other states are of assistance.
In State v. Erickson (1976) N.D., 241 N.W.2d 854, the defendant, who had been charged with negligent homicide in connection with driving while under the influence of aleohol, challenged the admissibility of his blood-alcohol test results, alleging that the state had failed to show that the test was "fairly administered" as required by North Dakota statute. The laboratory technician who had performed the analysis of the defendant's blood, with the use of a gas chromatograph, had not conducted a spot check of the "known solution" to determine that it contained the chemicals it was supposed to contain. The defendant relied upon an earlier North Dakota case in which the court had held that the results of a breathalyzer test conducted by a police officer who had failed to perform a similar spot check of the ampoules containing the "known solution" were inadmissible because the test was not shown to have been properly administered. See State v. Salhus (1974) N.D., 220 N.W.2d 852. The Supreme Court of North Dakota first noted a considerable distinction between a police officer, whom it characterized as a layman in the field of toxicological testing, in conducting a relatively simple breathalyzer test, and a laboratory technician who performed blood tests in a controlled environment and under the supervision of the State Toxicologist, who was a state officer charged with specific statutory duties. The court also recognized a distinction in the use of a gas chromatograph as opposed to a breathalyzer machine, in that the gas chromatograph is a far more sophisticated and complex instrument, requiring for its operation the services of highly trained technicians or chemists. Given these distinctions, the court stated:
""We believe a laboratory technician performing a service under the supervision and direction of the State Toxicologist is performing an official act which is entitled to a disputable presumption of regularity pursuant to § 31-11-08(15), NDCC, until contradicted by other evidence. No contradictory evidence was introduced. Therefore, the presumption stands." Erickson, supra, 241 NW.2d at 865 (citations omitted). (Emphasis supplied)
The Erickson case is distinguishable from the instant case in three respects. First, there were statutory guidelines in that state upon which the admission of test results were predicated. However, the specific statutory requirement that the test be "properly administered" is no more precise than the general requirement, heretofore discussed, that the tests be conducted accurately. Second, the laboratory technician in Erickson labored under a statutory duty which, in the court's view, lent greater reliability to the procedures used and the results derived. Notwithstanding these distinctions, the approach adopted by the North Dakota Supreme Court states a viable solution to the immediate problem. That is, where a trained and experienced laboratory technician, chemist, or toxicologist is employed to conduct testing and analyses of blood specimens in a laboratory whose business is the securing, handling, and analysis of blood specimens, among other types of specimens, the testing procedures and results are entitled to a disputable presumption of regularity until and unless contradicted by other evidence. Third, and most importantly, unlike the Erickson case, in the case before us there was evidence which tended to place the accuracy: of the test in question. In State v. Tiernan (1978) Iowa, 206 N.W.2d 898, the defendant challenged the admissibility of the results because the state did not show that a particular tube in the gas chromatograph had heated properly, nor that certain moisture did not alter the test results. However, the chemist, in that case, was able to dispel the doubts as to the accuracy of the results. The chemist testified that if for some reason, the tube in question had not heated properly, the analysis would have revealed an impossibly high reading and the test would have been rejected. The *641chemist also testified that the presence of moisture, similarly, would have invalidated the entire test procedure rather than merely altering the results obtained. Given this reassurance, the Supreme Court of lowa held that the state had established a "sufficient foundation of the scientific principles and condition of the devices for admission of the evidence." 206 N.W.2d at 901.
In the case before us Carey's testimony, initially, was sufficient to create a presumption that she utilized proper procedures and obtained an accurate result. Her testimony evinces a thorough understanding of the principles and procedures of gas chromatography. She is highly trained and possesses years of experience in laboratory work. Under such cireum-stances, and in the absence of contradictory evidence, the accuracy of her test procedure and result may be presumed. However, contradictory evidence was introduced which was sufficient to cast considerable doubt on the propriety of the procedures followed on this particular occasion, and thus, on the accuracy of the test result. Dr. Frajola, also an expert, testified that Carey's failure to conduct a preliminary test on the control sample to determine that it was devoid of alcohol, in his opinion, subjected the results to inaccuracy. His testimony is sufficient to rebut the presumption of regularity and to cast upon the state the burden of going forward with other evidence relative to the accuracy of the procedures and results.
