As You Sow v. Conbraco Industries

Opinion MUNTER, J. * This appeal concerns the Safe Drinking Water and Toxic Enforcement Act of 1986 (Health & Saf. Code, §§ 25249.5-25249.13), adopted by the vote of the people at the November 1986 election as Proposition 65 (hereafter referred to as the Act or Proposition 65). The Act regulates the discharge of specific toxins into California’s drinking water. A plaintiff alleging an unlawful discharge bears the burden of proving that a discharge has in fact occurred. California Code of Regulations, title 22, former section 12901 (hereafter Regulation 12901), * 1 which was adopted pursuant to the Act, previously established a hierarchy of testing methods, called “methods of analysis,” to be used in determining whether there has been a discharge. Any testing method, other than a “method of analysis,” could not be used to prove a discharge under the Act. In this case, plaintiff alleged that plumbing parts manufactured and/or distributed by defendants discharged lead into sources of California drinking water. The sole issue tried below was whether, in its reliance upon a particular designated testing method, plaintiff met its burden of proving, under Regulation 12901 and therefore under Proposition 65, that defendants’ products caused discharges. Following trial to the court, judgment was entered in favor of defendants, the court finding and concluding that plaintiff’s designated testing method did not meet the requirements of any of the “methods of analysis” permitted by Regulation 12901. While this appeal was pending, Regulation 12901 was repealed. *437 Plaintiff contends that the trial court misinterpreted Regulation 12901 and that reversal of the judgment follows. We agree that the trial court misinterpreted Regulation 12901 but disagree with the contention that the misinterpretation warrants reversal. Plaintiff further contends that, as a consequence of the repeal of Regulation 12901, plaintiff should have a new trial without being subjected to the requirements of that Regulation. We conclude that the repeal does not entitle plaintiff to another trial. STATUTORY AND REGULATORY BACKGROUND A. Proposition 65 Proposition 65 adopts a dual-pronged approach based on protection and information. The Act accomplishes these twin objectives by prohibiting any person in the course of doing business from knowingly discharging or releasing a listed toxic chemical into a source of drinking water (discharge provision) or from knowingly and intentionally exposing any individual to such chemicals without first providing a warning (warning requirement). (Health & Saf. Code, §§ 25249.5, 25249.6.) Section 25249.8 requires the state to publish and update annually a list of known cancer-causing chemicals (carcinogens) and reproductive toxins (teratogens). Lead and lead compounds are listed carcinogens and teratogens. (Regs., § 12000, subds. (b), (c).) “Proposition 65 establishes a series of shifting burdens. In the first instance, a plaintiff must show that a discharge has occurred. Once this burden has been met, the defendant may show, inter alia, that the amount of the discharge is not significant.” (Mateel Environmental Justice Foundation v. Edmund A. Gray Co. (2003) 115 Cal.App.4th 8, 18 [9 Cal.Rptr.3d 486] (Mateel), italics added; see also Health & Saf. Code, § 25249.9.) B. Regulation 12901 Regulation 12901 was adopted in 1988 by the California Health and Welfare Agency (HWA) 2 to implement Proposition 65. (HWA, Final Statement of Reasons (1988) Enactment, § 12901, p. 6.) Subdivision (g) of *438 Regulation 12901 provides, “For purposes of [Health and Safety Code] Sections 25249.5 and 25249.6 of the Act, no discharge, release or exposure occurs unless a listed chemical is detectable as provided in this section.” (Regs., § 12901, subd. (g), italics added.) In turn, subdivision (a) of Regulation 12901 defines “any detectable amount” as follows: “For purposes of Section 25249.11, subdivision (c) of the Act, the term ‘any detectable amount’ means a level detected using a method of analysis referred to in this section. For purposes of this section, ‘method of analysis’ refers to the method of detection or detection and calculation for a listed chemical in a specific medium, including, but not limited to, water, air, food, or soil, and shall include methods and procedures concerning the number of samples and the frequency and site of sampling that are specific for the listed chemical in question.” (Regs., § 12901, subd. (a), italics added.) “Subdivisions (b) through (e) of Regulation 12901 describe a hierarchy of methods of analysis that are to be used in detection. Under subdivision (b), Tier 1 tests are those that have been adopted or employed by one of a specified list of state and local agencies, including the California Department of Health Services. Under subdivision (c), Tier 2 tests are those that have been adopted or employed by a federal agency. Under subdivision (d), Tier 3 tests are those that are generally accepted in the scientific community, as evidenced by publication in professional journals and compilations. Finally, under subdivision (e), Tier 4 tests are those that are scientifically valid.” (Mateel, supra, 115 Cal.App.4th at p. 19.) C. Repeal of Regulation 12901 On June 4, 2004, while the instant appeal was pending, OEHHA published a notice of proposed rulemaking, regarding proposed amendments to Regulation 12901. 3 As stated in the June 4, 2004 notice, OEHHA was considering amending the regulation as a result of perceived difficulties in interpreting and applying it. (OEHHA, Notice of Proposed Rulemaking (2004) § 12901, p. 3.) The stated purpose of the amendments was to provide a greater level of certainty to parties subject to Proposition 65. (OEHHA, Notice of Proposed Rulemaking, supra, at p. 3.) The June 4, 2004 notice began a 45-day public comment period for OEHHA to accept written comments. (Id. at p. 1.) Subsequent to publishing the notice, OEHHA published an initial statement of reasons detailing the substance of the proposed amendments, which *439 included a revised definition of the term “medium.” 4 (OEHHA, Initial Statement of Reasons (2004) Amendments to § 12901, p. 4.) Under the proposed revision, “medium” was defined as “the substance, or mixture of substances, that creates the exposure (i.e. water, air or soil mixed with or containing the chemical in question) that closely duplicates the actual exposure conditions.” (OEHHA, Proposed Amendments (2004) p. 1, italics added.) “The new definition clarifies that the term ‘medium’ refers to the substance, or mixture of substances, that creates the exposure (i.e. water, air, soil, food mixed with or containing the chemical in question) or that closely duplicates the actual exposure conditions.” The proposed amendments were “intended to clarify the fact that the medium to be tested must be analogous to the item causing the real world exposure or discharge, but that the test need not be performed on a sample obtained during an actual exposure situation or from the actual discharge.” (OEHHA, Initial Statement of Reasons, supra, at p. 4.) Following the public comment period, OEHHA ultimately reached the conclusion that the proposed amendments did not resolve the legal and technical concerns. (OEHHA, Final Statement of Reasons (2005) Repeal § 12901, p. 1.) Accordingly, OEHHA decided to proceed with the repeal of Regulation 12901. (Id. at p. 3.) In reaching this conclusion, OEHHA determined that Regulation 12901 was “not necessary” and that test methods should instead be evaluated using the “existing body of statutory and case law concerning the conduct of scientific tests and the admissibility of scientific evidence.” (Id. at p. 2, citing Evid. Code, §§ 210 & 500; see People v. Kelly (1976) 17 Cal.3d 24 [130 Cal.Rptr. 144, 549 P.2d 1240]; People v. Leahy (1994) 8 Cal.4th 587 [34 Cal.Rptr.2d 663, 882 P.2d 321]; People v. Bonin (1989) 47 Cal.3d 808 [254 Cal.Rptr. 298, 765 P.2d 460].) In the final statement of reasons for repealing Regulation 12901, OEHHA provided a summary and response to comments to the proposed amendments. OEHHA reported that “[s]everal persons representing various regulated businesses did not support repeal of the regulation and further alleged that the regulatory amendments were proposed by OEHHA in an effort to influence the outcome of pending litigation.” (OEHHA, Final Statement of Reasons, Repeal, supra, at p. 3.) In response to this comment, OEHHA stated: “The regulatory action proposed by OEHHA was partially initiated in response to court decisions in which it appeared that the regulation was difficult to understand and apply, particularly in the context of consumer product exposures. The proposed regulation was not developed with the intent to influence the outcome of any pending court cases.” (Id. at p. 4, italics added.) *440 Additionally, in response to comments about the effect of the repeal on small businesses, OEHHA stated, “given that the regulation was used, primarily as an affirmative defense to litigation that was filed against businesses, it is true that small businesses will no longer be able to derive a benefit from the regulation, nor incur a detriment from it.” (OEHHA, Final Statement of Reasons, supra, at pp. 17-23, italics added.) That statement was repeatedly made by the agency. (Ibid.) PROCEDURAL HISTORY As You Sow (AYS) sued 36 companies that import, manufacture, distribute, market, and/or sell in-line plumbing devices 5 in California. It alleged in its first two causes of action that defendants had violated the discharge and warning provisions of Proposition 65 by selling plumbing parts that discharge lead into sources of California drinking water and by failing to warn consumers of this exposure. In the remaining four causes of action, AYS alleged that these Proposition 65 violations were unlawful business practices under the unfair competition law. (Bus. & Prof. Code, § 17200 et seq.) The trial court trifurcated the case. The first stage was limited to whether AYS could prove that a discharge of lead had