Opinion

In Prudential-LMI Com. Insurance v. Superior Court (1990) 51 Cal.3d 674 [ 274 Cal.Rptr. 387 , 798 P.2d 1230 ] (Prudential-LMI), we examined the issue of allocation of indemnity among insurers in a first party property insurance case, where a loss had occurred over several policy periods but was not discovered until several years after it commenced. We found the “manifestation of loss rule” applicable, holding that the insurer insuring the property at the time appreciable property damage becomes manifest is solely responsible for indemnifying the insured once coverage is established. (Id. at p. 699.) We expressly reserved the question of what rules should apply in third party liability insurance cases involving continuous or progressively deteriorating damage or injury. We recognized there are substantial analytical differences between first party property and third party liability policies, and cautioned that we were intimating no view as to the application of our decision in either the third party liability or commercial liability (including toxic tort) context. ( Prudential-LMI, supra, 51 Cal.3d at pp. 679, 694; see also Garvey v. State Farm Fire & Casualty Co. (1989) 48 Cal.3d 395, 405-408 [ 257 Cal.Rptr. 292 , 770 P.2d 704 ] (Garvey).)

In this case we address the issue reserved in Prudential-LMI. Specifically, we must determine whether four comprehensive general liability (CGL) policies issued by defendant and respondent Admiral Insurance Company (Admiral) to plaintiff and appellant Montrose Chemical Corporation of California (Montrose) obligate Admiral to defend Montrose in lawsuits seeking damages for continuous or progressively deteriorating bodily injury and property damage that occurred during the successive policy periods. These losses, it is alleged, were caused by Montrose’s disposal of hazardous wastes at times predating the commencement of Admiral’s policy periods.

As explained below, we conclude that the standard CGL policy language, such as was incorporated into Admiral’s policies in issue in this case, provides coverage for bodily injury and property damage that occurs during *655 the policy period. In the case of successive policies, 1 bodily injury and property damage that is continuous or progressively deteriorating throughout several policy periods is potentially covered by all policies in effect during those periods. Stated in the insurance industry’s parlance, we conclude the “continuous injury” trigger of coverage should be adopted for third party liability insurance cases involving continuous or progressively deteriorating losses. 2 In this case, because the potential of coverage arose under Admiral’s policies, so too did its duty to defend Montrose in the underlying lawsuits.

As will further be explained, we also conclude, with respect to the “loss-in-progress” rule codified in Insurance Code 3 sections 22 and 250, that in the context of continuous or progressively deteriorating property or bodily injury losses insurable under a third party CGL policy, as long as there remains uncertainty about damage or injury that may occur during the policy period and the imposition of liability upon the insured, and no legal obligation to pay third party claims has been established, there is an insurable risk within the meaning of sections 22 and 250 for which coverage may be sought under such a policy.

We shall therefore affirm the judgment of the Court of Appeal reversing the summary judgment granted in favor of Admiral.

I

Facts and Procedural Background

From 1947 until 1982, Montrose manufactured the pesticide dichlorodiphenyl-trichlorethane (DDT) at its plant in Torrance, California. In 1972, *656 the federal government prohibited all domestic use of DDT. Montrose continued to manufacture the chemical for export at the Torrance facility until the plant closed in 1982.

Between January 1960 and March 1986, seven different carriers, ending with Admiral, furnished CGL policies to Montrose. Admiral issued four policies to Montrose, covering the period from October 13, 1982, to March 20, 1986. The remaining six CGL insurers involved in this litigation are not parties to this appeal. 4 Admiral’s policies obligate it to “pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of . . . bodily injury, or . . . property damage to which this insurance applies, caused by an occurrence. . . “Occurrence” is defined as “an accident, including continuous or repeated exposure to conditions, which results in bodily injury or property damage neither expected nor intended from the standpoint of the insured.”

The broad issue before the trial court was whether any of the seven CGL carriers, including Admiral, were obligated to defend Montrose in five actions pending against it in connection with Montrose’s disposal of toxic or hazardous wastes at several locations in California. Admiral joined in an interim defense agreement to provisionally fund Montrose’s defense (to this date the parties apparently still disagree as to whether such agreement was entered into subject to a complete reservation of rights, a matter of no direct concern in this appeal). When Montrose filed its declaratory relief action, Admiral moved for summary judgment on the issue of its duty to defend given the effective dates and terms of coverage of its policies. The trial court found there was no potential for coverage under Admiral’s policies, and thus that Admiral had no duty to defend the liability actions. We next briefly summarize the facts of the underlying actions as established by the evidence submitted in support of, and in opposition to, Admiral’s summary judgment motion.

1. The Stringfellow cases.

In an action initiated in 1983—United States v. J.B. Stringfellow (U.S. Dist. Ct. (C.D.Cal.)) No. C-83-2501 HLH—the United States and the State of California sued Montrose and numerous other businesses under the Comprehensive Environmental Response, Compensation, and Liability Act ( 42 U.S.C. § 9601 et seq.; hereafter CERCLA), as well as various state *657 environmental law provisions, seeking reimbursement for response costs incurred pursuant to the investigation, removal, and remediation of toxic waste contamination at and near the state-licensed class I hazardous waste disposal site known as the Stringfellow acid pits in Riverside County. The government also seeks damages for injury to natural resources, abatement of conditions, and cleanup at and near the Stringfellow site. The basis for the federal law claim against Montrose is strict liability under CERCLA for generating toxic waste shipped to the site.

The Stringfellow waste disposal site opened in 1956 and closed in 1972. Chemical wastes generated by Montrose were deposited there between 1968 and 1972, when Montrose paid a hauling company to transport byproducts of its DDT manufacturing process to the state-approved and licensed disposal facility. As early as 1970, toxic wastes were detected seeping from the site, and in 1975 the Santa Ana Regional Water Quality Control Board declared the site a public nuisance. It is noteworthy that the Stringfellow site was selected and designed as a hazardous waste disposal facility by the State of California, and that the site was used for that purpose by many defense contractors. In 1989, the State of California was found jointly and severally liable for the cleanup, both on strict liability and various fault-based common law grounds, due to its actions in designing, licensing and supervising the facility.

According to the allegations of the CERCLA complaint, the property damage commenced in 1956 and continued throughout the periods when Admiral’s CGL policies issued to Montrose were in effect. No bodily injury is alleged in the CERCLA action.

In a second lawsuit, a consolidated private party toxic tort action— Newman v. J.B. Stringfellow (Super. Ct. Riverside County, No. 165994MF) 5 —numerous plaintiffs seek damages from Montrose and other defendants for bodily injury and property damage alleged to have resulted from the release of contaminants at the Stringfellow site. Plaintiffs allege that the bodily injury and property damage occurred on a continuous basis, commencing in 1956 and extending to the present. Specifically, plaintiffs allege that 27 wrongful deaths occurred between 1982 and 1986 (the period Admiral’s policies were in effect), and that property damage was continuous throughout that same period.

*658 Although both Stringfellow cases involve allegations of progressively deteriorating property damage 6 caused by contaminants being released into, or migrating through, soil, groundwater, and surface water, only Newman v. Stringfellow additionally seeks damages for bodily injuries. According to the plaintiffs in both Stringfellow cases, between February 1982 and February 1983, the concentration of trichloroethylene (a suspected human carcinogen) tripled in the groundwater located between the Stringfellow site and the town of Glen Avon. On August 31, 1982, six weeks prior to commencement of the policy term under the first of Admiral’s policies issued to Montrose, Montrose was notified by the federal Environmental Protection Agency (EPA) that it considered Montrose a potentially responsible party (PRP) for money expended for response activities at the Stringfellow site. At about the same time, Montrose notified its environmental impairment liability (EIL) carrier, International Insurance Company, about the Stringfellow allegations, but did not notify Admiral. 7

2. The Levin Metals Cases.

The three remaining actions—Parr-Richmond Terminal Co. v. Levin Metals Corp. (U.S. Dist. Ct. (N.D.Cal.)) No. C-85-4776 SC, Levin Metals Corp. v. Parr-Richmond Terminal Co. (U.S. Dist. Ct. (N.D.Cal.)) Nos. C-84-6273 SC and 84-6324 SC, and Levin Metals Corp. v. Parr-Richmond Terminal Co. (Super. Ct. Contra Costa County, No. 255836)—are all interrelated. Each arises out of a state court action brought by Levin Metals against Parr-Richmond, alleging that real property sold by Parr-Richmond to Levin Metals in Contra Costa County in 1981 was contaminated by hazardous waste. 8 The suits allege both on-site and off-site contamination of soil, groundwater, and surface water, and seek damages for fraud based on Parr-Richmond’s failure to disclose the alleged contamination. All chemical processing at the Parr-Richmond Terminal site ceased in 1964 or 1965; the basis of Montrose’s alleged CERCLA liability is that it shipped chemicals to *659 the site prior to that time, which chemicals were then formulated into chemical products by an independent company, and that the formulator’s disposal of chemical waste byproducts in turn caused or contributed to the contamination. According to the plaintiffs in the Levin Metals cases, the environmental contamination at the Parr-Richmond site was discovered by them no later than August 1982. After the lawsuits were filed, Parr-Richmond cross-complained against Montrose and others for contribution and indemnity.

Although the Levin Metals cases were further complicated by Parr-Richmond’s efforts to avoid CERCLA liability and other related federal actions, for purposes of this appeal we need focus only on the lawsuits filed against Montrose for indemnity and contribution for allegedly contaminating the property in question in Contra Costa County during a period beginning in 1947, and continuing through the effective dates of Admiral’s policy periods.

3. Proceedings on Summary Judgment.

Montrose tendered defense of these actions to its seven CGL insurers, including Admiral. In 1986, Montrose sued the carriers in a declaratory relief action, seeking a declaration that the insurers had a duty to both defend and indemnify Montrose in all five underlying actions. 9 All the carriers except Admiral agreed to defend subject to a reservation of rights. In 1989, Admiral moved for summary judgment and summary adjudication of issues, urging the trial court to find (i) that it had no duty to defend or indemnify *660 Montrose in the Levin Metals cases because the circumstances which trigger coverage, within the meaning of the coverage clauses and definitions in its policies, did not occur during the policy periods, and (ii) that it had no duty to defend or indemnify Montrose in the Stringfellow cases because the contamination alleged in those actions was an uninsurable loss-in-progress prior to the effective date of the first policy issued by Admiral (Oct. 13, 1982).

The trial court granted summary judgment in favor of Admiral on each ground. First, with respect to the Levin Metals cases, the court held that coverage for third party claims of progressive property damage under a CGL policy is “triggered” when the damage is first discovered; in essence, an application of the “manifestation” or “manifestation of loss” rule we later adopted in Prudential-LMI, supra, 51 Cal.3d 674 , for progressive losses in first party property insurance cases. The trial court reasoned there was no possibility of coverage under Admiral’s policies because the third party Levins Metal claimants (although not Montrose, the insured) allegedly discovered contamination at the Parr-Richmond site no later than August 1982, before the start of Admiral’s first policy term.

Second, with respect to the Stringfellow cases, the trial court found that coverage was further barred under the “loss-in-progress” rule codified in sections 22 and 250. Those statutory provisions will be examined in greater detail below; for present purposes it will suffice to note the rule provides that insurance is a contract that indemnifies against a loss or losses arising from contingent or unknown events (§ 22), and that any such contingent or unknown event may be insured against subject to the limitations of the Insurance Code (§ 250). Relying on the PRP letter that Montrose received from the EPA in August 1982, informing Montrose it might be responsible for response and other cleanup costs at the Stringfellow site, the trial court concluded coverage was barred for all claims relating to the site because, prior to the commencement of Admiral’s policies issued to Montrose, Montrose knew its liability for property damage and/or bodily injury stemming from contamination at the site was “likely.”