I am satisfied that in most cases, highly trained medical personnel who perform laboratory analyses, which often provide the basis for medical diagnosis and treatment, will perform their tasks carefully and in a manner which is acceptable in the relevant medical or scientific communities. In Woolley v. Hafner's Wagon Wheel, Inc. (1961) 22 Ill.2d 413, 176 N.E.2d 757, the Illinois Supreme Court states:
"If the routine and procedures of a laboratory are shown by the evidence as having been commonly accepted by the medical profession, and the business of the laboratory is the security, handling, and analysis of blood specimens, amongst other types of specimens, these routines and procedures ought to be acceptable to the courts."
However, even highly trained specialists are susceptible to mistake, oversight, and even carelessness. Therefore, when the opponent brings forth evidence of an actual error or omission in the testing procedure, the proponent of the evidence should be able to attest to the propriety of the procedures used and to vouch for the reliability of the test results. This is because the proponent is in a far better position, and has at his command the facts necessary, to prove that his procedure was proper and his conclusion, reliable. In the instant case, to require the challenger to show that the control sample did contain ethyl alcohol, such as would render the test result inaccurate, would be to impose an undue burden and a virtually insurmountable obstacle.
It may be that the failure to test the eontrol sample for alcohol would not necessarily control the reliability of the test results. It appears logical to one untrained in science that the effect of a control sample which did, per chance, contain alcohol would be to show a smaller variance in the alcohol content of Orr's blood vis-a-vis the control sample, and therefore, a lower blood-alcohol level than what truly existed in Orr's blood. If this hypothesis were correct, it would operate in Orr's favor. Without evidence to support it, however, the conclusion is sheer supposition.
I have searched the record for evidence which might assuage my doubt as to the reliability of the test procedure, as conducted on this occasion, and have found none. Carey did not testify that she conducted the test in compliance with Methodist Hospital procedures. She did not testify that the test, as she conducted it, complied with standard or accepted procedures in the relevant medical or scientific communities. She did not testify that, despite Dr. Frajo-la's testimony, it. was not necessary to conduct a preliminary test on the control sample in order to obtain accurate test results. *642She did not testify that the possibility of her control sample containing ethyl alcohol before she added the known solution was remote at best. In fact, there is no testimony regarding the purity of the sample Carey obtained from the hematology lab. Finally, she did not testify that her omission, if it was an omission, would not control the accuracy of the results. Unlike the chemist in Tiernan, supra, 206 N.W.2d 898, who was able to answer the defendant's challenges to the proper functioning of the gas chromatograph, Carey's testimony fails to address the challenge to her procedures posed by Dr. Frajola, and, I am unable, on the record before me, to ascertain whether she might have successfully defended her procedures.
In short, the State simply failed to offer any facts which would allow the trial court or us to conclude that the blood-alcohol analysis was conducted, in this instance, in a manner calculated to produce accurate and reliable results. An adequate foundation was not established. Therefore, in my view, the blood-alcohol results were erroneously admitted.
My conclusion finds no contradiction in the case of Reid v. State (1978) 267 Ind. 555, 372 N.E.2d 1149, wherein the supreme court stated:
"The fact that a scientific test is subject to error if not properly conducted is not a reason for rejecting evidence adduced by such a test. The persuasiveness of the evidence is in large measure dependent on the expertise of the witness who conducted it and in the final analysis is to be determined by the jury after the opportunity for eross-examination."