occurred. The first stage was further bifurcated into phases LA and IB. The scope of phase IA was confined to determining whether AYS’s designated testing method complied with Regulation 12901. If phase IA resulted in a favorable ruling to AYS, phase IB would address the remaining elements of AYS’s prima facie case. At trial, AYS sought to establish that a test method entitled “National Sanitation Foundation Intemational/American National Standards Institute Standard 61, Section 8” (NSF 61 § 8) qualified as a “method of analysis” under Regulation 12901. AYS relied upon deposition testimony taken from Dr. David Spath, who was employed by the California Department of Health Services (DHS) as the head of the Division of Drinking Water and Environmental Management. AYS also relied on the testimony of its designated expert Dr. Marc Edwards, a water chemistry and corrosion expert. Defendants relied upon the testimony of their four designated experts: Dr. Steven Reiber, a water chemistry and corrosion expert; Dr. Donald Langmuir, a water chemistry expert; Jerome Gilbert, the former general manager and chief engineer of two California water utility districts; and Rand Ackroyd, a member of NSF’s joint committee, which oversees the development and revision of NSF standards 60 and 61. *441 Following a seven-day bench trial on phase LA, and following posttrial briefing, the trial court issued its “Phase IA Trial Decision,” holding that AYS had failed to carry its burden of proving that its designated testing method was an acceptable “method of analysis.” The trial court rested its decision on alternative bases, which we discuss after reviewing the evidence presented at trial. This timely appeal followed. EVIDENCE AT TRIAL NSF 61 was developed by the National Sanitation Foundation (NSF), 6 a private, nonprofit organization that promotes its own voluntary product performance standards and certifies that products meet those standards. NSF 61 contains five different test protocols pertaining to a variety of products that contact drinking water. Section 8 is the specific section for mechanical devices, which include in-line plumbing devices such as those at issue in this appeal. NSF 61 § 8 is a laboratory protocol that tests for hundreds of different chemicals, including lead, on a pass/fail basis. NSF 61 § 8 exposes in-line plumbing devices to two extraction solutions at pH levels of 5 and 10 for a dwell period of not less than 12 hours. The extraction solutions are then analyzed for the presence of the many chemicals covered under the protocol. A. Purpose of NSF 61 § 8 At trial, Mr. Jerome Gilbert, the former general manager and chief engineer of two San Francisco Bay Area utility districts, 7 testified that the purpose of NSF 61 § 8 is to provide a basis for a voluntary product certification to be used by public water utilities when purchasing products. The certification standard allows water utilities to know the material components of the parts before they purchase them. Mr. Rand Ackroyd, a member of the NSF 60/61 joint committee, 8 testified that the intent of NSF 61 § 8 has never been to predict how products will perform in actual use. According to Dr. Steven Reiber, a water chemistry and corrosion expert, the NSF 61 § 8 test was not designed for the purpose of determining the amount of lead, if any, that may leach from a product under *442 typical conditions of use. Dr. Reiber further opined that NSF 61 § 8 is not relevant for testing products in “a real world drinking water situation.” AYS’s sole expert witness at trial, Dr. Marc Edwards, a water chemistry and corrosion expert, acknowledged that NSF 61 § 8 is not intended to identify how a specific product will react when used in a real world situation. Dr. Edwards further testified that there is no proven, direct relationship between the test results derived from NSF 61 § 8 and a given product’s lead-leaching propensity once actually installed and used. Dr. Edwards recognized that the goal of NSF 61 § 8 is not to test the actual drinking water. Rather, the goal of NSF 61 § 8 is to “identify problems with specific products in the same water .... [f] ... the synthesized water.” Expert testimony from both sides was to the effect that no study has ever been performed to establish whether there is a relationship between the lead-leaching results obtained from NSF 61 § 8 tests and the lead leaching that would occur in actual drinking water. B. NSF 61 § 8 Test Medium All of the experts agreed that the NSF 61 § 8 testing medium was not California drinking water. However, AYS’s expert, Dr. Edwards, testified that it would be “unreasonable” to use actual drinking water as the medium because it can vary greatly depending on location, time of year, even by time of day. As such, he opined that drinking water could not be used to compare the lead-leaching propensities of different products because it “simply is not reproducible.” A defense expert testified that the vast majority of Californians receive their drinking water from, public utilities. Additional evidence was presented that water utilities treat drinking water to ensure that it is noncorrosive, creating a coating, commonly referred to as a “passivation layer,” on the interior surface of the parts that come into contact with water in the distribution system. In his report, Dr. Reiber opined that the NSF 61 § 8 testing medium prevents the formation of a protective passivation layer on in-line plumbing products and destroys any such layer that may have been formed. Dr. Reiber further testified that NSF 61 § 8 employs an “extraction solution . . . which by its very name is intended to mobilize and release metals from the test specimens.” Mr. Gilbert opined that these extraction solutions are not intended to be representative of drinking water. Rather, according to Mr. Ackroyd, the intent of NSF 61 § 8 was to create the worst possible conditions for extracting metals. Additionally, Dr. Edwards conceded *443 that the NSF 61 § 8 testing solution was not found anywhere in nature. However, Dr. Edwards opined that, based on his reading of the NSF 61 § 8 protocol, he believed the extraction solutions were intended to be representative of drinking water and to “crudely” represent lead leaching in practice. He further testified that the pH 5 solution was no more aggressive than California rainwater. Dr. Donald Langmuir, a water chemistry expert, disagreed with this opinion, stating that rainwater would “never” have a pH of 5. Defendants’ experts testified that there are significant differences between the NSF 61 § 8 testing medium and California drinking water. Dr. Langmuir described the differences as “rather vast.” Specifically, the NSF 61 § 8 testing medium is maintained at pH levels (5 and 10) that are well outside the 6.5 to 8.5 range preferred by California water utilities. Dr. Edwards conceded that he was not aware of any utility in California that provides drinking water with a pH of 5 or below. There was conflicting testimony as to whether the NSF 61 § 8 extraction solutions would have the same or similar effect upon defendants’ plumbing products as would California drinking water. Mr. Ackroyd testified that the pH levels of the extraction solution would aggressively cause lead leaching that otherwise would not occur when the products were exposed to California drinking water. Dr. Reiber testified that the extraction solutions would induce a corrosion rate “roughly 300 fold higher” than what occurs in “a typical drinking water environment on a relatively well-passivated brass surface.” Dr. Edwards’s testimony was to a different effect; he opined that the presence of the chemical orthophosphate in the pH 5 extraction solution would inhibit lead leaching and would cause less lead leaching than would otherwise occur if the products were exposed to California drinking water. Both Mr. Ackroyd and Dr. Reiber disagreed with that opinion. Specifically, Mr. Ackroyd testified that the intent of adding orthophosphate to the NSF 61 § 8 solutions is not to inhibit lead leaching, but rather is to bring the pH level down to 5 and maintain it at that low level. Dr. Reiber testified that orthophosphate can act as an inhibitor in higher pH solutions, but that it would have no effect on the corrosion rate on copper alloys in a low pH 5 solution. C. NSF 61 § 8 Testing Procedures Addressing the specific testing procedures used by NSF 61 § 8, Mr. Ackroyd testified that, as a member of the NSF 60/61 joint standards committee, he participated in the development of the NSF 61 standard. Specifically, Mr. Ackroyd testified that NSF 61 § 8 does not specify the number of samples to be tested, the frequency of testing, the sampling sites, or the specific tests for lead. NSF 61 § 8 permits the test results from one product to be representative of a family of products. In his expert report, he characterized NSF 61 § 8 *444 as a “ ‘snapshot’ test in that it requires no follow-up or random testing for continuing compliance. One test on one device or its components is the requirement established by the protocol.” Mr. Ackroyd testified that NSF 61 § 8 testing procedures are often dictated by the specific laboratories running the test, which can cause disparate test results for the same product. For example, NSF 61 § 8 identifies the length of the “conditioning period” as “14-day-or-less exposure period.” Another variable is whether the “extraction period” will last 12 hours, 16 hours, or some amount of time in between. Additionally, there is no consistency regarding whether the entire device, a component of the device, or all of the components individually are to be tested. In his report, Mr. Ackroyd stated that many of the testing procedures are not based on science, but have evolved out of the negotiations between the manufacturers of the product and the individual laboratories conducting the tests. Furthermore, Mr. Ackroyd’s report discusses the numerous differences between the test conditions and the actual environment where defendants’ products are used. The following differences