Montrose appealed, and the Court of Appeal reversed the summary judgment order. The appellate court rejected a “manifestation of loss” or “discovery” trigger of coverage analysis (as employed in the first party insurance context), finding it incompatible with the language of Admiral’s third party CGL policies. It held that, because the underlying Levin Metals actions allege that continuous or progressively deteriorating property damage “occurred” throughout the period Admiral’s policies were in effect, potential *661 coverage under those policies was triggered, at least for purposes of the duty to defend. The court further held that the loss-in-progress rule did not bar coverage in the Stringfellow cases. It reasoned that Montrose’s potential liability to third parties for the progressive property damage alleged to have “occurred” throughout the period of Admiral’s policies was still “contingent,” and thus insurable, under section 250, even if damage as defined in the Admiral policies was inevitable, and notwithstanding Montrose’s earlier receipt of the PRP letter. The Court of Appeal remanded Admiral’s affirmative defense—that Montrose had concealed material facts prior to purchasing the CGL policies from Admiral—and further declined to address the insurer’s argument, not raised in the trial court, that coverage for progressive damage at the Stringfellow site is also barred under specific policy exclusions because Montrose “expected or intended” the progressive damage that occurred during Admiral’s policy periods. (§ 22.)

We granted Admiral’s petition for review to consider the complex and important issue of when potential coverage is triggered under a CGL policy where the underlying third party claims involve continuous or progressively deteriorating damage or injury, and how the loss-in-progress rule applies to such policies. 10

*662 II

Trigger of Coverage in Third Party Progressive Loss Cases

As noted, Admiral moved for summary judgment in the trial court on grounds that it had no duty to defend or indemnify Montrose in the Levin Metals cases because the circumstances which trigger coverage, within the meaning of the coverage clauses in its policies, did not occur during the policy periods, 11 and that it had no duty to defend or indemnify Montrose in the Stringfellow cases because the contamination alleged in those actions was an uninsurable loss-in-progress prior to the effective date of the first policy it issued to Montrose. Having convinced the trial court, but not the Court of Appeal, Admiral seeks to renew these claims. Admiral asserts in its brief on the merits that “the fact that the Stringfellow CERCLA action alleges continuing or progressive contamination does not establish there was an occurrence while Admiral’s policies were in effect.” Admiral submits that “all damage was caused by a single occurrence outside (i.e., prior to commencement of) Admiral’s policy period,” and urges that any determination that continuous or progressive damage or injury occurring during its *663 ensuing policy periods can itself trigger coverage, “ignore[s] the policy language and confuse[s] the consequences of the occurrence with the occurrence itself, i.e., the event that ‘resulted’ in damage.”

1. Preliminary considerations: distinguishing third party liability insurance from first party property insurance.

To properly analyze the trigger of coverage issues presented in this case, it is necessary to first clearly distinguish between third party liability insurance, the type of coverage here at issue, and coverage under a first party property insurance policy, such as the standardized homeowners policy in issue in Prudential-LMI, supra, 51 Cal.3d 674 .

As we observed in both Garvey, supra, 48 Cal.3d at page 399 , footnote 2, and Prudential-LMI, supra, 51 Cal.3d at pages 698-699, a first party insurance policy provides coverage for loss or damage sustained directly by the insured (e.g., life, disability, health, fire, theft and casualty insurance). A third party liability policy, in contrast, provides coverage for liability of the insured to a “third party” (e.g., a CGL policy, a directors and officers liability policy, or an errors and omissions policy). In the usual first party policy, the insurer promises to pay money to the insured upon the happening of an event, the risk of which has been insured against. In the typical third party liability policy, the carrier assumes a contractual duty to pay judgments the insured becomes legally obligated to pay as damages because of bodily injury or property damage caused by the insured. (Garvey, supra, 48 Cal.3d at p. 407 .)

The difference in the nature of the risks insured against under first party property policies and third party liability policies is also reflected in the differing causation analyses that must be undertaken to determine coverage under each type of policy. (Garvey, supra, 48 Cal.3d at p. 406 .) “ ‘Property insurance ... is an agreement, a contract, in which the insurer agrees to indemnify the insured in the event that the insured property suffers a covered loss. Coverage, in turn, is commonly provided by reference to causation, e.g., “loss caused by . . .” certain enumerated perils. [¶] The term “perils” in traditional property insurance parlance refers to fortuitous, active, physical forces such as lightning, wind, and explosion, which bring about the loss.’ ” (Ibid., quoting Bragg, Concurrent Causation and the Art of Policy Drafting: New Perils for Property Insurers (1985) 20 Forum 385, *664 386-387.) In contrast, “ 'the “cause” of loss in the context of a property insurance contract is totally different from that in a liability policy.'" (Garvey, supra, 48 Cal.3d at p. 406 , italics in original.) “[T]he right to coverage in the third party liability insurance context draws on traditional tort concepts of fault, proximate cause and duty. This liability analysis differs substantially from the coverage analysis in the property insurance context, which draws on the relationship between perils that are either covered or excluded in the contract. In liability insurance, by insuring for personal liability, and agreeing to cover the insured for his own negligence, the insurer agrees to cover the insured for a broader spectrum of risks.'' (Id. at p. 407, italics added.)

The parties’ expectations may also differ depending upon the type of coverage sought. First party property coverage is typically purchased in an amount sufficient to cover the insured’s maximum potential loss (e.g., fire insurance typically covers the value of the property insured). Hence, there is no reason for a first party insured to look to more than one policy in the event of loss (the policy in effect at the time of the fire). (See Garvey, supra, 48 Cal.3d at p. 406 .) Third party liability coverage differs substantially. As the Court of Appeal below observed, “[a]t best, the insured makes an educated guess about its potential exposure to third parties. At worst, the insured’s best guess falls far short of the mark.”

Yet another distinction between the two types of insurance coverage is that third party CGL policies do not impose, as a condition of coverage, a requirement that the damage or injury be discovered at any particular point in time. Instead, they provide coverage for injuries and damage caused by an “occurrence,” and typically define “occurrence” as an accident (or sometimes a “loss”), including a “continuous or repeated exposure to conditions,” that results in bodily injury or property damage during the policy period. The standardized CGL policy language (like the language in Admiral’s policies) will be reviewed in greater detail below. As will be seen, nothing about this language suggests a manifestation or discovery requirement as a prerequisite for triggering coverage. (See, e.g., Trizec Properties v. Biltmore Const. Co. (11th Cir. 1985) 767 F.2d 810 , 813 [no requirement in standard CGL policy that damages “manifest” themselves during the policy period].)

Another important difference between first and third party policies is that first party insurance policies require the insured to bring any action against the insurer within 12 months after “inception of the loss.” (Prudential-LMI, *665 supra, 51 Cal.3d at pp. 682-687.) Before an action is filed under such a policy, there must be a dispute between the insured and insurer. Before there can be a dispute, the insured must (or reasonably should) know it has suffered a “loss.” (Id. at pp. 686-687.) By contrast, third party liability policies do not include a 12-month limitations period in which the insured must bring an action against the insurer (although the policies may contain express notification requirements). It is the damaged or injured third party who initiates the action against the insured. If coverage is ultimately established, it is the insurer that in turn must indemnify the insured for “all sums which the insured shall become legally obligated to pay.” Hence, there is no “inception of the loss” language in a standard CGL policy, and, as will become apparent, no corollary need to apply the definition of “inception of the loss” that this court articulated in Prudential-LMI, supra, 51 Cal.3d at pp. 682, 699. (Cf. § 2071 [standard form fire insurance policy].)

Unfortunately, some courts have failed to draw these critical distinctions when discussing coverage issues under first and third party insurance policies. In the third party liability insurance context, some reported cases have muddied the waters by seemingly failing to distinguish between disputes arising between an insured and insurer, and actions among several CGL carriers that seek a judicial declaration allocating a loss already paid out to the insured under one or more such policies. In suits between an insured and an insurer to determine coverage, interpretation of the policy language and, in the case of ambiguous policy language, the expectations of the parties, will typically take precedence. The existence of excess or “secondary insurance” policies, “other insurance” clauses, or similar policy language decreeing the manner of apportionment of liability under multiple policies may also factor into the coverage analysis.

In contrast, where two or more CGL carriers turn to the courts to allocate the cost of indemnity for a paid loss, different contractual and policy considerations may come into play in the effort to apportion such costs among the insurers. The task may require allocation of contribution amongst all insurers on the risk in proportion to their respective policies’ liability limits (such as deductibles and ceilings) or the time periods covered under each such policy. Reported cases whose analyses fail to take these distinctions into account, although purporting to clarify or settle an underlying “trigger of coverage” issue, may shed more darkness than light on the matter.

The proper analysis and resolution of a trigger of coverage issue may also depend on whether the CGL policy in issue insures against liability to third *666 parties for bodily injury, property damage, or both. As will be shown, the coverage clauses in Admiral’s policies do not distinguish between the nature of the underlying harm (bodily injury or property damage) that triggers the insured’s liability coverage. Accordingly, Montrose and Admiral appear to agree that under a plain reading of that unambiguous aspect of the policy language, whatever be the circumstances (or timing of the circumstances) that will potentially trigger liability coverage under the policies, coverage will apply uniformly under such circumstances whether the claims be for bodily injury, or property damage, alleged in the underlying third party lawsuits.

Finally, the proper resolution of a trigger of coverage issue in any given case may turn on whether the court is addressing underlying facts involving a single event resulting in immediate injury (e.g., an explosion causing instantaneous bodily injuries and destruction of property), a single event resulting in delayed or progressively deteriorating injury (e.g., a chemical spill), or a continuing event (referred to in CGL policies as “continuous or repeated exposure to conditions”) resulting in single or multiple injuries (e.g., exposure to toxic wastes or asbestos over time). Significantly, in the present case we are dealing both with claims of continuous or progressively deteriorating bodily injury (the Newman v. Stringfellow lawsuit), and progressively deteriorating property damage (the Stringfellow and Levin Metals cases), all arising from continuous or repeated exposure to hazardous waste contamination over time, allegedly including the periods when Admiral’s policies were in effect.

With these considerations in mind, we turn next to the express language of the contracts of insurance here in issue, looking first to the relevant principles of insurance policy interpretation that must govern our construction of the contested provisions.

2. Admiral’s policy language and the applicable rules of interpretation.

Insurance policies are contracts and, therefore, are governed in the first instance by the rules of construction applicable to contracts. Under statutory rules of contract interpretation, the mutual intention of the parties at the time the contract is formed governs its interpretation. (Civ. Code, § 1636.) Such intent is to be inferred, if possible, solely from the written provisions of the contract. (Id., § 1639.) The “clear and explicit” meaning of these provisions, interpreted in their “ordinary and popular sense,” controls judicial interpretation unless “used by the parties in a technical sense, or unless a special meaning is given to them by usage.” (Id., §§ 1638, 1644.) If *667 the meaning a layperson would ascribe to the language of a contract of insurance is clear and unambiguous, a court will apply that meaning. (See AIU Ins. Co. v. Superior Court (1990) 51 Cal.3d 807, 822 [ 274 Cal.Rptr. 820 , 799 P.2d 1253 ] (AIU); Reserve Insurance Co. v. Pisciotta (1982) 30 Cal.3d 800, 807 [ 180 Cal.Rptr. 628 , 640 P.2d 764 ]; Crane v. State Farm Fire & Cas. Co. (1971) 5 Cal.3d 112, 115 [ 95 Cal.Rptr. 513 , 485 P.2d 1129 , 48 A.L.R.3d 1089 ].)

In contrast, “[i]f there is ambiguity . . . it is resolved by interpreting the ambiguous provisions in the sense the promisor (i.e., the insurer) believed the promisee understood them at the time of formation. (Civ. Code, § 1649.) If application of this rule does not eliminate the ambiguity, ambiguous language is construed against the party who caused the uncertainty to exist. (Id., § 1654.)” (AIU, supra, 51 Cal.3d at p. 822.) “This rule, as applied to a promise of coverage in an insurance policy, protects not the subjective beliefs of the insurer but, rather, ‘the objectively reasonable expectations of the insured.’ (AIU, supra, at p. 822 .) Only if this rule does not resolve the ambiguity do we then resolve it against the insurer. (See AIU, supra, at p. 822 .)” (Bank of the West v. Superior Court (1992) 2 Cal.4th 1254, 1265 [ 10 Cal.Rptr.2d 538 , 833 P.2d 545 ]; see also Cooper Companies v. Transcontinental Ins. Co. (1995) 31 Cal.App.4th 1094, 1101-1102 [ 37 Cal.Rptr.2d 508 ].)