Although the above language might lead one to believe that test procedures were at issue in Reid, an examination of the case reveals otherwise. In that case, a trace metal detection technique (TMDT) was administered to the defendant's hand to detect minute traces of metal remaining upon flesh which had come into contact with metal objects. The test procedure itself was simple, and involved only the spraying of the defendant's hand with a commercial ly prepared chemical solution which is then placed under a special light. The light reveals a purplish glow where there are metal tracings. The challenge to admissibility, in Reid, was not to the propriety of the test procedure used, as in the instant case, nor the accuracy of the results. Rather, the defendant challenged the scientific reliability of the entire technique, likening the TMDT to a lie detector test. The defendant urged that the entire TMDT was inherently unreliable. Thus, the issue in Reid was the scientific validity of the test, per se. This is not an issue in the instant case concerning the gas chromatograph. All parties concede the scientific validity and reliability of gas chromatography, if conducted properly. Therefore, under the facts of this case, Reid is inapposite.
Likewise, the instant case is distinguished from Martin v. Roberts (1984) Ind., 464 N.E.2d 896. In that case, the supreme court overturned Martin v. Roberts (2d Dist.1983) Ind.App., 452 N.E.2d 182, and held that the expert opinion testimony of a state trooper as to the speed of a vehicle was admissible despite the trooper's failure to use any formula, calculation, or principle in adducing the speed at which the vehicle had been traveling, and his failure to take into account the structure, weight, or load of the vehicle. Our supreme court determined that any deficiency in accounting for these elements went to the weight of the trooper's estimate of speed rather than to its admissibility.
The focus of Martin v. Roberts, as is relevant here, is upon the admissibility of an expert's testimony as to his opinion, based upon his credentials as an expert and his measurements and visual observations at the accident scene, although he did not personally witness the accident. The instant case does not concern the admissibility of expart opinion, although the testimony of an expert is necessary to the admissibility of the blood test results. Rather, the focus of the instant case is upon the admissibility of a scientific fact-the objective and precise measurement of the alcohol content of a particular blood sample based *643upon technical principles of scientific measurement and testing, the accuracy of which is controlled by the manner in which the measurement is taken. Naomi Carey's testimony as to the blood-alcohol content of Orr's blood sample was not a statement of her opinion as to the level of alcohol content. To the contrary, it was testimony of scientific fact. For that reason, any defect in the procedures used to ascertain that fact, which affect the accuracy of the fact must necessarily render the fact, or the test result, inadmissible and prejudicial to the defendant. I disagree with the majority conclusion that we are confronted merely with conflicting expert opinion. In the case before us we do not have conflicting expert opinions regarding the propriety of the test procedure. If we did, I1 would agree that the test results might be admitted and the trier of fact be advised that they may weigh the conflicting opinions and according to that assessment lend such weight, if any, as might be appropriate.
Here, however, we have only an opinion by one expert that an admittedly omitted test procedure was necessary to assure the accuracy of the test result. We do not have a conflicting opinion that the omitted procedure was not necessary either because result accuracy could not be adversely affected thereby, or that accuracy was assured by other procedures used instead of the omitted procedure.
The blood-alcohol analysis was not the only evidence of Orr's intoxication. Officer Goff and Officer Madison testified that they smelled alcohol on Orr's breath at the scene of the accident. Officer Carrico testified that he smelled alcohol on Orr's breath at the hospital; that Orr exhibited loud and boisterous behavior; and that Orr's speech was slurred. Officers Goff and Carrico both testified that in their many years of law enforcement experience they each had numerous occasions to observe persons who were intoxicated and that in their opinion, based upon their observation of Orr's behavior, he was intoxicated. The passengers in the car passed by Orr, just before he struck the decedents' car, testified that they saw Orr's head as he passed them but they did not see his head as his car veered to the left, implying that he had fallen over on the seat. Indeed, Orr was found entrapped on the floor of his car, his body stretched cross-wise and his head on the passenger side. Finally, Orr admitted that he had been drinking at two westside Indianapolis taverns just prior to the accident.
These facts alone may have been sufficient to satisfy the jury that Orr was intoxicated, and thus support the verdict. However, I am unable to conclude that the jury's decision was not significantly influenced by the improperly admitted blood-al-eohol test results or that the remaining evidence is so overwhelming as to support the verdict as a matter of law. See Mulry v. State (2d Dist.1980) Ind.App., 399 N.E.2d 413. I would reverse and remand for a new trial.