were highlighted: (1) the products are tested using aggressive “extraction waters” instead of actual drinking water; (2) the products are tested after the first 17 days of exposure to the “extractant medium,” while in the real world in-line products have a product life of more than 30 years; (3) the test uses a static dump-and-fill procedure instead of flowing water; (4) the complete devices are not required to be tested, and as such the tests do not take into account “the interior geometry of the product, flow, or other issues that apply in the real world [;]” (5) the products are tested with room temperature water, while in a real world plumbing system water is either colder than room temperature or warmer than room temperature; and (6) the test assumes that every drop running through the products is for human consumption, while in actual use less than 5 percent of the hundreds of gallons of water flowing through the products on a daily basis is available for consumption. D. Scientific Validity On the one hand, AYS’s expert testified that NSF 61 § 8 is a “scientifically valid test for purposes of measuring lead leaching propensity of specific products.” On the other hand, the defense presented evidence that NSF 61 § 8 is not scientifically valid as a test for detecting lead from a copper alloy product in California drinking water. Additionally, there was conflicting evidence on the issue of whether NSF 61 § 8 generates reproducible test results. Dr. Reiber testified that the ability of a testing method to generate reproducible test results is an “essential component” of scientific validity. Mr. Ackroyd further testified that NSF 61 *445 § 8 test results can vary depending on the laboratory conducting the test. Dr. Edwards was of a contrary opinion, testifying that any variances in test results were negligible. The experts also disagreed about whether the NSF 61 § 8 testing waters would have the same or similar effects upon defendants’ plumbing products as would California drinking water. However, as previously noted, the experts for both sides agreed that no study has ever been performed to establish whether there is a relationship between the leach-leading results obtained from NSF 61 § 8 tests and the lead leaching that would occur in the actual drinking water. TRIAL COURT’S DECISION The trial court found as follows. NSF, a private nonprofit organization, develops and promotes testing standards under which manufacturers can voluntarily submit their products for certification. The test medium used in NSF 61 § 8 is not actual California drinking water (i.e., it is not the water that flows through the defendants’ products in real world use), but rather is a synthesized water compound to which various chemicals are added. The valves are tested by submersing them in this medium for a number of days. Thereafter, test samples are taken from the valves to determine whether lead has leached into the medium. Importantly, the trial court held that, on the “key factual disputes,” AYS’s expert was no more persuasive than defendants’ experts, and that therefore AYS failed to carry its burden of proof on those issues. The “key factual disputes” on which AYS failed to persuade included the issue of whether AYS’s test medium was “sufficiently identical to real California drinking water so as to be appropriate for testing under Prop. 65.” On this issue the trial court found as follows: “In particular, the plaintiff asserts that the synthesized water is identical in effect to the range of actual drinking water in California, while the defense claims that the solution itself precipitates lead leaching that would not occur with actual drinking water. The defense also asserts that the method of soaking the tested products for protracted periods of time is significantly different than actual use conditions where replenished water flows constantly through the products. Again, the defense claims that the soaking causes leaching rather than just detecting it. [f] Both sides presented qualified and credible expert evidence to support their positions. Significantly, however, neither the plaintiff nor the defense bolstered their expert evidence by running the NSF 61 § 8 test with actual California drinking water to compare results with those obtained with the protocol test water.” *446 Interpreting Regulation 12901 to require that the “medium” must be the specific real world medium, the trial court held that the NSF 61 § 8 test medium did not qualify as a “method of analysis.” However, after noting that “plaintiff asserts that the synthesized water is identical in effect to the range of actual drinking water in California,” the court rejected that contention based on the evidence. Thus, it found that: “Even if this court were to interpret Prop 65 and [Regulation] 12901 to allow for a method of analysis utilizing a medium not identical to the real world medium so long as the test medium behaved exactly like the real world medium, plaintiff has failed to meet its burden of proof that such is the case with NSF 61 § 8. According to plaintiff’s expert, the NSF 61 § 8 test medium behaves like California drinking water. According to the defense expert, it does not, but instead causes the leaching of lead from the defendants’ products. Each side’s expert was well-qualified and provided credibly supported opinions regarding how the test medium compared to actual California drinking water. Neither was more believable than the other. As such, plaintiff failed to meet its burden of proof on this point. [][] Of particular note is the fact that neither expert ran the NSF 61 § 8 test on a defendant’s product first with the test water and then retesting the same product substituting actual drinking water for the test water. A comparison of the two results likely would have shed light on whether or not the test water behaves like actual drinking water. No such comparison was presented to this court and, given that the plaintiff had the burden of proof on this point, the court must conclude that NSF 61 § 8 has not been shown to be an acceptable method of analysis even under plaintiff’s advocated standard.” (Fns. omitted.) Thus, fairly read, one of the trial court’s findings was that AYS failed to prove that the NSF 61 § 8 test medium “behaves like California drinking water.” As reviewed above, defendants’ evidence was to the effect that AYS’s test medium does not “behave like” California drinking water, but rather causes the leaching of lead from the defendants’ products. As weighed by the trial court, AYS’s evidence to the contrary was no more persuasive than was defendants’ evidence. In making this determination, the trial court did not rely upon its interpretation of Regulation 12901 that the test medium must be actual California drinking water, but rather cited the failure of AYS, which had the burden of proof, to use actual California drinking water to shed further light on its claim that the test medium “behaves like” California drinking water. In other words, under this alternative analysis given by the trial court, the failure to use actual California drinking water was not viewed as constituting a stand-alone disqualifying factor, but rather as giving rise to a negative inference about the utility of the NSF 61 § 8 test medium to predict what would happen in the real world of California drinking water. Beyond all that, the trial court further found that, whether or not the “specific medium” requirement of Regulation 12901. must be met, AYS failed *447 to prove that NSF 61 § 8 is scientifically valid to test for lead in California drinking water. The trial court found “[a]s is set forth above, the parties’ experts disagreed as to whether NSF 61 is scientifically valid to test for lead in California drinking water, and this court found each side equally convincing. Thus, plaintiff did not meet its burden of proof on this point.” That this finding did not depend upon any perceived requirement that the test medium be actual California drinking water appears from the trial court’s opinion; “As is set forth above, the parties disagree as to whether [Regulation] 12901 applies to the plaintiff’s burden of proof. Plaintiff claims that it need only demonstrate that a method of analysis is scientifically valid, not one that complies with [Regulation] 12901. Defendants claim that this argument is of no moment because Tier 4 allows for the utilization of a scientifically valid method of analysis, which defendants say is the same standard advocated by plaintiff. The potential significance of whether [Regulation] 12901 applies to a plaintiff’s burden of proof is not whether the ‘scientifically valid’ standard applies but rather whether the ‘specific medium’ requirement must be satisfied. As is set forth below, in this case, this question is not determinative of anything.” The trial court made further findings, and reached further conclusions, in the context of reaching its conclusion that NSF 61 § 8 does not meet the requirements of any of the four tiers of Regulation 12901. Those findings and conclusions are discussed below. DISCUSSION A. Standard of Review A trial court’s findings of fact are reviewed to determine whether they are supported by substantial evidence. (Foundation for Taxpayer & Consumer Rights v. Garamendi (2005) 132 Cal.App.4th 1354, 1366 [34 Cal.Rptr.3d 354].) A trial court’s interpretation of an administrative regulation is a legal determination and is reviewed de novo. (Delta Air Lines, Inc. v. State Bd. of Equalization (1989) 214 Cal.App.3d 518, 525 [262 Cal.Rptr. 803].) However, our interpretation, even if it differs' from that of the trial court, does not end our review. “Our task is to determine whether the judgment should be affirmed or reversed. Thus, we review the judgment for reversible error, not merely to determine whether the trial court’s interpretation . . . was correct, but whether the judgment is correct on any theory. [Citation.] ‘No rule of decision is better or more firmly established by authority, nor one resting upon a sounder basis of reason and propriety, than that a ruling or decision, itself correct in law, will not be disturbed on appeal merely because given for a wrong reason. If right upon any theory of the law applicable to the case, it must be sustained regardless of the considerations which may have moved *448 the trial court to its conclusion.’ [Citation.]” (Hiser v. Bell Helicopter Textron Inc. (2003) 111 Cal.App.4th 640, 647-648 [4 Cal.Rptr.3d 249].) “We review the trial court’s ruling, not its reasoning.” (Bridgestone Corp. v. Superior Court (2002) 99 Cal.App.4th 767, 774 [121 Cal.Rptr.2d 673]; see also Florio v. Lau (1998) 68 Cal.App.4th 637, 653 [80 Cal.Rptr.2d 409].) Against the background of these legal principles, we must determine initially whether the trial court’s interpretation of Regulation 12901 is correct. If not, we then must determine whether, as applied to a correct interpretation, the trial court’s findings of fact, as supported by substantial evidence, warrant affirmance or require reversal. B. Specific Medium AYS contends that the trial court’s interpretation of the “specific medium” as requiring actual California drinking water, for determining the existence of a discharge under the Act, is an unduly narrow interpretation that contravenes both the holding in Mateel, supra, 115 Cal.App.4th 8, and the very purpose of Proposition 65. We agree. In determining the meaning of the term “specific medium,” as used in subdivision (a) of Regulation 12901, we apply the same standards as if interpreting a statute. (Consumer Cause, Inc. v. Weider Nutrition Internat., Inc. (2001) 92 Cal.App.4th 363, 368 [111 Cal.Rptr.2d 823].) “Our foremost aim is to ascertain the intent of the agency issuing the regulation to effectuate the purpose of the law. [Citations.] When the agency’s intent cannot be discerned directly from the language of the regulation, we may look to a variety of extrinsic aids, including the purpose of the regulation, the legislative history, public policy, and the regulatory scheme of which the regulation is a part. [Citation.] Whenever possible, we will interpret the regulation to make it workable and reasonable. [Citation.]” (Manriquez v. Gourley (2003) 105 Cal.App.4th 1227, 1235 [130 Cal.Rptr.2d 209].) Under Health and Safety Code section 25249.5, the Proposition 65 discharge prohibition, “No person in the course of doing business shall knowingly discharge or release a chemical known to the state to cause cancer or reproductive toxicity into water or onto or into land where such chemical passes or probably will pass into any source of drinking water, notwithstanding any other provisions or authorization of law except as provided in Section 25249.9.” Section 25249.9 sets forth exemptions from the discharge prohibition, with the burden of proof on the exemption issue being expressly put on the defendant. As applicable here, “Section 25249.5 shall not apply to any discharge or release that meets both of the following criteria: [][] (1) The discharge or release will not cause any significant amount of the discharged *449 or released chemical to enter any source of drinking water; [f] (2) The discharge or release is in conformity with all other laws and with every applicable regulation, permit, requirement, and order.” (§ 25249.9, subd. (b).) Proposition 65 defines “significant amount” as “any detectable amount except an amount which would meet the exemption test in subdivision (c) of Section 25249.10 if an individual were exposed to such an amount in drinking water.” (§ 25249Í11, subd. (c).) The act does not define “detectable amount.” However, Regulation 12901, subdivision (a) defines this term as follows: “For purposes of Section 25249.11, subdivision (c), of the Health and Safety Code, the term ‘any detectable amount’ means a level detected using a method of analysis referred to in this section. For purposes of this section, ‘method of analysis’ refers to the method of detection or detection and calculation for a listed chemical in a specific medium, including, but not limited to, water, air, food, or soil, and shall include methods and procedures concerning the number of samples and the frequency and site of sampling that are specific for the listed chemical in question.” (Reg., § 12901, subd. (a), italics added.) However, the term “specific medium” is not further defined in the regulations. In Mateel, supra, 115 Cal.App.4th 8, the plaintiff alleged that defendant manufacturers of plumbing parts caused discharges of lead into drinking water in violation of Proposition 65. (115 Cal.App.4th at p. 15.) The issue addressed by the court centered around the Proposition 65 regulations that (a) establish a hierarchy of the four permissible “tiers” or tests that may be used to measure whether a listed toxin has been discharged, and (b) provide that a discharge occurs if, and only if, the toxin is detectable using a test from the highest available tier. (Reg., § 12901.) That issue arose because defendants had obtained summary judgment on the grounds that a “Tier 1” test, called the “Lead and Copper Rule,” existed for detecting lead discharges from the plumbing parts at issue, but that the plaintiff failed to use that test as its chosen method of analysis. (Mateel, supra, at pp. 16, 21-22.) The appellate court reversed, holding that, because the Lead and Copper Rule was designed and adopted to monitor for lead and copper levels only in public water systems, it was not a Tier 1 test for lead from sources other than public water systems. (Mateel, supra, 115 Cal.App.4th at pp. 20-22.) “When an agency designs, adopts, or employs a test in a manner that renders it specific to a particular context or environment, that test qualifies as a method of analysis only when applied in the same context or environment.” (Id. at p. 22, italics added.) Because of “detailed context-specific sampling procedures . . . [that] limit the applicability of the Lead and Copper Rule,” it “cannot be applied outside *450 a public water system.” (Mateel, supra, 115 Cal.App.4th at p. 22.) Further, the particular multiple sampling requirements of the Lead and Copper Rule caused it to be “of no use for detecting lead discharges from a single specific plumbing part such as the galvanized steel parts at issue” in the case. (Ibid.) The court was unpersuaded by the fact that the Lead and Copper Rule involved the testing for lead in water. “[T]he fact that this case involves testing for lead in water, and the Lead and Copper Rule also involves testing for lead in water, does not qualify the Lead and Copper Rule as a Tier 1 test for the purposes of this case. [I] ... A scientifically valid, state-adopted test suitable for measuring a particular toxin in a particular medium in one unique environment may not be a suitable way to test for that toxin in that medium in any other environment.” (Mateel, supra, 115 Cal.App.4th at pp. 22-23.) The court further stated that given the unique context of the Lead and Copper Rule, an entire water distribution system, there was “no likelihood that regulated entities in wholly different contexts will already be using the method to self-monitor.” (Id. at p. 23.) Accordingly, the Mateel court held that the Lead and Copper Rule was not a Tier 1 test for purposes of that case. (Id. at p. 24.) In so holding, the court reasoned that “Regulation 12901’s testing hierarchy was intended to eliminate threshold disputes over conflicting test results using different methodologies and over which tests should be used. It was not intended to create an insuperable barrier to identifying actual discharges when they occur.” (Ibid.) The defendants in Mateel argued that summary judgment was proper on the alternative basis that several of the plaintiff’s tests, including an NSF test, were inadmissible for failing to comply with the standards for scientific reliability imposed by People v. Kelly, supra, 17 Cal.3d 24. (Mateel, supra, 115 Cal.App.4th at pp. 24-25.) However, the trial court did not exclude plaintiff’s tests on that basis. Rather, the exclusion was based on the trial court’s incorrect interpretation that the Lead and Copper Rule was a valid Tier 1 test. (Id. at p. 26.) The Mateel court declined to decide whether the challenged tests were admissible under Kelly, stating “it is not for us to determine whether the trial court’s ruling might still be correct on a factual andevidentiary basis not yet considered by it. That task is reserved for the trial court.” (Ibid., italics added.) The Mateel court then determined that the trial court, on remand, was required to reconsider the defendants’ motions in limine and to determine whether the challenged tests were admissible under Kelly. (Mateel, supra, 115 Cal.App.4th at p. 26.) The court concluded its decision with the following passage: “To assist the trial court on remand, we address one final issue. The Plumbing Distributors contend that Mateel’s tests are categorically invalid because they were not performed on drinking water, as opposed to any other *451 water, and Regulation 12901 requires that tests be performed for the listed chemical in a ‘specific medium.’ Nothing in Regulation 12901 or the supporting Final Statement of Reasons supports such a narrow construction of the phrase ‘specific medium.’ In particular, Regulation 12901 defines a method of analysis as a ‘method of detection or detection and calculation for a listed chemical in a specific medium, including, but not limited to, water, air, food, or soil . . . .’ [Citation.] The examples given show that ‘water’ is a specific medium under Regulation 12901, and that a test for lead in ‘water’ will qualify as a method of analysis so long as it meets the other requirements of Regulation 12901. Mateel’s tests are not categorically invalid for any alleged failure to use ‘drinking water.’ ” (Id. at p. 26, fn. omitted.) Preliminarily, we agree with AYS that the Mateel court’s interpretation of Regulation 12901 on the subject of water as a “specific medium” is not appropriately characterized as dictum, as it was not “general argument or observation unnecessary to the decision. . . .” (See, e.g., United Steelworkers of America v. Board of Education (1984) 162 Cal.App.3d 823, 834-835 [209 Cal.Rptr. 16].) However, even if the court’s interpretation can fairly be characterized as irrelevant to the primary issue of whether the trial court erred in determining that the Lead and Copper Rule was an appropriate Tier 1 test, the interpretation was directly responsive to an argument raised by defendants and was intended to guide the trial court on remand. Thus, it is not properly characterized as mere dictum. (Ibid.) We also agree with AYS that the trial court’s interpretation of Regulation 12901 as requiring actual California drinking water is an unduly narrow construction of the term “specific medium.” 9 The trial court opined that the language of Regulation 12901 “mandates that the medium of the method of analysis must be the specific real world medium in which the listed chemical is to be detected.” However, nothing in Regulation 12901 or the final statement of reasons in support of the enactment requires an exact identity of medium. Moreover, as a remedial statute, Proposition 65 must be “broadly construed” to effectuate its remedial purpose. (People ex rel. Lungren v. Superior Court, supra, 14 Cal.4th at pp. 313-314.) As previously quoted, in the initial statement of reasons in support of the proposed amendments to Regulation 12901, OEHHA, in its own words, sought to “clarify” that the “medium” requirement means that the medium to *452 be tested must “closely duplicates the actual exposure conditions” or “must be analogous to the item causing the real world exposure or discharge,” but that the test need not be performed on a sample obtained from an actual discharge. (OEHHA, Initial Statement of Reasons, supra, at p. 4, italics added.) OEHHA explained that, in its view, the need for clarification arose from difficulties that litigants and the courts were having in interpreting and applying the regulation. (OEHHA, Notice of Proposed Rulemaking, supra, at p. 3.) An amendment to a law that “clarifies” it is merely a statement of what that law has always been (see In re Marriage of McClellan (2005) 130 Cal.App.4th 247, 255 [30 Cal.Rptr.3d 5]), and an agency’s interpretation of its own prior regulation, although not binding on a court, may properly be used in determining the effect of the prior regulation. (See MW Erectors, Inc. v. Niederhauser Ornamental & Metal Works Co., Inc. (2005) 36 Cal.4th 412, 434 [30 Cal.Rptr.3d 755, 115 P.3d 41].) With these legal principles in mind, we find the agency’s interpretation of the meaning of the word “medium” in Regulation 12901 to be persuasive and also to be consistent with California law pertaining to the use of experimental evidence. (See Yamaha Corp. of America v. State Bd. of Equalization (1998) 19 Cal.4th 1, 7 [78 Cal.Rptr.2d 1, 960 P.2d 1031].) To establish a foundation for “experimental evidence,” it must be shown that the experiment is relevant and was conducted “under the same or similar conditions” to those events at issue. (People v. Bonin, supra, 47 Cal.3d at pp. 846-848; Culpepper v. Volkswagen of America, Inc. (1973) 33 Cal.App.3d 510, 521 [109 Cal.Rptr. 110].) However, the conditions need only be “substantially identical, not absolutely identical.” (Culpepper, at pp. 521-522.) We do not read Mateel as holding that any water suffices as a “specific medium” in all cases involving tests for drinking water. While stating that plaintiff’s tests were not “categorically invalid” for any alleged failure to use “drinking water,” the Mateel court expressly declined to make evidentiary rulings, in the first instance, on the admissibility of such tests. (Mateel, supra, 115 Cal.App.4th at pp. 26-27.) That issue was reserved for the trial court. (Id. at p. 26.) In short, as applied to the instant case, the test medium, to have validity, must “closely duplicate,” or be “substantially identical to,” the drinking water at issue. 10 As shown above, and as further discussed below, the trial court, on disputed evidence, found and determined that AYS did not prove that there *453 was close duplication or substantial identity. 11 In other words, AYS was found to have failed to meet its burden of proving that the NSF 61 § 8 extraction solutions or testing conditions were closely duplicative of, or substantially identical to, a real world discharge scenario involving defendants’ products. These findings and determinations were supported by substantial evidence. Furthermore, the trial court’s ruling is consistent with Mateel’s contextual analysis. The Mateel court held that the appropriate context for the Lead and Copper Rule was its application as a test for a public water system and not for testing specific plumbing parts. (Mateel, supra, 115 Cal.App.4th at p. 22.) Specifically, the court noted that it was of little consequence that the Lead and Copper Rule involved testing for lead in water, as that testing method was not valid outside of the context of analyzing an entire public water system. (Ibid.) Here, the undisputed evidence demonstrated that NSF 61 § 8 is used to provide the basis for voluntary product certifications to public water utilities when purchasing products. The disputed evidence supported the trial court’s determination that AYS failed to prove that NSF 61 § 8 has relevance outside of this context. The general consensus of the experts was that the test is not intended to predict how products will perform in actual use. AYS’s own expert conceded that there is no relationship between the test results derived from NSF 61 § 8 and a given product’s lead-leaching propensity once actually installed and used. Furthermore, the evidence supported the conclusion that NSF 61 § 8 test conditions are dramatically different from the environment under which the products are actually used. AYS contends that invalidating a test because the laboratory is a different “environment” would create an “insuperable barrier” to enforcing Proposition 65 cases. Thus, AYS argues that Mateel should not be read as holding that a given test may not be used outside of its “environmental context” (i.e., a test for an entire watersystem is of no use for testing individual plumbing parts). However, even accepting AYS’s interpretation arguendo, we reach the same conclusion. AYS’s own expert testified that the goal of NSF 61 § 8 is not to test drinking water, but is to “identify problems with specific products in . . . [][]... the synthesized water.” *454 According to AYS’s expert, the NSF 61 § 8 testing medium behaves like drinking water. However, according to one of the defense experts, the pH levels of the testing medium would aggressively cause lead leaching that otherwise would not occur when the products were exposed to California drinking water. The defense expert also provided evidence that NSF 61 § 8 testing conditions are substantially different from the actual environment in which defendants’ products are used. The trial court found all of the experts, whether called by AYS or defendants, were well-qualified and provided credibly supported opinions regarding how the test medium compared to actual California drinking water. The court further found that neither side’s expert testimony was more believable than the other side’s, and found that AYS failed to meet its burden of proof on this point. It is an established principle that the credibility of witnesses and the weight to be given their testimony are matters within the sole province of the trier of fact, here the trial court. (Smith v. Regents of University of California (1997) 56 Cal.App.4th 979, 985, fn. 5 [65 Cal.Rptr.2d 813].) An appellate court will not reweigh the evidence, but rather views the record in the light most favorable to the prevailing party and resolves all evidentiary conflicts and indulges all reasonable inferences in support of the judgment. (In re Marriage of Mix (1975) 14 Cal.3d 604, 614 [122 Cal.Rptr. 79, 536 P.2d 479].) Here, the record on appeal contains ample evidence supporting the findings of fact by the trial court. Furthermore, when those findings are viewed in the context of a correct interpretation of the “specific medium” requirement, the •inexorable conclusion is that NSF 61 § 8 was not an appropriate proxy medium. C. Four Tiers Each of the four tiers contains the “specific medium” requirement. Therefore, the foregoing analysis applies to each of those tiers and supports the trial court’s conclusion that NSF 61 was not an appropriate method of analysis. But even apart from the “specific medium” requirement, the trial court’s decision is affirmable for AYS’s failure to meet other requirements of the tiers. 1. Tier 1 The first tier of Regulation 12901 specifies that where a listed state agency “has adopted or employs a method of analysis for a listed chemical in a *455 specific medium, such method shall be the method of analysis for that chemical in that medium.” (Reg., § 12901, subd. (b).) At trial, while conceding that NSF 61 § 8 had not been “adopted” by a listed state agency, AYS argued that that testing methodology qualified as a Tier 1 method of analysis because it was “employed” by the DHS, which is one of the listed agencies in Regulation 12901, subdivision (b). In support of that claim, AYS relied on “Policy Memo 2002-04,” dated June 27, 2002, and the testimony of its author, Dr. David Spath, the Chief of the Division of Drinking Water and Environmental Management at DHS. The purpose of the memo, as stated on its face, was to establish “approaches” to be used by DHS drinking water management staff prior to the adoption of various water works regulations. One such approach is: “For new projects and construction by water systems . ... HD ... [f] If there is a site-specific reason to do so, include conditions in the system’s permit that require all materials to be Standard 61 listed by an ANSI accredited certifier. Suggested language: ‘All materials in contact with drinking water must be tested and certified as meeting specification of ANSI/NSF Standard 61—2001. This requirement shall be met under testing conducted by a product certification organization accredited for this purpose by the American National Standards Institute.’ ” Dr. Spath testified that this memorandum directs interim approaches for determining what products can be used in public water systems. AYS argued that the policy memo establishes that the DHS “employs” NSF 61 for purposes of Tier 1. There is no authority defining “employs” for the purposes of Regulation 12901. In the trial court, AYS defined the term using the following dictionary definitions: “to make use of, as an instrument, a means, a material, etc., for a specific purpose” or “to use or apply or apply for a specific purpose” or “[t]o apply (a thing) to some definite purpose, to use as a means or instrument, or as a material.” As AYS argued below, “the term ‘employ’ should be interpreted to mean that the Department of Health Services . . . must make use of, or apply, the method of analysis for some ‘definite purpose.’ ” The trial court agreed that, in the context of the instant case, AYS’s proffered definition was appropriate. The trial court then held that the “definite purpose” manifested in Regulation 12901 is measuring the presence of listed substances in a specific medium. Based on the evidence presented, the trial court concluded that DHS did not employ NSF 61 § 8 for purposes of Regulation 12901. Preliminarily, AYS contends that the trial court’s interpretation of “employ” was overly restrictive in that it excludes otherwise valid tests employed by an agency merely because that agency uses the test for non-Proposition 65 purposes. AYS further argues that “[i]t is illogical and contrary to OEHHA’s *456 intent to require enforcers and the regulated community to seek out new tests when an agency has employed a test for non-Proposition 65 purposes that also provides meaningful information for Proposition 65 compliance and enforcement.” However, nothing in the trial court’s decision precludes the use of a test with multiple purposes. As found by the trial court, the evidence established that, at least on an interim basis, DHS uses the NSF 61 to specify standards for materials to be used in public water projects “if there is site-specific reason for doing so.” The memo reflects that the use of NSF 61 was intended as a product certification standard where there was a “site-specific reason to do. so.” As further noted by the trial court, AYS provided no evidence as to what would constitute a “site-specific” reason. Moreover, the reference to NSF as a product certification standard is consistent with Mr. Gilbert’s testimony that water utilities use NSF 61 § 8 when making their purchasing decisions because it establishes what materials are contained in a given part, not how the part will actually perform in terms of releasing metals when it is installed. Against the foregoing background, the trial court concluded that the purpose for which DHS uses NSF 61 is not the same as attempting to detect lead in California drinking water. 12 Thus, the court said: “For example, the specification of a standard for materials in water systems may reflect an extremely conservative approach to material quality, far above what might reasonably be needed to prevent leaching of lead into the drinking water. Similarly, the reference to NSF 61 as a materials standard may reflect a judgment by the DHS that the use of non-California drinking water as a test medium is not important to, or is more convenient for, deciding what types of valves to use in public water systems.” It was AYS’s burden, under Tier 1, to prove that the DHS “employed” AYS’s testing methodology, and the evidence was sufficient to support the trial court’s conclusion that AYS failed to meet that burden of proof. In other words, substantial evidence supports the trial court’s determination that NSF 61 § 8 did not qualify as a valid Tier 1 test. 2. Tier 3 13 Tier 3 of Regulation 12901 applies only where Tiers 1 and 2 do not. Tier 3 specifies that where “a method of analysis for a listed chemical in a specific *457 medium which is generally accepted by the scientific community, as evidenced by its publication in compilations by professional and scientific associations or societies, such as the Association of Official Analytical Chemists, or in peer-reviewed technical journals published by such associations or societies, such method shall be the method of analysis for that chemical in that medium.” (Reg., § 12901, subd. (d).) At trial, AYS relied, in part, on expert testimony and the use of NSF 61 § 8 by 30 states (other than California) to support its contention of “general acceptance” in the scientific community. Dr. Edwards testified that: “NSF 61 is not only accepted as the measure of lead-leaching propensity of specific plumbing products, it’s also the only standard, the only test that I am aware of with any sort of acceptance that measures that target parameter . . . .” “It’s the—indeed, it’s the only standard we have to address lead leaching or metal leaching from specific products. That is widely understood.” As noted by the trial court, this evidence merely tends to show that NSF 61 § 8 is viewed in the relevant scientific community as reliable for comparing how various products will leach lead under the NSF 61 § 8 test conditions. Nothing is said about how the NSF test is viewed in the relevant scientific community as related to potential lead leaching outside of the NSF test conditions. In fact, Dr. Edwards acknowledged that there is no correlation between the test results derived from NSF 61 § 8 and a given product’s lead-leaching propensity once actually installed and used. The trial court found that the expert witness testimony “does not . . . establish that the scientific community has accepted NSF 61 § 8 as a method of detecting the presence of lead in California drinking water.” 14 That finding was supported by the evidentiary record. AYS’s evidence that 30 other states use the NSF 61 standard is equally unavailing, as it fails to establish that other states use NSF 61 to test the performance of plumbing parts once they have been installed. As stated by the trial court, “[t]he question to be asked is not whether other states use the standard, but rather what do they use it for?” The answer, as shown by the evidence, is that all 30 states identified by AYS as using NSF 61 refer to it within the context of product certification standards for products used in public water systems. Against the background of this evidence, the trial court found that “[n]one of the evidence presented establishes that other states use NSF 61 to test the *458 performance of valves once they have been installed,” and concluded that the evidence “is not sufficient to qualify NSF 61 as a method of analysis under Tier 3.” The finding is supported by the trial record. AYS further argues that NSF 61 § 8 meets the publication requirement because the test is published by a NSF-led consortium of drinking water professionals and also appears in the standards published by the AWWA. The contention is similarly without merit. First, even assuming that NSF can be considered a professional or scientific association, its self-publication of NSF 61, which it sells for use, cannot be deemed evidence of general acceptance in the scientific community. Second, Dr. Edwards’s testimony that AWWA publishes standards mentioning NSF 61 does not equate to evidence that AWWA endorses the test for purposes of detecting lead in drinking water. In fact, Mr. Gilbert, a former president of AWWA, testified that AWWA does not accept NSF 61 § 8 beyond its voluntary certification utility. Further, Mr. Ackroyd testified that the reference to NSF 61 in the AWWA standards does not mean that AWWA requires the test or approves of it. In finding that “[p]laintiff has also failed to meet its burden to establish general acceptance in the scientific community through evidence of publication by professional and scientific associations or societies,” the trial court noted that “[n]one of the publications presented by plaintiff deal with NSF 61 § 8 in the context of being a test for detecting lead in drinking water, let alone for such detection in California drinking water.” That finding and that conclusion find support in the record. We find substantial evidence supports the trial court’s determination that NSF 61 § 8 did not qualify as a valid Tier 3 test. 3. Tier 4 As applied to cases where the first three tiers of Regulation 12901 do not apply, the fourth tier specifies that where “a scientifically valid method of analysis has been developed for a listed chemical in a specific medium, such method shall be the method of analysis for that chemical in that medium.” (Reg. § 12901, subd. (e).) The trial court concluded that NSF 61 § 8 “has not been shown to be an appropriate method of analysis under Tier 4” for two different reasons. First, because the trial court “found each side equally convincing” on the issue of “whether NSF 61 is scientifically valid to test for lead in California drinking water,” the court held that “plaintiff did not meet its burden of proof on this *459 point.” Second, and “in addition,” the court found that “plaintiff failed to demonstrate that NSF 61 § 8 was ‘developed for a listed chemical in a specific medium’ as is required under Tier 4.” Each of these independently sufficient bases for rejecting application of Tier 4 is supported by the record. 15 We find substantial evidence supports the trial court’s determination that NSF 61 § 8 did not qualify as a valid Tier 4 test. D. The Effect of the Repeal of Regulation 12901 AYS contends that the repeal of Regulation 12901 requires that this case be remanded for a new trial. In support of this contention, AYS argues that repeal of Regulation 12901 constitutes a “purely procedural change in law,” warranting “immediate application.” Conversely, defendants argue that remand is improper because it would deprive them of a “complete defense” and would impermissibly change the legal effects of past events. Alternatively, defendants argue that remand is not warranted as the existing law compels the same result. “Generally, the same rules of construction and interpretation which apply to statutes govern the construction and interpretation of administrative regulations.” (Union of American Physicians & Dentists v. Kizer (1990) 223 Cal.App.3d 490, 504-505 [272 Cal.Rptr. 886], citing Cal. Drive-In Restaurant Assn. v. Clark (1943) 22 Cal.2d 287, 292 [140 P.2d 657].) “ ‘[I]t is an established canon of interpretation that statutes are not to be given a retrospective operation unless it is clearly made to appear that such was the legislative intent.’ [Citation.]” (Evangelatos v. Superior Court (1988) 44 Cal.3d 1188, 1207 [246 Cal.Rptr. 629, 753 P.2d 585].) A retroactive statute is one that changes the legal effects of past events. (Aetna Cas. & Surety Co. v. Ind. Acc. Com. (1947) 30 Cal.2d 388, 391 [182 P.2d 159] (Aetna Casualty); Tapia v. Superior Court (1991) 53 Cal.3d 282, 288-289 [279 Cal.Rptr. 592, 807 P.2d 434] (Tapia).) In determining whether a statute would have an improper retrospective application, courts have broadly distinguished between “substantive” and “procedural” statutes. (Tapia, supra, at p. 289; Brenton v. Metabolife Internat., Inc. (2004) 116 Cal.App.4th 679, 688 [10 Cal.Rptr.3d 702] (Brenton).) Substantive laws, imposing new or additional liability, invoke the general disfavor of retroactivity because they change the legal consequences *460 of the parties’ past conduct. (Aetna Casualty, supra, 30 Cal.2d at p. 394; Tapia, supra, 53 Cal.3d at p. 298; Russell v. Superior Court (1986) 185 Cal.App.3d 810, 814 [230 Cal.Rptr. 102].) “In contrast to changed substantive statutes, applying changed procedural statutes to the conduct of existing litigation, even though the litigation involves an underlying dispute that arose from conduct occurring before the effective date of the new statute, involves no improper retrospective application because the statute addresses conduct in the future.” (Brenton, supra, 116 Cal.App.4th at p. 689; see also Tapia, supra, at pp. 288-289.) Such statutes are “ ‘ “actually prospective in nature since they relate to the procedure to be followed in the future.[”] ’ [Citations.]” (Tapia, supra, at p. 288.) For this reason, it has been said that “ ‘it is a misnomer to designate [such statutes] as having retrospective effect.’ ” (Ibid.) In other words, procedural statutes are applied “prospectively” when they are “applied to a trial occurring after the law’s effective date, regardless of when the underlying . . . cause of action arose.” (Id. at p. 289.) Moreover, our Supreme Court has made it clear that “it is the law’s effect, not its form or label, which is important.” (Tapia, supra, 53 Cal.3d at p. 289.) In other words, “ ‘it is not significant whether the statute is labeled [as being] substantive or procedural in nature.’ [Citation.] Rather, what is determinative is the effect that application of the statute would have on substantive rights and liabilities.” (Moore v. State Bd. of Control (2003) 112 Cal.App.4th 371, 378 [5 Cal.Rptr.3d 116].) With these principles in mind, we conclude that Regulation 12901 is properly characterized as a substantive regulation. Subdivision (g) of Regulation 12901 expressly provides that for purposes of Proposition 65, “no discharge, release or exposure occurs unless a listed chemical is detectable as provided in this section.” (Reg. § 12901, subd. (g).) Additionally, the final statement of reasons in support of the enactment of Regulation 12901 provides that subdivision (g) “means that no violation of Proposition 65 can be found unless there exists a method of analysis which meets the requirements of this regulation.” (HWA, Final Statement of Reasons, supra, at p. 5.) Furthermore, OEHHA acknowledged in its Final Statement of Reasons in Support of Repeal that the regulation had been “used[] primarily as an affirmative defense to litigation that was filed against businesses” and that they “will no longer be able to derive a benefit from the regulation . . . .” (OEHHA, Final Statement of Reasons, Repeal, supra, at pp. 17-23.) We agree that the repeal of Regulation 12901 constitutes more than a procedural change. In Morris v. Pacific Electric Ry. Co. (1935) 2 Cal.2d 764 [43 P.2d 276] (Morris), the court refused to retroactively apply an amendment of the former Vehicle Act, which eliminated a finding of negligence per se based on *461 speeding. (2 Cal.2d at pp. 768-769.) At the time of trial, the law provided that speeding was unlawful and constituted negligence per se. (Id. at pp. 766-767.) As such, a plaintiff was unable to recover damages suffered in an accident if the plaintiff’s speed contributed to the accident. (Id. at p. 767.) However, the trial court refused to instruct the jury on this existing principle of law. In absence of this instruction, the plaintiff prevailed, even though there was evidence that he was speeding immediately before the accident. (Ibid.) On appeal, the plaintiff argued that the refusal to give the negligence per se instruction, even if in error, was not prejudicial in light of an amendment to the law that occurred while the appeal was pending. (Morris, supra, 2 Cal.2d at pp. 767-768.) Thus, during the appeal, the Legislature had changed the law to provide that speeding was no longer negligence per se, and that the opposing party bore the burden of establishing that driving a vehicle at excessive speed constituted negligence. (Id. at p. 768.) The plaintiff sought to apply the amended law to his case, arguing that it was a “mere procedural change—a change in the rule of evidence—which is applicable alike to all actions, including those pending appeal at the time of its enactment . . . .” (Ibid.) However, the court rejected the plaintiff’s argument on the ground that “[a]t the time of the accident involved herein the defendant had an absolute defense, based upon the finding which the jury could have made in its favor on the evidence. ... To apply the amendment now so as to cut off such a substantial defense would be arbitrary and unjust.” (Morris, supra, 2 Cal.2d at pp. 768-769.) In so holding, the court relied on the principle that “the legislature may not, under pretense of regulating procedure or rules of evidence, deprive a party of a substantive right, such as a good cause of action or an absolute or a substantial defense . . . .” (Id. at p. 768.) AYS argues that Morris is not applicable to the instant case because defendants fail to demonstrate that the repeal of Regulation 12901 affects substantial rights. By this argument, AYS does violence to the plain language of subdivision (g), which essentially precludes a Proposition 65 claim unless a discharge is detectable under Regulation 12901. AYS also overlooks the discussion in the Final Statement of Reasons in Support of Repeal to the effect that a defense was being eliminated by virtue of the repeal. Rather, AYS asserts that Regulation 12901 is merely a procedural “screening mechanism” for establishing discharges under Proposition 65 that should be applied prospectively on remand. In support of this contention, AYS cites to Brenton, supra, 116 Cal.App.4th 679, for the proposition that “appellate *462 courts consistently apply new procedural rules to cases before them, notwithstanding whether the trial has concluded.” That proposition is neither remarkable nor pertinent, and Brenton is easily distinguishable from the case at bar. The Brenton court held that a posttrial amendment (Code Civ. Proc., § 425.17) to the anti-SLAPP statute (Code Civ. Proc., § 425.16) removing certain claims from that statute’s reach would apply to that case, thereby providing an independent basis for affirming the trial court’s order denying the defendant’s anti-SLAPP motion as to certain claims. (Brenton, supra, 116 Cal.App.4th at pp. 688-691.) The court rested its decision upon its conclusion that “Section 425.17 is properly characterized as a procedural statute applicable to pending actions.” (Id. at p. 689.) In reaching that conclusion, the court first noted that the anti-SLAPP statute is “ ‘merely ... a procedural screening mechanism for determining whether a plaintiff can demonstrate sufficient facts ... to permit the matter to go to a trier of fact.’ [Citation].” (116 Cal.App.4th at p. 689.) As such, the court further observed that “[s]ection 425.17 does nothing more than amend section 425.16 to except certain claims from applicability of the statutorily conferred remedy of the screening mechanism provided by section 425.16; section 425.17 does not impose new, additional or different liabilities based on past conduct or deprive Mil [the defendant] of any substantive defense to the action.” (Brenton, supra, at pp. 689-690, italics added.) Brenton does not support remand of this matter. Here, unlike in Brenton, the repeal of Regulation 12901 would deprive defendants of a substantive defense. Even assuming arguendo that Regulation 12901 could be considered “procedural,” the result in the instant case would be no different. As Brenton acknowledges, the labeling of a law is not dispositive. (Brenton, supra, 116 Cal.App.4th at p. 689.) Our Supreme Court has held that “[i]f substantial changes are made, even in a statute which might ordinarily be classified as procedural, the operation on existing rights would be retroactive because the legal effects of past events would be changed, and the statute will be construed to operate only in futuro unless the legislative intent to the contrary clearly appears. [Citations.]” (Aetna Casualty, supra, 30 Cal.2d at p. 394, italics added; see also In re Marriage of Buol (1985) 39 Cal.3d 751, 758-759 [218 Cal.Rptr. 31, 705 P.2d 354] [Legislature may not apply changes in the rules of evidence or procedure when the changes directly affect vested rights]; Brown v. Ferdon (1936) 5 Cal.2d 226, 231, 234-236 [54 P.2d 712] [change in Civ. Code could not be applied retroactively where it affected substantive rights].) *463 For example in Beeman v. Burling (1990) 216 Cal.App.3d 1586, 1606-1607 [265 Cal.Rptr. 719], the court refused to allow