We explained further in AIU, supra, 51 Cal.3d at page 822, that “[i]n the insurance context, we generally resolve ambiguities in favor of coverage. (See, e.g., State Farm Mut. Auto. Ins. Co. v. Jacober (1973) 10 Cal.3d 193, 197 [ 110 Cal.Rptr. 1 , 514 P.2d 953 ]; Bareno v. Employers Life Ins. Co. (1972) 7 Cal.3d 875, 878 [ 103 Cal.Rptr. 865 , 500 P.2d 889 ]; Continental Casualty Co. v. Phoenix Constr. Co. (1956) 46 Cal.2d 423, 437 [ 296 P.2d 801 , 57 A.L.R.2d 914 ].) Similarly, we generally interpret the coverage clauses of insurance policies broadly, [in order to protect] the objectively reasonable expectations of the insured. (See, e.g., Garvey v. State Farm Fire & Casualty Co.[, supra,) 48 Cal.3d 395, 406 ; Reserve Insurance Co. v. Pisciotta, supra, 30 Cal.3d at p. 808 .) These rules stem from the fact that the insurer typically drafts policy language, leaving the insured little or no meaningful opportunity or ability to bargain for modifications. (See, e.g., Garcia v. Truck Ins. Exchange (1984) 36 Cal.3d 426, 438 [ 204 Cal.Rptr. 435 , 682 P.2d 1100 ]; Bareno, supra, 7 Cal.3d at p. 878 .) Because the insurer writes the policy, it is held ‘responsible’ for ambiguous policy language, which is therefore construed in favor of coverage.” (Fn. omitted; see also Mehr et al., Principles of Insurance (8th ed. 1985) p. 137.)

Is the language of Admiral’s contracts of insurance here in issue “clear and explicit,” and thus controlling (Civ. Code, §§ 1638, 1644)—or is *668 it ambiguous, requiring us to interpret the coverage clauses broadly in order to protect the objectively reasonable expectations of Montrose, the insured? Some courts, including the Court of Appeal below, have concluded that the varying judicial constructions placed on the definition of occurrence in the standard form CGL policy themselves attest to the inherent ambiguity in that definition. (See California Union Ins. Co. v. Landmark Ins. Co. (1983) 145 Cal.App.3d 462, 472 [ 193 Cal.Rptr. 461 ].) One commentator has gone so far as to suggest that “[t]he word ‘occurrence’ itself is ambiguous because the injury process is not a definite, discrete event.” (Note, Developments in the Law—Toxic Waste Litigation (1986) 99 Harv. L.Rev. 1458, 1579.) Although any such ambiguity would ultimately have to be resolved in favor of the reasonable expectations of the insured (Bank of the West v. Superior Court, supra, 2 Cal.4th at p. 1265 ; AIU, supra, 51 Cal.3d at p. 822; Garvey, supra, 48 Cal.3d at p. 406 ), we find that the express language of Admiral’s policies of insurance, when read as a whole, unambiguously provides potential coverage for the continuous and progressively deteriorating bodily injury and property damage alleged to have occurred during Admiral’s policy periods.

Turning to the express policy language, Admiral contracted with Montrose to “pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of . . . bodily injury, or .. . property damage to which this insurance applies, caused by an occurrence. . . .” (Italics added.) “[Pjroperty damage to which this insurance applies” is defined in Admiral’s policies as “(1) physical injury to or destruction of tangible property which occurs during the policy period, including the loss of use thereof at any time resulting thereform . . . .” (Italics added.) 12 “Bodily injury” to which the insurance applies is defined as “bodily injury, sickness or disease sustained by any person which occurs during the policy period, including death at any time resulting therefrom.” (Italics added.) We find no ambiguity in this language; it clearly and explicitly provides that the occurrence of bodily injury or property damage during the policy period is the operative event that triggers coverage.

*669 Furthermore, “occurrence” is defined in Admiral’s policies as “an accident, including continuous or repeated exposure to conditions, which results in bodily injury or property damage neither expected nor intended from the standpoint of the insured.” (Italics added.) When read together with the aforementioned clauses defining covered bodily injury and property damage, this policy language unambiguously distinguishes between the causative event—an accident or “continuous and repeated exposure to conditions”— and the resulting “bodily injury or property damage.” It is the latter injury or damage that must “occur” during the policy period, and “which results” from the accident or “continuous and repeated exposure to conditions.” In this case, it is the third party litigants’ bodily injuries and property damage, which are alleged to have been continuous or progressively deteriorating throughout Admiral’s policy periods, and which allegedly resulted from the continuous and repeated exposure to toxic chemicals for which the insured, Montrose, is an allegedly responsible party, that triggers potential coverage under the policies in question.

3. Settled case law and the drafting history of the standardized CGL policy language confirm that coverage is triggered by damage or injury occurring during the policy period.

Admiral contends that to read its CGL policies as providing that coverage is triggered when damage or injury occurs within the policy periods as a result of an “occurrence” is to “ignore[] the policy language and confuse[] the consequences of the occurrence with the occurrence itself, i.e., the event that ‘resulted’ in damage.” (Ante, at p. 663.) Admiral in essence urges that coverage under a CGL policy is established at the time the “occurrence” (i.e., the precipitating act or event) first gives rise to appreciable damage or injury, and that policies that commence after an “occurrence” and some consequent appreciable damage or injury cannot be on the risk for progressive damage or injury that occurs during such subsequent policy periods. The relevant cases and interpretative authorities which have construed the industry-standardized CGL policy language lend no support to Admiral’s position.

California courts have long recognized that coverage in the context of a liability insurance policy is established at the time the complaining party was actually damaged. In Remmer v. Glens Falls Indem. Co. (1956) 140 Cal.App.2d 84 [ 295 P.2d 19 , 57 A.L.R.2d 1379 ] (Remmer), the court was asked to interpret the definition of “occurrence” as that term was used in a CGL policy. The precise issue in Remmer was whether the act of defectively grading and filling a lot constituted the sole occurrence giving rise to coverage under the policy’s “one occurrence” provision, or whether subsequent injury (an alleged maintenance of a nuisance on the graded lot *670 adjoining the third party claimants’ property) also triggered liability coverage under the policy. Relying on cases from California and other jurisdictions, the Remmer court formulated the following rule: “The general rule is that the time of the occurrence of an accident within the meaning of an indemnity policy is not the time the wrongful act was committed, but the time when the complaining party was actually damaged.” (Id. at p. 88.)

The Remmer formulation, which distinguishes between a wrongful act and the injurious result of that act, and holds that the triggering of liability coverage under a CGL policy is established at the time the complaining third party was actually damaged, has been embraced by such noted experts as Appleman (7A Appleman, Insurance Law & Practice (1979 rev.) § 4501.03, p. 256) and Couch (11 Couch, Insurance (2d ed. 1982) § 44:8, p. 194.) It can be found in American Jurisprudence Second (43 Am.Jur.2d (1982 rev.) Insurance, § 243, p. 324), has been accepted by the courts of many other states, and has been cited by federal courts interpreting the law of still other states. (See Annot., Event Triggering Liability Insurance Coverage as Occurring Within Period of Time Covered by Liability Insurance Policy Where Injury or Damage is Delayed—Modem Cases (1993) 14 A.L.R.5th 695 , § 6 and cases cited therein.) Indeed, as stated by the Idaho Supreme Court, “This rule is followed in every jurisdiction that has considered the issue except Louisiana.” (Millers Mut. Fire Ins., etc. v. Ed Bailey, Inc. (1982) 103 Idaho 377 [ 647 P.2d 1249, 1251 ].)

Although the Court of Appeal concluded that potential coverage was triggered under Admiral’s policies by damage or injury occurring during the policy periods, the court did not trace this long-standing interpretation of how liability coverage is triggered under a CGL policy to the rule formulated in Remmer . Instead, the court independently looked to the drafting history of the standard CGL policy language for support for its conclusion that no reasonable construction, other than that described above, could be placed on the insurance industry’s use of such policy language.

Admiral contends that evidence of the drafting history of the standardized CGL policy provisions and definitions, and available interpretative materials, are irrelevant and should not have been considered by the Court of Appeal in construing the language of its CGL policies issued to Montrose. Most courts and commentators have recognized, however, that the presence of standardized industry provisions and the availability of interpretative literature are of considerable assistance in determining coverage issues. (See, e.g., Maryland Casualty Co. v. Reeder (1990) 221 Cal.App.3d 961, 968 [ 270 Cal.Rptr. 719 ].) Such interpretative materials have been widely cited and *671 relied on in the relevant case law and authorities construing standardized insurance policy language. As one court has suggested, “where two insurers dispute the meaning of identical standard form policy language—the meaning attached to the provisions by the insurance industry is, at minimum, relevant.” (Fireman’s Fund Ins. Co. v. Aetna Casualty & Surety Co. (1990) 223 Cal.App.3d 1621, 1629 [ 273 Cal.Rptr. 431 ].) On the other hand, as another court has observed, “[w]hile insurance industry publications are helpful in understanding the scope of coverage insurers are trying to delineate in any given policy, they are by no means dispositive.” (American Star Ins. Co. v. Insurance Co. of the West (1991) 232 Cal.App.3d 1320, 1330 [ 284 Cal.Rptr. 45 ], italics in original.) In this case, we find the drafting history relevant in evaluating Admiral’s argument that, from a public policy standpoint, the insurance industry will be harmed by the adoption of a continuous injury trigger that the industry assertedly never anticipated would be applied to these policies.

Standard CGL policy language was revised by insurance industry drafters in several important respects starting in 1966. Prior to that year, third party general liability policies covered bodily injuries and damages caused by “accidents.” (American Home Prod. v. Liberty Mut. Ins. Co. (S.D.N.Y. 1983) 565 F.Supp. 1485, 1501 , affd. as mod. (2d Cir. 1984) 748 F.2d 760 .) In 1966, the National Bureau of Casualty Underwriters and the Mutual Insurance Rating Board, the predecessor organizations to the Insurance Services Office (ISO), 13 changed the standard form policy from an “accident-based” to an “occurrence-based” format. (Ibid., see also New Castle County v. Hartford Acc. and Indem. Co., supra, 933 F.2d at p. 1181; Pasich, Insurance Coverage for Environmental Claims (Jan. 1989) L.A.Law., p. 23, fn. 12; 3 Cal. Insurance Law & Practice, supra, Property and Liability Insurance, § 49.04, at p. 49-10.) It is reasonable to infer that the insurance industry knew precisely what the change entailed.

In comments addressing the question of coverage under the new CGL policies for progressive personal injury or property damage resulting over an extended period of time, one of the drafters explained that “[i]n some exposure type cases involving cumulative injuries it is possible that more than one policy will afford coverage.” (Elliott, The New Comprehensive General Liability Policy, in Liability Insurance Disputes (PLI, Schreiber edit. *672 1968), pp. 12-3 to 12-5; see also Obrist, The New Comprehensive General Liability Insurance Policy—A Coverage Analysis (Defense Research Inst. Monograph 1966) p. 6 [same]; Nachman, The New Policy Provisions for General Liability Insurance (1965) 18 CPU Annals 197, 200 [same].)

