specially concurring:
I agree that in determining whether a presumption that a defendant was under the influence of alcohol arises under section 18-3-106(2)(c), 8B C.R.S. (1986), the statutory standard' “0.10 percent or more by weight of alcohol in the defendant’s blood” requires a calculation based on grams of alcohol per 100 milliliters of blood. I write separately to emphasize what I believe to be the narrow applicability of the reasoning that produces this seemingly anomalous result.
In construing a statute we must ascertain and give effect to the intent of the legislature. Kern v. Gebhardt, 746 P.2d 1340, 1344 (Colo.1987). If the statutory language is clear, we look no further to determine that intent. Kane v. Town of Estes Park, 786 P.2d 412, 415 (Colo.1990). Only when the language is ambiguous may we consider extraneous sources, such as legislative history, to arrive at the proper meaning. Griffin v. S.W. DeVanney & Co., Inc., 775 P.2d 555, 559 (Colo.1989).
The term “percent by weight” has a clear, commonly understood meaning. In order for a result expressed as a “percent” to have meaning, the units being compared must be the same. See 2 Erwin, Defense of Drunk Driving Cases § 15.02 (3d ed. 1991). The majority acknowledges this. Maj. op. at 507 n. 5. “Percent by weight,” therefore, clearly denotes a calculation employing the same units of weight for both alcohol and blood. Under normal circumstances it would be indisputable that blood test results expressed as grams (a unit of weight) per 100 milliliters (a unit of volume) do not state a percent by weight. Such a mathematical calculation cannot produce a result that can be properly characterized in terms of “percent.”
The “percent by weight” statutory test, however, cannot be divorced from the context for which it was created. Blood-alcohol tests are common, and have been for years, as evidentiary aids in determining whether a person drove a motor vehicle while impaired by alcohol to a statutorily impermissible degree. See Commonwealth v. Brooks, 366 Mass. 423, 319 N.E.2d 901, 904 (1974). The usual scientific practice in conducting such tests is to determine grams of alcohol per 100 milliliters of blood. Id. 319 N.E.2d at 905; 2 Erwin, Defense of Drunk Driving Cases § 15.02. The results of blood-alcohol tests so conducted are typically expressed in “percentage” terms even though “[sjtrictly speaking, this [mathematical expression] is not a true percentage by weight of alcohol in the blood. ...” Id. Indeed, such testing practice and method of expressing results are commonly adopted by scientists even when the relevant statutes specifically refer to the standard as percent by weight, see Brooks, 319 N.E.2d at 905-07 and 905-06 n. 6. This is an area then— hopefully unusual — in which science attaches a meaning to a term different from the clear, common, and technically correct meaning of the words.1 For the reasons *510set forth by the majority, I agree that it is this scientific méaning that the legislature intended to adopt in enacting the “percent by weight” standard.
In some circumstances, due process concerns would arise from adopting a statutory construction so at odds with the plain meaning that would be ascribed to the words by persons not acquainted with the scientific usage. I think that is not the case here. Principles used in evaluating void for vagueness challenges to statutes have application by analogy. We have held that a vagueness challenge implicates two constitutional concerns: statutory language must be clear enough to afford a practical guide for law-abiding behavior, and must also provide sufficient guidance to assure that the statute can be enforced in a consistent and even-handed manner. People v. Schoondermark, 699 P.2d 411, 415-16 (Colo.1985); People v. Gross, 670 P.2d 799, 800 (Colo.1983). It cannot reasonably be believed that the test at issue is one to which persons tailor their conduct. The evidence showed that the grams per 100 milliliters standard produces results that are only slightly higher than a true weight per weight standard. It is simply unrealistic to believe that persons limit their alcohol consumption in a way so finely calibrated as to avoid the presumption under the weight per weight test but sufficient to reach the level that gives birth to the presumption under the weight per volume test. Moreover, the record suggests that testing laboratories have consistently used the grams per 100 milliliters method of expressing blood-alcohol test results for use in Colorado courts. The law has therefore been applied in an even-handed manner. Under the circumstances, the utilization of the grams of alcohol per 100 milliliters of blood standard did not violate defendant Rickstrew’s right to due process of law.2
I would explicitly recognize that the adoption of the grams per 100 milliliters standard is not consistent with the plain, common meaning of the statutory language. The construction is proper in these limited and unusual circumstances, however, because it is consistent with legislative intent, comports with scientific usage, and does not infringe upon the due process rights of persons subjected to the test.
I concur in the judgment of the court.
ERICKSON, J., joins in this special concurrence.. To arrive at the meaning ascribed to the term "percent by weight" in scientific practice in conducting blood-alcohol tests, it is necessary not only to cpnclude that "percent by weight” is to be based on a comparison of weight to volume but that a particular unit of weight (the gram) and a particular unit of volume (the milliliter) are to be employed. Not only that, but the volume to be used is 100 milliliters. Nothing inherent in a "weight per volume” method of *510expressing blood-alcohol test results requires that weight be measured in grams or volume in 100 milliliter quantities. Only common scientific testing practice supports the use of these units and quantities of weight and volume.
. As applied to Rickstrew, both the weight per weight test and the weight per volume test produced a result in excess of that necessary to trigger the statutory presumption. See maj. op. at 508.