retroactive application to an amendment to Code of Civil Procedure section 473. There, the appeal was from an order denying a motion for relief from default. (Beeman, supra, at p. 1590.) The trial court had denied the motion for failure of counsel to make the requisite showing of excusable neglect. (Id. at p. 1602.) While the appeal was pending, section 473 was amended to eliminate the requirement that counsel show excusable neglect. The new law made relief from default mandatory whenever an attorney attests that his or her mistake or inadvertence resulted in the default at issue. (Beeman, supra, at p. 1604.) Appellant argued that the change in law was merely procedural and could be applied to his case without violating the rule against retroactive application. (Beeman, supra, 216 Cal.App.3d at p. 1605.) The court refused to apply the amended statute retroactively. “[Appellant’s position, which rests on the assumption that ‘procedural’ rules such as Code of Civil Procedure section 473 are retroactive, is overly simplistic and begs the key question whether past events will be affected by retroactive application of the amendment. If the amended version of section 473 were applied retroactively in this case, appellant would have an opportunity to go forward and defend this case on the merits, despite the fact that under the law existing at the time of the default judgment, he had clearly lost that right. As such, the retroactive application of this amendment would substantially affect substantive rights; under Aetna and its progeny this is impermissible. Accordingly, the amended version of Code of Civil Procedure section 473 cannot be applied retroactively to provide appellant relief from default.” (Beeman, supra, at p. 1607.) Similarly, in Nelson v. A. H. Robins Co. (1983) 149 Cal.App.3d 862, 870-871 [197 Cal.Rptr. 179], this division declined to retroactively apply an amendment to Code of Civil Procedure section 581a. There, a plaintiff’s personal injury claim against the manufacturer of an intrauterine device was dismissed in February 1980 for failing to serve the amended complaint on the manufacturer within three years after the commencement of the action pursuant to section 581a. (Nelson, supra, at p. 865.) However, after the trial court dismissed the action, the Supreme Court issued its decision in Hocharian v. Superior Court (1981) 28 Cal.3d 714 [170 Cal.Rptr. 790, 621 P.2d 829], which set forth guidelines for trial courts in determining whether a plaintiff may be excused from mandatory dismissal under section 581a upon a showing of due diligence. (Nelson, supra, at p. 868.) Subsequently, in 1982, section 581a was amended, providing that reasonable diligence was not applicable under the new law. (Nelson, at p. 870.) Since Hocharian did not change existing law, its teaching would apply to the Nelson case, and remand would be required for further hearing, unless the *464 1982 amendment was retroactive. (Nelson, supra, 149 Cal.App.3d at pp. 868-869.) Thus, the issue arose whether the 1982 amendment was retroactive. We determined that the amendment of Code of Civil Procedure section 581a met the definition of a “procedural” statute “because it neither created a new cause of action nor eliminated a defense.” (Nelson, supra, 149 Cal.App.3d at p. 870.) Nevertheless, we declined to apply the amendment retroactively due to its impermissible effect on past events. (Id. at p. 871.) “Commencing in February 1980 plaintiff was entitled to relief from the trial court’s error in failing to conduct a ‘Hocharian hearing’ on the motion by Robins for an order dismissing the action pursuant to section 581a, subdivision (a). For purposes pertaining to the construction of the 1982 enactment which amended the statute, the error was a ‘past event’ because it preceded the effective date of the enactment by almost three years. In the absence of a provision in the enactment expressly declaring or implying that it was to be given retroactive effect, it may not be construed in such manner that ‘the legal effects of past events would be changed.’ [Citation.]” (Ibid., italics added.) Here, the sole issue at phase IA of the trial was whether AYS’s testing procedure qualified as a method of analysis under Regulation 12901. AYS even stipulated to the regulation’s application in the case management order trifurcating the trial. AYS further stipulated that if it could not meet its prima facie burden of establishing that its testing method complied with Regulation 12901, the case would not further proceed and judgment would be entered in favor of defendants. The findings made, and the conclusions reached, by the trial court are “past events” in that they preceded the effective date of the repeal by nearly two years. Furthermore, there is nothing in the final statement of reasons in support of the repeal expressly declaring that it was intended to be given retroactive effect. To the contrary, OEHHA stated that the proposed regulatory action “was not developed with the intent to influence the outcome of any pending court cases.” (OEHHA, Final Statement of Reasons, Repeal, supra, at p. 4, italics added.) We conclude that the repeal of Regulation 12901 is not to be given retroactive application. 16 *465 DISPOSITION The judgment is affirmed. Defendants are entitled to their costs on appeal. Reardon, Acting R J., and Sepulveda, J., concurred. A petition for a rehearing was denied January 4, 2006, and appellant’s petition for review by the Supreme Court was denied March 15, 2006, S140591. George, C. J., did not participate therein. Footnotes [*] Judge of the San Francisco Superior Court, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution. [1] All further references to regulations are to title 22 of the California Code of Regulations. [2] Section 25249.12 of the Health and Safety Code authorizes the lead agency designated by the Governor to “adopt and modify regulations, standards, and permits as necessary to conform with and implement this chapter and to further its purposes.” In 1987 the HWA was the initial “lead agency” designated by the Governor to oversee Proposition 65. (People ex rel. Lungren v. Superior Court (1996) 14 Cal.4th 294, 309 [58 Cal.Rptr.2d 855, 926 P.2d 1042].) In 1991, when the California Environmental Protection Agency was created, the Governor transferred “lead agency” responsibilities to the Office of Environmental Health Hazard Assessment *438 (OEHHA), which had been transferred to that newly created agency. (Id. at p. 310 & fn. 6.) For simplicity, we refer to the “lead agency” as OEHHA. [3] We grant defendants’ request to take judicial notice of the June 4, 2004, notice of proposed rulemaking. (Evid. Code, § 452, subd. (c).) [4] Though not requested by the parties, we take judicial notice of both the June 2004 initial statement of reasons and the proposed amendments, on our own motion. (Evid. Code, § 452, subd. (c).) [5] The products identified in the complaint are copper alloy (including bronze and brass) backflow preventers, pressure regulators, pressure reducing valves, ball valves, gate valves, check valves, stop valves, plumbing valves, and strainers that come into contact with drinking water in drinking water distribution systems. [6] NSF is now known as NSF International. For simplicity, we refer to the organization as NSF. [7] Mr. Gilbert was also the former executive officer for the State Water Resources Control Board and former president of the American Water Works Association (AWWA). [8] NSF appoints a joint committee of 30 users/consumers, industry, and regulatory members to develop and revise its standard on an ongoing basis. [9] Contrary to AYS’s contention, the trial court’s interpretation does not result in a partial repeal of Evidence Code section 600, subdivision (b), or otherwise conflict with that section. Section 600, subdivision (b) provides that “[a]n inference is a deduction of fact that may logically and reasonably be drawn from another fact or group of facts found or otherwise established in the action.” Nothing in the trial court’s interpretation of Regulation 12901 can reasonably be read as constituting a complete bar to the use of circumstantial evidence. [10] The record reflects that actual water can be a variable medium, depending on when a sample is taken. However, the fact that the various constituents of actual water can vary does not mean that a proxy medium is impossible to create. For example, as the trial court suggested, a representative sample could be formulated by taking random samples. [11] Even though the use of actual California drinking water as a “specific medium” under the regulations was not required, the trial court was entitled to consider, and give weight to, the fact that AYS did not run the NSF 61 § 8 test on defendants’ products in order, in the trial court’s own words, to “shed light on whether or not the test water behaves like actual drinking water.” When juries are fact finders, they are routinely instructed that, “You may consider the ability of each party to provide evidence. If a party provided weaker evidence when it could have provided stronger evidence, you may distrust the weaker evidence.” (CACI No. 203.) Of course, the same principle applies to a judge in a court-tried case. [12] The trial court did not rest this conclusion upon the failure to use California drinking water as a specific medium, but rather upon the disconnect between the purpose of the NSF 61 test and anything to do with detecting lead in California drinking water. [13] AYS did not claim that NSF 61 § 8 qualifies as a method of analysis under Tier 2. [14] This finding is not dependent upon the failure to use California drinking water as a specific test medium. [15] Neither reason depends upon any requirement that the medium be California drinking water. [16] By reason of our conclusion that the repeal of Regulation 12901 is not entitled to retroactive application, we need not address defendants’ alternate argument that the same result as occurred at trial would reoccur under existing law if the matter were remanded.