By 1966, the insurance industry was also demonstrating its awareness of potential coverage issues involving continuous or progressively deteriorating bodily injury and property damage. Richard H. Elliott, then secretary of the National Bureau of Casualty Underwriters, wrote the following regarding the adoption of the occurrence-based CGL policy, which standard form policy retained the term “accident” within its definition of occurrence: “The new policy will afford coverage on an ‘occurrence’ basis. ‘Occurrence’ is defined as ‘an accident, including injurious exposure to conditions, which results, during the policy period, in bodily injury and property damage neither expected nor intended from the standpoint of the insured.’ Note that this definition includes the word ‘accident.’ This has been done in order to clarify the intent with respect to time of coverage and application of policy limits, particularly in situations involving a related series of events attributable to the same factor. Under such circumstances only one accident or occurrence is intended as far as the application of policy limits is concerned. For example, the liability of a contractor arising out of the derailment of ten or twelve freight cars as a result of a collision with a piece of his equipment is intended to be subject to one application of the occurrence limit of the policy. Retention of the word ‘accident’ is limiting in this sense and no other.” (Elliott, The New Comprehensive General Liability Policy, in Liability Insurance Disputes, supra, at p. 12-5, italics added.)

Secretary Elliot’s comments leave little doubt that the definition of “occurrence” in the newly drafted standard form CGL policy was intended to provide coverage when damage or injury resulting from an accident or “injurious exposure to conditions” occurs during the policy period. The term “accident” was left in the definition of occurrence for the purpose of circumscribing the policy limits applicable to each occurrence. The drafters did not intend to require that an “accident” in the literal sense, e.g., a sudden precipitating event, occur during the policy period in order to trigger potential coverage for ensuing damage or injury. “The reference to ‘injurious exposure to conditions [resulting in] . . . bodily injury [or property damage]’ eliminates any requirement that the injury result from a sudden event. Although it is most common that an injury takes place simultaneously with the exposure, there are many instances of injuries taking place over an extended period of time before they become evident [for example, the slow ingestion of foreign substances or the inhalation of noxious fumes]. In these and similar cases, the definition of ‘occurrence’ identifies the time of loss for *673 purposes of applying coverage—the injury must take place during the policy period.” (3 Cal. Insurance Law & Practice, supra, Property and Liability Insurance, § 49.12, at pp. 49-20-49-21, fns. omitted.)

As these materials demonstrate, the drafters of the standard occurrence-based CGL policy, and the experts advising the industry regarding its interpretation when formulated in 1966, contemplated that the policy would afford liability coverage for all property damage or injury occurring during the policy period resulting from an accident, or from injurious exposure to conditions. Nothing in the policy language purports to exclude damage or injury of a continuous or progressively deteriorating nature, as long as it occurs during the policy period. Nor is there any basis for inferring that an insured’s understanding and reasonable expectations regarding the scope of coverage for damage or injury occasioned during the effective period of an occurrence-based CGL policy would have been otherwise. 14

We have shown how the clear and explicit language of Admiral’s policies supports the conclusion that potential coverage is triggered by the occurrence of bodily injury or property damage during the policy periods, as a result of an accident or the “continuous or repeated exposure to conditions.” We next review the relevant reported decisions, from California, the federal courts, and other state courts, that have sought to construe the industry-standardized CGL policy language to determine how continuous injury or damage triggers potential coverage under such policies. As will be seen, the weight of authority, consistent with our own interpretation of Admiral’s express policy language, is that bodily injury and property damage that is continuous or progressively deteriorating throughout successive CGL policy periods, is potentially covered by all policies in effect during those periods.

4. Survey of case law and authorities discussing triggering of coverage under CGL policies where injury or damage is continuous over successive policy periods.

The issue of trigger of coverage in continuous injury or damage cases has been explored by many courts. (See Annot. (1993) 14 A.L.R. 5th 695 .) Courts have recognized several “triggers” as a means of identifying the nature and timing of damage or injury that will give rise to liability coverage under an occurrence-based CGL policy. The courts have generally viewed the timing of damage or injury under occurrence-based CGL policies in four ways: at the date of exposure to the injurious or damage-causing event or conditions; at the date of the first occurrence of “injury in fact”; at the date *674 of manifestation or discovery of the damage or injury; and over the continuous period from exposure through manifestation and beyond, where the damage or injury is ongoing, continuous, or progressively deteriorating throughout a policy period or successive policy periods. At this point it will be helpful to briefly outline the various trigger theories formulated by the courts.

The exposure (or continuous exposure) trigger. This trigger of coverage theory, first applied in cases involving asbestos-related bodily injuries, focuses on the date on which the injury-producing agent first contacts the body. The exposure theory apportions the cost of indemnity among those insurers whose policies were in effect from that point in time onward. In effect, under this theory, damage or injury is deemed to commence from the first contact of the injury-producing agent with the injured party. The leading case espousing this trigger of coverage analysis is the Sixth Circuit’s decision in Ins. Co. North America v. Forty-Eight Insulations (6th Cir. 1980) 633 F.2d 1212 , clarified 657 F.2d 814 , cert. den. (1981) 454 U.S. 1109 [ 70 L.Ed.2d 650 , 102 S.Ct. 686 ] (Forty-Eight Insulations). The court in FortyEight Insulations found that the covered occurrence of injury commenced with the immediate contact of an asbestos fiber with the lungs, even though the progressive disease typically took some 20 years to develop. (633 F.2d at pp. 1215, 1218-1220.) The court reasoned that because of the cumulative and progressively deteriorating nature of the disease, it had to be distinguished from the ordinary accident or injury situation, and further, that because the injury is a continuing one, the insurers who furnished comprehensive general liability policies would expect the scope of their policies’ coverage to parallel the applicable theory of liability.

The manifestation (or manifestation of loss) trigger. This trigger of coverage, which, as already explained, was adopted by this court in the first party property insurance context in Prudential-LMI, supra, 51 Cal.3d 674 , holds the insurer insuring the property at the time appreciable property damage first becomes manifest solely responsible for indemnification to the insured. For purposes of applying the rule, the time at which the property damage becomes manifest (also the point of “inception of the loss”) is “that point in time when appreciable damage occurs and is or should be known to the insured, such that a reasonable insured would be aware that his notification duty under the policy had been triggered.” (Id. at p. 699.)

In Prudential-LMI, supra, 51 Cal.3d 674 , we identified three reasons supporting the application of the manifestation theory in the first party property insurance context. First, application of that trigger of coverage meets the reasonable expectations of the insureds who, in seeking to insure against perils to their property, would normally look to their present carrier *675 for coverage. (Id. at p. 699.) Second, “the underwriting practices of the insurer can be made predictable because the insurer is not liable for a loss once its contract with the insured ends unless the manifestation of loss occurred during its contract term.” (Ibid.) Third, since the insured is required under a standard first party property insurance policy to file suit against the insurer within 12 months after “inception of the loss,” and since inception of the loss is the date on which appreciable damage occurs and is or should be known to the insured, the definition of manifestation of loss and inception of the loss must be one and the same, that is to say, “that point in time when appreciable damage occurs and is or should be known to the insured, such that a reasonable insured would be aware that his notification duty under the policy had been triggered.” ( Prudential-LMI, supra, 51 Cal.3d at pp. 686-687, 699.) These policy reasons led us to conclude, “in conformity with the loss-in-progress rule, [that first party property] insurers whose policy terms commence after initial manifestation of the loss are not responsible for any potential claim . . . .” (Id. at p. 699.) 15

The continuous injury (or multiple) trigger. Under this trigger of coverage theory, bodily injuries and property damage that are continuous or progressively deteriorating throughout successive policy periods are covered by all policies in effect during those periods. The timing of the accident, event, or conditions causing the bodily injury or property damage, e.g., an insured’s negligent act, is largely immaterial to establishing coverage; it can occur before or during the policy period. Neither is the date of discovery of the damage or injury controlling: it might or might not be contemporaneous with the causal event. It is only the effect—the occurrence of bodily injury or property damage during the policy period, resulting from a sudden accidental event or the “continuous or repeated exposure to conditions”—that triggers potential liability coverage. The appellate cases in which this trigger of coverage was developed are discussed in greater detail below.

The injury-in-fact trigger. Under an injury-in-fact trigger, coverage is first triggered at that point in time at which an actual injury can be shown, *676 retrospectively, to have been first suffered. This rationale places the injury-in-fact somewhere between the exposure, which is considered the initiating cause of the disease or bodily injury, and the manifestation of symptoms, which, logically, is only possible when an injury already exists. (See Abex Corp. v. Maryland Cas. Co. (D.C. Cir. 1986) 790 F.2d 119 [ 252 App.D.C. 297 ] [asbestos]; American Home Products Corp. v. Liberty Mut. Ins. (2d Cir. 1984) 748 F.2d 760 [pharmaceuticals].) In the context of continuous or progressively deteriorating injuries, the injury-in-fact trigger, like the continuous injury trigger, affords coverage for continuing or progressive injuries occurring during successive policy periods subsequent to the established date of the initial injury-in-fact. However, the injury-in-fact trigger, unlike the exposure trigger, when applied in asbestos cases excludes from coverage the period from initial exposure to the date on which the injury-in-fact was first suffered. 16

*677 As already indicated, in the case before us, Montrose urges our adoption of a continuous injury trigger of coverage. Admiral in turn, in its briefs, urges us to apply a manifestation trigger of coverage. At oral argument, however, counsel for Admiral appeared to deviate from this position, arguing instead that an injury-in-fact trigger, and not a manifestation trigger, should be applied. 17 We shall give Admiral the benefit of the doubt and consider which, if any, of the recognized trigger of coverage theories should be applied here. The precise question, of course, is what result follows under the language of the policies of insurance to which the parties agreed, including the standardized definitions that were incorporated into those policies. As will be seen, most courts that have analyzed the issue have found the continuous injury trigger of coverage applicable to the standard occurrence-based CGL policy.

One of the first cases to apply a continuing injury theory of loss allocation in the context of progressive property damage was Gruol Construction Co. v. Insurance Co. of North America (1974) 11 Wn.App. 632 [ 524 P.2d 427 ] (Gruol). In that case, a contractor prevailed in an action against his insurer who had failed to defend him under his general liability policy in a third party construction defect suit for recovery of dry rot damage to a building. The contractor’s improper piling of dirt against the building had caused the dry rot. The court held that the injury was a continuous process which began *678 at the time of the negligent construction and continued through the manifestation of the dry rot damage, “ ‘even though there [was] a lapse of time between the initial negligent act and the occurrence of the ultimate damage (Id. at p. 636 [ 524 P.2d at p. 430 ].) Thus the holding of Gruol was that, when warranted by the facts, property damage should be deemed to occur over the entire process of the continuing injury. An insurer would become liable at any point in the process for the entire loss up to the policy limits, even though the continuing injury or progressively deteriorating damage may extend over several policy periods.

The first reported California case to discuss the triggering of potential coverage under third party liability insurance policies, where continuous or progressively deteriorating property damage was involved, was California Union Ins. Co. v. Landmark Ins. Co., supra, 145 Cal.App.3d 462 (California Union). That case involved a gradual leak of water from a swimming pool which caused damage to adjoining property. The parties stipulated that the pool began to leak in June 1979, and that a crack in the pool was the sole cause of the ensuing property damage. Damage to the adjoining property occurred between July 1979 and November 1980. Landmark Insurance Company (Landmark) was on the risk from July 1978 to July 1980. California Union Insurance Company (Cal Union) provided liability coverage from July 1980 to July 1981. (Id. at pp. 467-469.)

The source of the leakage damage in California Union was discovered during an inspection of the pool in October 1980, at a point in time following expiration of the Landmark policy and during the term of the successive Cal Union policy. At trial, the two carriers contested liability for the damage that occurred between October 1980 (discovery of the leak) and November 1980 (repair of the source of the damage). Landmark had undertaken repairs prior to the expiration of its policy in July 1980, but apparently repaired only the damage to the slopes of the adjoining property, and not the as-yet undiscovered source of the damage: the leaking pool. Landmark contended that the postdiscovery damage (that which occurred after October 1980) constituted a separate occurrence within the definition of that term in the Cal Union policy, and was therefore Cal Union’s sole responsibility. Cal Union in turn argued the damage was a continuation of a single occurrence that began during the period of coverage provided by the Landmark policies, and was thus the sole responsibility of Landmark. (California Union, supra, 145 Cal.App.3d at p. 468 .)

The trial court held that each manifestation of damage should be treated as a separate occurrence under the policies, rejecting Cal Union’s position that separate incidents of manifestation of damage which are attributable to the *679 same underlying cause are merely manifestations of the same continuous “occurrence” of damage. (California Union, supra, 145 Cal.App.3d at p. 469 .) The Court of Appeal reversed, concluding that the trial court’s ruling contravened the express language of each insurer’s policies. (Ibid.)

On appeal, both insurers in California Union readily acknowledged that under the rule of Remmer, supra, 140 Cal.App.2d 84 (ante, at pp. 669-670), a coverable “occurrence” arose under their policy language at the point at which the complaining party was actually damaged, not the time at which the initial damage-causing act or conditions transpired. (California Union, supra, 145 Cal.App.3d at p. 470 .) The California Union court agreed, pointing out that the precise facts of Remmer were distinguishable from those in the case before it. Observing that the dangerous condition in Remmer (a defectively graded lot) had failed to manifest any damage for a period of five years, the California Union court noted that in the case before it the leaking pool was a “continuous active force at work” during the eighteen months between the time of the “wrongful act” (the crack in the pool that first gave rise to the water damage to the adjoining property) and the manifestation of the actual loss. (Id. at p. 473.) Focusing on the identical “one occurrence” language in Cal Union’s and Landmark’s CGL policies (“ ‘all . . . property damage arising out of continuous or repeated exposure to substantially the same general conditions shall be considered as arising out of one occurrence’ ”), the California Union court concluded that, given the continuing and progressively deteriorating nature of the pool leakage damage, the trial court’s determination that each manifestation of damage was a separate occurrence conflicted with the “one occurrence” policy language in each insurer’s policies. (Id. at p. 469.)

The California Union court next surveyed several California appellate decisions which, up to that time, had attempted to set, for definitional purposes, the timing of occurrences of damage or injury transpiring prior to, during, or after the effective periods of successive third party liability insurance policies. (California Union, supra, 145 Cal.App.3d at pp. 471-472, 474, and cases cited.) Although each such case “[w]ithout exception . . . involved delays in varying periods of time between the wrongful act and [manifestation of] the actual loss,” the California Union court observed that none had involved actual continuous or progressively deteriorating damage or injury. (Id. at p. 473.) The court also took note of the settled authorities holding that an insurer’s obligation to indemnify an insured for manifested *680 losses may continue even after the term of the policy expires. (Ibid.) 18 Even in the third party liability insurance context, an insurer’s liability for a still insured and continuing event is not terminated by the expiration of the policy term. (California Union, supra, 145 Cal.App.3d at p. 475 ; accord, Harman v. American Casualty Co. of Reading, Pa. (C.D.Cal. 1957) 155 F.Supp. 612 .) As stated in California Union: “[I]n a ‘one occurrence’ case involving continuous, progressive and deteriorating damage, the carrier in whose policy period the damage first becomes apparent remains on the risk until the damage is finally and totally complete, notwithstanding a policy provision which purports to limit the coverage solely to those accidents/occurrences within the time parameters of the stated policy term.” (California Union, supra, 145 Cal.App.3d at p. 476 .)

Having found under settled principles of law that insurer Landmark remained obligated to indemnify the insured for the pool leakage damage which commenced prior to, but continued to progressively deteriorate after, expiration of Landmark’s policy, the California Union court then turned to the unsettled question of whether the successive insurer, Cal Union, was also on the risk for the damage occurring during its successive policy period. Although it was true that the force producing the continuing pool leakage was already set in motion when Cal Union came on the risk with the initiation of its successive policy, that damage-causing force continued into the period of Cal Union’s policy, further damage occurred during that policy *681 period, and substantial corrective procedures were necessary and performed after the October-November 1980 damage manifested itself during the period of Cal Union’s policy. (California Union, supra, 145 Cal.App.3d at p. 476 .)

The California Union court concluded Cal Union was on the risk to indemnify the insured for the continuous property damage occurring through its successive policy period. 19 The court placed primary reliance on three cases. The first was Gruol, supra, 11 Wn.App. 632 [ 524 P.2d 427 ]. Gruol , as noted above (ante, at pp. 677-678), held that progressive property damage should be deemed to occur over the entire process of the continuing injury, with a CGL carrier liable at any point in the process for the entire loss up to the policy limits, even though the continuous or progressively deteriorating damage extends over successive policy periods. ( 11 Wn.App. at p. 636 [524 P.2d at pp. 430-431].)

The second case relied on by the California Union court was the Sixth Circuit Court of Appeals’ decision in Forty-Eight Insulations, supra, 633 F.2d 1212 . As noted above, Forty-Eight Insulations was the leading case on the exposure theory of coverage, holding that due to the continuing and cumulative nature of asbestos-related diseases, insurers providing CGL coverage commencing at the time of the worker’s initial exposure to asbestos particles would be held potentially liable to defend and indemnify the insured manufacturer of asbestos products in underlying third party actions alleging bodily injury claims against the insured. (Ante, at p. 674.) Recognizing that Forty-Eight Insulations was an asbestos products liability case, the California Union court nonetheless concluded that because the Sixth Circuit Court of Appeals’ decision involved a CGL policy covering “single accident/occurrence, continuing damage claims,” the basic rationale of that decision was applicable to the continuous injury trigger analysis being invoked for the ongoing property damage at issue in California Union. (California Union, supra, 145 Cal.App.3d at p. 478 .)

*682 The third case relied on by the California Union court was Keene, supra, 667 F.2d 1034 . 20 Like Forty-Eight Insulations, Keene was a products liability case in which the manufacturer of insulation products containing asbestos brought an action for a declaratory judgment seeking a determination of the obligations of four liability insurance carriers to defend and indemnify it in pending products liability litigation. Holding that the “occurrence” which caused the bodily injury took place substantially before the manifestation of the ultimate injury (asbestosis), the District of Columbia Circuit Court of Appeals found each insurer on the risk between the initial exposure and the manifestation of disease to be potentially liable for indemnification and defense costs. (667 F.2d at pp. 1046-1047; California Union, supra, 145 Cal.App.3d at p. 478 .)

The Keene court based its rationale primarily on the expectations of the parties and the ambiguities it perceived as inherent in the standard CGL policy language. Applying the presumption requiring ambiguities to be construed in favor of the insured, the Keene court reasoned that Keene Corporation (the insured) could have reasonably expected that it was covered for future liabilities: “A latent injury, unknown and unknowable to Keene at the time it purchased [liability] insurance, must, at least, be covered by an insurer on the risk at the time it manifests itself.” (Keene, supra, 667 F.2d at p. 1044 .) In the context of a progressive disease like asbestosis, where the medical evidence indicates that the disease can develop or manifest as late as 20 or more years after exposure, the continuous injury trigger of potential coverage adopted in Keene , consistent with the expectations of the insured, fixes the timing of the injury at the point of initial exposure of the injured third party to the injury-causing agent, at the time of manifestation of symptoms of bodily injury, and during the development and progression of the disease in between those points in time. The Keene court also broadly interpreted bodily injury to mean “any part of the single injurious process that asbestos-related diseases entail.” (Id. at p. 1047.)

Fireman’s Fund Ins. Co. v. Aetna Casualty & Surety Co., supra, 223 Cal.App.3d 1621 (Fireman’s Fund) was the next California appellate decision postdating California Union to directly address the question of coverage for continuous property damage or losses under successive third party CGL *683 policies. In Fireman’s Fund, two liability carriers insured a contractor that had undertaken the restoration of the exterior facade of a hotel. The first carrier (Fireman’s Fund) was on the risk when construction defects (spalling and cracking of the restored facade) were first discovered; the second carrier (Aetna) was on the risk when the defects progressed and when their cause became known. (Fireman’s Fund, supra, 223 Cal.App.3d at p. 1623 .) On cross-motions for summary judgment, based upon stipulated facts and purporting to rely on the rationale of Home Ins. Co. v. Landmark Ins. Co. (1988) 205 Cal.App.3d 1388 [ 253 Cal.Rptr. 277 ] (Home) (a first party property insurance case holding that only the first of two successive insurers, the carrier on the risk on the date of first manifestation of the property damage, was liable for the entire claim), the trial court determined Fireman’s Fund was solely responsible to indemnify the contractor for an arbitration award returned against it. The Fourth District Court of Appeal (the same court that decided Home) affirmed the trial court’s order, rejecting Fireman’s Fund’s argument that the analysis of Home was inapposite because that case involved first party property insurance coverage, not third party general liability insurance. (Fireman’s Fund, supra, 223 Cal.App.3d at pp. 1623-1624.)

Although the Fireman’s Fund court was construing standardized third party CGL policies, the court refused to apply the continuous injury trigger of coverage analysis adopted for third party liability insurance policies in California Union, supra, 145 Cal.App.3d 462 . The Fireman’s Fund court observed that it had already considered and rejected the reasoning of California Union in Home, supra, 205 Cal.App.3d 1388 , and opined that coverage under successive third party CGL policies, in essence, should require no different analysis than that which was applied in the first party property insurance context in Home . In short, the Fireman’s Fund court applied the manifestation trigger of coverage which it had earlier adopted in Home. (Fireman’s Fund, supra, 223 Cal.App.3d at pp. 1626-1627.) The court indicated that “[t]o the extent Home’s rationale rests on the loss-in-progress rule, it, too, is fully applicable to a third party claim.” (Id. at p. 1627, fn. omitted.) And the Fireman’s Fund court reasoned further that, “[l]ike the situation in Home , here the issues arise in a context where the claimant [i.e., the insured] has been fully satisfied and the case involves allocating loss between insurers. Home is, therefore, dispositive. Contrary to Fireman’s Fund’s contention, Garvey v. State Farm Fire & Casualty Co., supra, 48 Cal.3d 395 does not change the result.” (Id. at p. 1628.)

The Fireman’s Fund court made clear in its opinion that our admonishments in Garvey , respecting the differences between first party property and third party liability insurance, had little impact on that court’s determination *684 to apply its earlier trigger of coverage analysis in Home to the third party liability insurance case before it. The Fireman’s Fund court suggested: “In Garvey , the Supreme Court held it is ‘important to separate the causation analysis necessary in a first party property loss case from that which must be undertaken in a third party tort liability case.’ (Garvey v. State Farm Fire & Casualty Co., supra, 48 Cal.3d at p. 406 , italics added.) However, although there are important differences between property damage insurance and liability insurance—not the least of which is causation analysis—the issues here do not even remotely involve causation. Garvey neither holds nor suggests that all legal principles developed in first party cases are inapplicable in third party cases. Thus, even if Home’s rationale was solely based upon first party insurance provisions (which it is not), Garvey does not prohibit its application to liability coverage.” (Fireman’s Fund, supra, 223 Cal.App.3d at p. 1628 , italics in original.)

The Fireman’s Fund court failed to engage in any meaningful discussion of what factors set first party property insurance policies apart from third party comprehensive liability insurance policies. Nor did the court set forth in its opinion, or make any attempt to analyze, the standard definition of “occurrence” found in the standard form CGL policy. 21 Finally, apparently satisfied with its earlier observation in Home that, “[b]y its terms, section 22 [the codified loss-in-progress rule] applies to both first-party [property insurance] and third-party [liability insurance] cases” (Home, supra, 205 Cal.App.3d at p. 1395, fn. 4 ), the Fireman’s Fund court made no further effort to analyze how application of the loss-in-progress rule might differ in the third party liability insurance context.

Only one reported California decision followed Fireman's Fund in holding the manifestation trigger of coverage applicable in cases of continuous or progressively deteriorating damage or injury under successive third party CGL policies. In Pines of La Jolla Homeowners Assn. v. Industrial Indemnity (1992) 5 Cal.App.4th 714 [ 7 Cal.Rptr.2d 53 ], the Fourth District Court of Appeal once again, relying on its earlier holding in Fireman’s Fund, supra, 223 Cal.App.3d 1621 , concluded that a manifestation theory should be applied in determining the trigger of potential coverage applicable under several CGL policies for continuous property damage resulting from construction defects. (5 Cal.App.4th at pp. 721-722.)

*685 Most recently, the Fourth District Court of Appeal decided Zurich Ins. Co. v. Transamerica Ins. Co. (Cal.App.) (Zurich). Zurich involved a declaratory relief action, brought by one of four liability insurers who provided successive periods of coverage to a construction company, to determine the respective defense and indemnity obligations of each insurer with regard to three underlying construction defect actions against the company pertaining to a condominium project.

The Fourth District Court of Appeal in Zurich repudiated the rationale of its earlier decision in Fireman’s Fund, supra, 223 Cal.App.3d 1621 . The court acknowledged that “[t]he manifestation rule developed in the first party context is not appropriately applied across the board,” and that “[i]n-stead, in this [third party] liability context, a ‘continuing injury’ trigger should be used, because property damage occurred . . . and continued . . . [over a period of several years].” The court “[n]oted that the manifestation rule presupposes that the first party insured will be on site to observe the damage; it is in the nature of a discovery rule. However, third party liability policies do not contain in their occurrence sections any discovery requirement or a policy limitations period for the filing of an action after manifestation of a defect.” The Zurich court also retreated from the “rather narrow definition of contingency” espoused in Fireman’s Fund in connection with that opinion’s discussion of the loss-in-progress rule, and reached new conclusions regarding the applicability of that rule in third party liability insurance cases, consistent with the conclusions we reach today respecting the applicability of that rule to this case. (Post, at pp. 689-692.)

Accordingly, to the extent the decisions in Fireman’s Fund, supra, 223 Cal.App.3d 1621 , and Pines of La Jolla Homeowners Assn. v. Industrial Indemnity, supra, 5 Cal.App.4th 714 , are inconsistent with the principles discussed herein, those decisions are hereby disapproved.

5. Various practical and policy considerations further support adoption of the continuous injury trigger of coverage for the third party claims of continuous or progressively deteriorating damage or injury brought under the CGL policies in this case.

Our foregoing review of the standard CGL policy language, as incorporated into Admiral’s policies, as well as the relevant cases and authorities that have construed that language, lead us to conclude that the continuous injury trigger of coverage should be adopted for claims of continuous or progressively deteriorating damage or injury under the third party CGL policies in issue in this case.

*686 We have shown why Admiral’s express policy language supports application of the continuous injury trigger of coverage. We have explained that, contrary to Admiral’s arguments in its briefs, it has long been understood that the standard form CGL policy provides liability coverage for damage or injury occurring during the policy period which results from an accident, or from continuous or repeated exposure to injurious conditions. There is no requirement that the sudden, accidental damage-causing act or event, or the conditions giving rise to the damage or injury, themselves occur within the policy period in order for potential liability coverage to arise. We have also explained how retention of the term “accident” within the standard definition of occurrence in the “occurrence-based” policies drafted after 1966 was intended to serve the one-occurrence rule, and was never intended to impose a requirement that the damage-causing accident, event, or conditions occur within the policy period. We have noted the settled rule that an insurer on the risk when continuous or progressively deteriorating damage or injury first manifests itself remains obligated to indemnify the insured for the entirety of the ensuing damage or injury. And we have reviewed the rationale of California Union, supra, 145 Cal.App.3d 462 , and the decisions cited and relied on therein, which, together with the weight of more recent authorities, 22 conclude that where successive CGL policies have been *687 purchased, bodily injury and property damage that is continuing or progressively deteriorating throughout more than one policy period is potentially covered by all policies in effect during those periods.

Lastly, we have explained how first party insurance differs from third party liability insurance in many fundamental respects, and why the rationale of our holding in Prudential-LMI, supra, 51 Cal.3d 674 , adopting the manifestation trigger of coverage for first party cases, would be inapposite if applied in the context of third party liability insurance coverage.

Our conclusion that the continuous injury trigger of coverage should be applied to the third party CGL policies in this case is also in conformity with several important policy considerations. In Prudential-LMI, supra, 51 Cal.3d at page 699, we observed, as one policy reason favoring adoption of the manifestation trigger of coverage in first party property insurance cases, that “the underwriting practices of the insurer can be made predictable because the insurer is not liable for a loss once its contract with the insured ends unless the manifestation of loss occurred during its contract term.” Admiral here suggests that the general policy favoring the predictability of underwriting practices and reserves will be negatively affected by adoption of a continuous injury trigger in the third party CGL insurance context. We disagree. A number of factors undercut Admiral’s concerns.

First, leaving aside the availability of excess (multiple) policies or “other insurance” clauses, and absent express policy language decreeing the manner of apportionment of contribution among successive liability insurers, the courts will generally apply equitable considerations to spread the cost among the several policies and insurers. (See, e.g., CNA Casualty of California v. Seaboard Surety Co. (1986) 176 Cal.App.3d 598, 619-620 [ 222 Cal.Rptr. 276 ]; Olympic Ins. Co. v. Employers Surplus Lines Ins. Co. (1981) 126 Cal.App.3d 593, 601-602 [ 178 Cal.Rptr. 908 ].)

Second, in establishing reserves for the standard form occurrence-based CGL policies which replaced accident-based policies in 1966, the insurance industry, as we have shown, was fully aware of the intended scope of *688 coverage of the new policies, coupled with the specific provision providing coverage for continuous or repeated exposure to conditions causing property damage or bodily injury. Indeed, the drafting history of the standard occurrence-based CGL policy reflects that not only did the drafters understand the term occurrence to mean an accident or exposure to injurious conditions resulting in the occurrence of damage or injury during the policy period, they specifically considered and rejected the suggestion that language establishing a manifestation or discovery trigger of coverage be incorporated into the standard form CGL policy. Among the reasons relied on for rejecting the incorporation of such limitations into the standard definitions in the coverage clauses were several stated equitable concerns: the difficulty of applying such limitations or requirements in cases of continuing damage or injury over the course of successive policy periods, the uncertainty of who would bear the burden of a discovery requirement (i.e., the insured or third party claimants), the arbitrariness, from the carrier’s perspective, of telescoping all damage in a continuing injury case into a single policy period, and the fear that policyholders could be disadvantaged by such an approach. (See American Home Prod. v. Liberty Mut. Ins. Co., supra, 565 F.Supp. 1485, 1501-1502 , affd. as mod., supra, 748 F.2d 760 [surveying joint committee hearings and drafting materials].) In short, the insurance industry is on record as itself having identified several sound policy considerations favoring adoption of a continuous injury trigger of coverage in the third party liability insurance context. 23

Finally, we agree with Montrose that application of a manifestation trigger of coverage to an occurrence-based CGL policy would unduly transform it into a “claims made” policy. Claims made policies were specifically developed to limit an insurer’s risk by restricting coverage to the single policy in effect at the time a claim was asserted against the insured, without regard to the timing of the damage or injury, thus permitting the carrier to establish reserves without regard to possibilities of inflation, upward-spiraling jury awards, or enlargments of tort liability after the policy period. 24 The insurance industry’s introduction of “claims made” policies into the area of *689 comprehensive liability insurance itself attests to the industry’s understanding that the standard occurrence-based CGL policy provides coverage for injury or damage that may not be discovered or manifested until after expiration of the policy period. That understanding is clearly reflected in the higher premiums that must be paid for occurrence-based coverage to offset the increased exposure. (Pacific Employers Ins. Co. v. Superior Court, supra, 221 Cal.App.3d at pp. 1359-1360.) We agree with the conclusion of the Court of Appeal below that to apply a manifestation trigger of coverage to Admiral’s occurrence-based CGL policies would be to effectively rewrite Admiral’s contracts of insurance with Montrose, transforming the broader and more expensive occurrence-based CGL policy into a claims made policy. (Accord, Harford County v. Harford Mut. Ins., supra, 610 A.2d at pp. 294-295.)

We therefore conclude that the continuous injury trigger of coverage should be applied to the underlying third party claims of continuous or progressively deteriorating damage or injury alleged to have occurred during Admiral’s policy periods. Where, as here, successive CGL policy periods are implicated, bodily injury and property damage which is continuous or progressively deteriorating throughout several policy periods is potentially covered by all policies in effect during those periods.

III

The Loss-in-Progress Rule

Relying on the loss-in-progress rule (sometimes also referred to as the known loss rule), Admiral contends there was no potential liability coverage for, and consequently no duty to defend Montrose, in the String-fellow cases. We disagree.

Section 22 defines “insurance” as a “contract whereby one undertakes to indemnify another against loss, damage, or liability arising from a *690 contingent or unknown event.” (Italics added.) Section 250 provides that “any contingent or unknown event, whether past or future, which may damnify a person having an insurable interest, or. create a liability against him, may be insured against, subject to the provisions of this code.” (Italics added.) Accordingly, when a loss is “known or apparent” before a policy of insurance is issued, there is no coverage. (Prudential-LMI, supra, 51 Cal.3d at p. 695, & fn. 7.)

Critically, the requirement that the “event” be “unknown” or “contingent” is stated in the disjunctive in the rules embodied in sections 22 and 250. We long ago recognized that all that is required to establish an insurable risk is that there be some contingency. Even where subsequent damage might be deemed inevitable, “ ‘such “inevitability” does not alter the fact that at the time the contract of insurance was entered into, the event was only a contingency or risk that might or might not occur within the term of the policy.’ ” (Sabella v. Wisler (1963) 59 Cal.2d 21, 34 [ 27 Cal.Rptr. 689 , 377 P.2d 889 ], italics in original (Sabella).)

In Sabella , the insurer claimed that damage to the insured’s residence was not fortuitous and thus not covered because “the damage occurred as a result of the operation of forces inherent” in the underlying soil conditions (including uncompacted fill and defective workmanship in the installation of a sewer outflow that ultimately broke). Sabella rejected the insurer’s contention that the loss was “not fortuitous and hence not a ‘risk’ properly the subject of insurance.” ( 59 Cal.2d at p. 34 .) Relying on Snapp, supra, 206 Cal.App.2d 827 , Sabella held that even if it were inevitable that the damage would have occurred at some time during ownership of the house, the loss was covered because such loss was a contingency or risk at the time the parties entered into the policy. (Sabella, supra, 59 Cal.2d at p. 34 ; see also Prudential-LMI, supra, 51 Cal.3d at p. 696.)

According to Admiral, Montrose’s knowledge of the problems at the Stringfellow site defeats coverage. In particular, Admiral points to the fact of Montrose’s receipt of the PRP letter from the EPA on August 31, 1982, prior to the inception of the first of Admiral’s four successive CGL policies issued to Montrose. Admiral misses the point. The PRP notice is just what its name suggests—notice that the EPA considered Montrose a “potentially” responsible party. While it may be true that an action to recover cleanup costs was inevitable as of that date, Montrose’s liability in that action was not a certainty. There was still a contingency, and the fact that Montrose knew it was more probable than not that it would be sued (successfully or otherwise) is not enough to defeat the potential of coverage (and, consequently, the duty to defend). (Sabella, supra, 59 Cal.2d at p. 34 .)

*691 Moreover, since Admiral’s policies did not purport to cover damage or injury that occurred prior to the time those policies went into effect, and only covered those bodily injuries and damages (or continuing bodily injuries and damages resulting from “continuous or repeated exposure to conditions”) that might occur in the future during the policy periods, the existence and extent of such prospective injuries were clearly unknown and contingent, from Montrose’s standpoint, at the time Montrose first purchased its policies from Admiral.

Courts which have addressed the loss-in-progress issue have recognized that “[t]he point at which a threat of loss is so immediate that it may fairly be said that the loss was in progress and the insured knew of it at the time the policy was applied for or issued is generally a question of fact.” (Sentinel Ins. Co. v. First Ins. Co., supra, 875 P.2d at p. 920; Inland Waters Pollution Control v. National Union (6th Cir. 1993) 997 F.2d 172, 178 .) Indeed, several courts have observed that, in the context of third party CGL coverage, a loss-in-progress or known loss contention can seldom be successfully relied on by an insurer to defeat a duty to defend because the factual uncertainties needed to be resolved in order to establish the defense generally cannot be resolved on a motion for summary judgment (see Nestle Foods Corp. v. Aetna Casualty and Sur. Co. (D.N.J. 1993) 842 F.Supp. 125, 130-131 ), or until the insurer conclusively establishes no possibility of coverage (see American Bumper & Mfg. Co. v. Hartford Fire Ins. Co. (1994) 207 Mich.App. 60 [ 523 N.W.2d 841, 845-846 ]). (See also Maryland Casualty Co. v. Wausau Chemical Corp. (W.D.Wis. 1992) 809 F.Supp. 680, 697-698 [insurers’ contention that “principle of fortuity” barred coverage due to insured’s knowledge of “completed” or “highly likely” property damage before purchase of insurance policies precluded on summary judgment motion given existence of “disputed facts”].)

In the Court of Appeal, Admiral relied on Advanced Micro Devices, Inc. v. Great American Surplus Lines Ins. Co. (1988) 199 Cal.App.3d 791 [ 245 Cal.Rptr. 44 ] in support of its contention that if the insured knows or should have known that there was a problem, the loss is known and there is no insurable risk. Advanced Micro Devices, Inc., is inapposite. That case involved interpretation of an express exclusionary clause disavowing coverage of losses arising from “ ‘any known pre-existing conditions.’ ” (Id. at p. 794.) Although Admiral’s policies define “occurrence” as an accident resulting in bodily injury or property damage “neither expected nor intended from the standpoint of the insured,” this language is part of the coverage clauses and not an express exclusionary clause as was the policy provision at issue in Advanced Micro Devices, Inc. As an integral part of the occurrence-based *692 coverage clause, we must interpret it broadly, in order to protect the objectively reasonable expectations of the insured (Garvey, supra, 48 Cal.3d at p. 406 ), and consistent with the rule announced by this court in Sabella .

Although it is true that the loss-in-progress rule as codified in sections 22 and 250 draws no distinction between, and thus is applicable to, first party property insurance and third party liability insurance policies, the distinctions inherent in the two types of coverage necessarily result in a different analysis when the rule is applied in the liability insurance context. As we have explained, first party property insurance policies provide coverage for damage to the insured’s own property. In that context, insurance cannot be obtained for damage which has already occurred because the absence of risk precludes coverage. (§§ 22, 250.) Third party liability insurance policies, in contrast, afford coverage for “sums which the insured shall become legally obligated to pay as damages because of bodily injury or property damage.” In the liability insurance context, insurance cannot be obtained for a “known liability.”

Where, as here, there is uncertainty about the imposition of liability and no “legal obligation to pay” yet established, there is an insurable risk for which coverage may be sought under a third party policy. (Austero v. National Cas. Co. (1978) 84 Cal.App.3d 1, 27-28, 29 [ 148 Cal.Rptr. 653 ] [“The fundamental contractual duty of the insurer in the third party case is to pay such judgments as shall be recovered against the insured . . . .” “In the usual first party case the promise of the insurer is to pay money, due under the policy, to the insured upon the happening of the event, the risk of which has been insured against.” (Original italics deleted, new italics added.)]; Keene, supra, 667 F.2d at p. 1041 [in a CGL policy, “the uncertain loss is the possibility of incurring legal liability”]; Aetna Cas. & Sur. Co. v. Condict (S.D.Miss. 1976) 417 F.Supp. 63, 73 .)

The United States Court of Appeals, Second Circuit’s decision in City of Johnstown, N.Y. v. Bankers Standard Ins. (2d Cir. 1989) 877 F.2d 1146 , is particularly instructive given its analogous facts. In that case, prior to commencement of the liability policy term in issue, the city (the insured) was aware that releases from its dumpsite may have polluted surrounding groundwater, had been notified by the EPA that the “landfill posed a ‘potential environmental hazard,’ ” and had been sued by a family “whose well had reportedly been contaminated by the landfill.” (Id. at pp. 1147, 1151-1152.) The Second Circuit Court of Appeals held in City of Johnstown, N.Y., that because future injury and resulting liability to third parties *693 for damages remained contingent, a loss-in-progress rule would not bar coverage. 25

We therefore hold that, in the context of continuous or progressively deteriorating property damage or bodily injury insurable under a third party CGL policy, as long as there remains uncertainty about damage or injury that may occur during the policy period and the imposition of liability upon the insured, and no legal obligation to pay third party claims has been established, there is a potentially insurable risk within the meaning of sections 22 and 250 for which coverage may be sought. Stated differently, the loss-in-progress rule will not defeat coverage for a claimed loss where it had yet to be established, at the time the insurer entered into the contract of insurance with the policyholder, that the insured had a legal obligation to pay damages to a third party in connection with a loss.

Montrose’s receipt of the PRP letter prior to its purchase of Admiral’s policies did not establish any legal obligation to pay damages or cleanup costs in connection with the contamination at the Stringfellow site, such as would implicate the loss-in-progress rule and preclude Montrose from seeking to obtain the liability coverage sought. The PRP letter did no more than formally place Montrose on notice of the government’s asserted position and initiate proceedings that could result in subsequent findings and orders. (See Spangler Const, v. Indus. Crankshaft (1990) 326 N.C. 133 [ 388 S.E.2d 557, 559 ].) Moreover, the PRP letter referred only to the CERCLA cleanup of the Stringfellow site—it did not refer in any way to the injuries, wrongful deaths, or property damage alleged to have occurred off-site by the plaintiffs in Newman v. Stringfellow.

Accordingly, we conclude that, at least on the facts heretofore alleged in this declaratory relief action, the loss-in-progress rule does not bar potential coverage, or relieve Admiral of its duty to defend, under the policies issued by Admiral to Montrose.

IV

Conclusion

Although we have determined that the continuous injury trigger of coverage should be applied in this case, and that the loss-in-progress rule does not *694 serve to defeat the potential for coverage or Admiral’s duty to defend, we hasten to add that resolution of these issues in Montrose’s favor would appear not to mark the end of the coverage-related inquiries in this complex litigation.

We do not herein purport to reach the merits of whether coverage under Admiral’s policies for the injury and damage alleged in the five underlying lawsuits against Montrose can ultimately be established. (See ante, at p. 655, fn. 2.) Whether the damages and injuries alleged were in fact “continuous” is itself a matter for final determination by the trier of fact. (See, e.g., Carey v. Canada, Inc. v. California Union Ins. Co. (D.D.C. 1990) 748 F.Supp. 8 ; Triangle Publications v. Liberty Mutual Ins. Co. (E.D.Pa. 1989) 703 F.Supp. 367, 371 .) Nor do we determine the effect, if any, of any exclusions contained in Admiral’s policies on the duty to defend or the ultimate issue of coverage, or reach the merits of any affirmative defenses to coverage that might be available to Admiral.

As conceded by Montrose, factual questions remain surrounding the circumstance of Montrose’s receipt of the PRP letter and its alleged failure to advise Admiral of the same. An insured must make all required disclosures at the time it applies for coverage; the fact that the loss-in-progress rule does not defeat coverage does not itself obviate the possibility of a finding of fraudulent concealment. (§ 331 [“Concealment, whether intentional or unintentional, entitles the injured party to rescind insurance”]; §§ 332-337; § 338 [“An intentional and fraudulent omission, on the part of one insured, to communicate information of matters proving or tending to prove the falsity of a warranty, entitles the insurer to rescind”].) We do not express any view concerning what, if anything, ought to have been disclosed by Montrose to Admiral at the time of purchase of the initial policy, and thereafter upon each renewal, nor do we consider the validity or effect, if any, of the pollution exclusion provision contained in the last of Admiral’s four policies issued to Montrose. In short, we leave all necessary factual findings, and the ultimate question of the existence of coverage, for determination in the appropriate forum below. (Prudential-LMI, supra, 51 Cal.3d at p. 680, fn. 3; Garvey, supra, 48 Cal.3d at p. 406 .)

The judgment of the Court of Appeal is affirmed, and the matter remanded for further proceedings consistent with the views expressed herein.

Mosk, J., Kennard, J., Arabian, J., George, J., and Werdegar, J., concurred.

Throughout this opinion, any reference to “successive” policies is intended to also include policies or policy periods which are temporally separated from one another by gaps or lapses in the coverage periods.

Throughout this opinion, we will refer to the term “trigger of coverage.” In the third party liability insurance context, “trigger of coverage” has been used by insureds and insurers alike to denote the circumstances that activate the insurer’s defense and indemnity obligations under the policy. The term “trigger of coverage” should not be misunderstood as a doctrine to be automatically invoked by a court to conclusively establish coverage in certain categories of cases, or under certain types of policies. The word “trigger” is not found in the CGL policies themselves, nor does the Insurance Code enumerate or define “trigger of coverage.” Instead, “trigger of coverage” is a term of convenience used to describe that which, under the specific terms of an insurance policy, must happen in the policy period in order for the potential of coverage to arise. The issue is largely one of timing—what must take place within the policy’s effective dates for the potential of coverage to be “triggered”? Whether coverage is ultimately established in any given case may depend on the consideration of many additional factors, including the existence of express conditions or exclusions in the particular contract of insurance under scrutiny, the availability of certain defenses that might defeat coverage, and a determination of whether the facts of the case will support a finding of coverage.

All further statutory references are to the Insurance Code unless otherwise indicated.

The other CGL carriers and dates of coverage are: Insurance Company of North America (Jan. 1, 1960, to Jan. 1, 1969, and Jan. 15, 1981, to Jan. 15, 1986); American Motorists Insurance Company (Jan. 1, 1969, to Mar. 1, 1971); the Travelers Indemnity Company (Mar. 1, 1971, to July 1, 1977); National Union Fire Insurance Company (July 1, 1977, to Jan. 15, 1981); Canadian Universal Insurance Company, Ltd. (Mar. 20, 1980, to Mar. 20, 1982); and Centaur Insurance Company (Mar. 20, 1982, to October 13, 1982).

When it is necessary to distinguish between these two actions involving the Stringfellow site, we will refer to them as U.S. v. Stringfellow and Newman v. Stringfellow. References to the “Stringfellow cases” are intended to apply to both actions.

Progressive property damage, or progressively deteriorating damage, are terms that refer to damage that occurs over an extended period of time, often during the effective periods of several successive insurance policies. In the property damage context, “progressive” or “progressively deteriorating” damage typically might involve continuing damage caused by, or resulting from, natural causes such as soil subsidence or dry rot, or man-made causes such as the disposal of industrial pollutants or toxic wastes that leach through or onto property adjoining the insured’s land, or into the underlying water table.

For reasons not clear from the record, sometime prior to October 13, 1982, Stauffer Chemical Company, which at the time owned 50 percent of the stock of Montrose, notified all of Montrose’s CGL carriers except Admiral of the PRP letter. Montrose first advised Admiral about the Stringfellow allegations at the time Montrose submitted its application for a renewed policy of insurance dated February 15, 1985. Of course it is also true that Admiral thereafter renewed the CGL policy for 1985-1986.

We shall refer to these cases collectively as the Levin Metals cases.

It must be borne in mind that Admiral’s duty to defend Montrose is all that is directly at issue in this proceeding. The obligation to indemnify must be distinguished from the duty to defend. The duty to defend arises when there is a potential for indemnity. (See post, at pp. 662-663; Horace Mann Ins. Co. v. Barbara B. (1993) 4 Cal.4th 1076, 1081 [ 17 Cal.Rptr.2d 210 , 846 P.2d 792 ]; Gray v. Zurich Insurance Co. (1966) 65 Cal.2d 263, 276 [ 54 Cal.Rptr. 104 , 419 P.2d 168 ].) It may exist even when coverage is in doubt and ultimately is not established. (Horace Mann Ins. Co. v. Barbara B., supra, 4 Cal.4th at p. 1081; Saylin v. California Ins. Guarantee Assn. (1986) 179 Cal.App.3d 256, 263 [ 224 Cal.Rptr. 493 ].) The obligation to indemnify, on the other hand, arises when the insured’s underlying liability is established. (Civ. Code, § 2778, subd. 1; Clark v. Bellefonte Ins. Co. (1980) 113 Cal.App.3d 326, 336-337 [ 169 Cal.Rptr. 832 ].) Although an insurer may have a duty to defend, it ultimately may have no obligation to indemnify, either because no damages were awarded in the underlying action against the insured, or because the actual judgment was for damages not covered under the policy. (See City of Laguna Beach v. Mead Reinsurance Corp. (1990) 226 Cal.App.3d 822, 830 [ 276 Cal.Rptr. 438 ].) Moreover, in a declaratory relief action held before the insured’s liability to third party claimants has been established, the trial court will be unable to determine the amount of the insurer’s indemnity obligation. (Aitchison v. Founders Ins. Co. (1958) 166 Cal.App.2d 432, 439 [ 333 P.2d 178 ].)

This court’s recent opinion in Montrose Chemical Corp. v. Superior Court (1993) 6 Cal.4th 287 [ 24 Cal.Rptr.2d 467 , 861 P.2d 1153 ], decided during the pendency of this appeal, is to be distinguished from the instant case. That appeal involved a separate CERCLA action brought against Montrose in 1990 by the United States and the State of California alleging that Montrose’s operation of its Torrance facility caused environmental contamination that damaged land, water, and wildlife in the Los Angeles harbor basin and neighboring waters (United States v. Montrose Chemical Corporation of California (U.S. Dist. Ct. C.D.Cal.), 1990, No. CV 90-3122-AAH (Jrx)), and a related cross-complaint filed against Montrose by the Los Angeles County Sanitation District. (Montrose Chemical Corp. v. Superior Court, supra, 6 Cal.4th at p. 292 .)

In Montrose Chemical Corp. v. Superior Court, supra, 6 Cal.4th 287 , Montrose’s tender of defense was rejected and it brought a declaratory relief action against its various CGL insurers, seeking a declaration that each owed a duty to defend in the federal action and cross-complaint proceedings. The insurers denied they owed a duty to defend and asserted a number of affirmative defenses, including, as in this proceeding, several based on exclusion-of-coverage language contained in the various policies. Montrose moved for summary adjudication on the issue of the insurers’ duty to defend, arguing it was entitled as a matter of law to have its insurers defend it in the underlying CERCLA action because the allegations of the complaint, along with the terms of the CGL policies, created a potential for coverage, thereby triggering the defense duty. The insurers countered that Montrose had failed to establish it was entitled to summary adjudication, and that extrinsic evidence revealed a triable issue of fact regarding whether a potential for coverage existed, undercutting the basis for Montrose’s motion. (Montrose Chemical Corp. v. Superior Court, supra, 6 Cal.4th at p. 293 .) The trial court denied Montrose’s motion, concluding it had failed to make a prima facie showing that the CERCLA action created a potential for coverage because the allegations of the third party’s complaint, upon which Montrose was relying, were “neutral” regarding *662 whether the alleged contamination was caused by an “occurrence” within the meaning of the policies, or by Montrose’s regular business practices (which the trial court evidently viewed as outside the concept of “occurrence”). (Ibid.) The trial court also found that the insurers had adduced sufficient extrinsic evidence to create a triable issue of fact as to whether the CERCLA complaint alleged acts within the policies’ terms of coverage. (Id. at pp. 293-294.)

On Montrose’s petition for a writ of mandate, the Court of Appeal directed tire trial court to reconsider and grant the motion, finding Montrose had made a prima facie showing of potential coverage under the policies there in issue. We granted review and ultimately affirmed the judgment of the Court of Appeal, concluding Montrose had made a prima facie showing of potential coverage sufficient to trigger the insurers’ duty to defend. (Montrose Chemical Corp. v. Superior Court, supra, 6 Cal.4th at pp. 291, 294.) We explained that “the fact that toxic discharges occurred over a lengthy period during which Montrose operated its Torrance facility does not, without more, establish that Montrose expected or intended the property damage that allegedly resulted. [Citations.]” (Id. at p. 304.) And we found the insured’s allegations sufficient to raise the possibility that it would be liable for property damage covered by the policies, concluding further that “[ejxtrinsic evidence adduced by the insurers did not eliminate that possibility, but merely placed in dispute whether Montrose’s actions would eventually be determined not to constitute an occurrence or to fall within one or more of the exclusions contained in the policies.” (Ibid.) We did not, however, have occasion to address the trigger-of-coverage issues presented herein, because the timing of the circumstances giving rise to coverage in relation to the relevant CGL policy periods was not directly at issue in that appeal.

It should be noted that Admiral did not advance the loss-in-progress theory as applicable to the Levin Metals cases; Admiral has not contended that Montrose (as opposed to the third party claimants in the Levin Metals litigation) had knowledge of the contamination at the Parr-Richmond site prior to the commencement of Admiral’s policy periods.

The policy definition of “property damage to which this insurance applies” also includes “loss of use of tangible property which has not been physically injured or destroyed provided such loss of use is caused by an occurrence during the policy period.” Since the “loss of use” clause pertains only to property “which has not been physically injured or destroyed,” the sustaining of damage or injury to such property during the policy period by definition cannot be what triggers coverage for such losses. Because the parties have not directed us to any “loss of use” issue in this case, we have no occasion to decide whether coverage, under such a policy, for loss of use of tangible property which has not been physically injured or destroyed is dependent upon whether the loss of use of the property occurs during the policy period, or whether the occurrence which results in the loss of use occurs during the policy period.

ISO is a nonprofit trade association that provides rating, statistical, and actuarial policy forms and related drafting services to approximately 3,000 nationwide property or casualty insurers. Policy forms developed by ISO are approved by its constituent insurance carriers and then submitted to state agencies for review. Most carriers use the basic ISO forms, at least as the starting point for their general liability policies. (New Castle County v. Hartford Acc. and Indem. Co. (3d Cir. 1991) 933 F.2d 1162, 1181 ; see also 3 Cal. Insurance Law & Practice (1986) Property and Liability Insurance, § 49.04, at pp. 49-9 to 49-10.)

The 1973 standard form CGL policy language, which was incorporated in Admiral’s policies, was revised in a number of respects by ISO, but the insuring agreement and coverage-related definitions were substantially unaltered.

We are aware of only one appellate court decision that has adopted the manifestation trigger of coverage for bodily injuries in the context of third party liability insurance. In Eagle-Picher Industries, Inc. v. Liberty Mut. Ins. (1st Cir. 1982) 682 F.2d 12 , the United States Court of Appeals for the First Circuit concluded on the evidence before it that the injury resulting from inhalation of asbestos fibers did not “occur” until the symptoms of the disease asbestosis had manifested themselves. The asbestos manufacturer had no insurance prior to 1968, the period when most of the exposure took place. The manufacturer’s CGL insurance coverage began when the number of claims began accelerating. As was the case in Forty-Eight Insulations, supra, 633 F.2d 1212 , the court in Eagle-Picher, in adopting the manifestation trigger, made clear its intention to interpret the policies in a manner that would afford and maximize coverage on the particular facts of that case. (682 F.2d at p. 23.) The Eagle-Picher case therefore stands as somewhat of an aberration.

The injury-in-fact trigger has been applied in actions involving asbestos-related disease because symptoms of the disease often will not manifest themselves until decades after actual inhalation of asbestos fibers. Like the manifestation and continuous injury theories, the injury-in-fact theory assumes as a predicate that mere exposure to asbestos during the policy period is not enough to trigger coverage: “The plain language of the definition of ‘occurrence’ used in the CGL policy requires exposure that ‘results, during the policy period, in bodily injury’ in order for an insurer to be obligated to indemnify the insured. The unambiguous meaning of these words is that an injury—and not mere exposure—must result during the policy period. The CGL policies expressly distinguish exposure from injury; to equate the two . . . is to ignore this distinction. Any argument that mere exposure—without injury—triggers liability is simply unsound linguistically.” (Abex Corp. v. Maryland Cas. Co., supra, 790 F.2d at p. 127, italics in original; see also American Home Products Corp. v. Liberty Mut. Ins., supra, 748 F.2d at p. 764 .)

Unlike the manifestation trigger, however, the injury-in-fact trigger acknowledges that actual injury may “occur” before it has become manifest or been discovered. Under the injury-in-fact approach, coverage is triggered by “ ‘a real but undiscovered injury, proved in retrospect to have existed at the relevant time . . . irrespective of the time the injury became [diagnosable].’ ” (American Home Products Corp. v. Liberty Mut. Ins., supra, 748 F.2d at p. 766 .) That is, after an injury has been diagnosed, it may be inferred, from evidence establishing the “gestation period” and the stage to which the illness has advanced, that the harm or “injury-in-fact” actually began sometime earlier. (Id. at p. 765.) The injury-in-fact trigger therefore affords coverage for any ensuing continuing or progressively deteriorating injury that can be shown to have occurred during a successive policy period, regardless of whether such injury manifested itself or was discovered during that period.

In Armstrong World Industries, Inc. v. Aetna Casualty & Surety Co. (Cal.App.) (Armstrong), the First District Court of Appeal affirmed a trial court’s decision applying a “continuous injury” trigger to asbestos claims for which coverage was being sought under various CGL policies. The Armstrong court observed, however, that “[t]he trial court’s continuous trigger decision . . . [was] based upon factual findings that for asbestos claimants an injury-in-fact took place during each triggered policy period, even though the injury was not diagnosable and compensable during the policy period.” (Italics added.) Indeed, the *677 Armstrong court carefully noted that the trial court had “relied upon medical evidence to make factual findings on the physiological processes that actually occur upon inhalation of asbestos fibers and continue until death . . . .” in determining to apply a “continuous injury” trigger in that case.

Although the Armstrong court’s trigger of coverage discussion appears largely consistent with our analysis of the applicable principles of third party CGL coverage in the present case, because we do not here face the unique facts of asbestos-related bodily injury claims, we deem it appropriate that trigger of coverage questions specifically involving asbestos claims be left for decision, in the first instance, on an appropriate record in a case in which they are squarely presented.

Counsel for Admiral argued: “The occurrence in this case is when the dumping of toxic waste resulted in appreciable damage. Whether it was discoverable or not is not the issue. The issue is when it resulted in appreciable damage. And regardless of when that happened, it certainly happened a long time before the Admiral policies incepted. . . . When it occurred, which may or may not be the date of manifestation. It’s all going to depend on the facts of the particular case.”

On close scrutiny, it can be seen that Admiral is not advancing a true injury-in-fact trigger of coverage theory in lieu of a manifestation theory. As explained, under the injury-in-fact theory, continuing injury occasioned during the policy period which occurs subsequent to the point in time at which the injury-in-fact can first be pinpointed is subject to coverage. It is the period from initial exposure to the point at which the injury-in-fact is retrospectively first established for which no coverage is afforded. Admiral, in contrast, appears to be arguing that once an injury-in-fact is established, even retrospectively, all potential coverage is cut off from that point onward, and only the insurer on the risk at the time the injury-in-fact first “occurs” is liable to indemnify the insured, regardless of whether the injury-in-fact ever manifested itself.

One of the cases most frequently cited for this proposition is Snapp v. State Farm Fire & Cas. Co. (1962) 206 Cal.App.2d 827 [ 24 Cal.Rptr. 44 ] (Snapp), the first California case to discuss a manifestation theory in the first party property insurance context. (See Prudential-LMI, supra, 51 Cal.3d at pp. 694-696.) Snapp was a declaratory relief action to determine the extent of an insurer’s liability, both before and after the expiration of a first party standard form fire insurance policy, for damage to the insured premises resulting from a landslide. The effective date of the coverage was November 15,1956, and it was to continue for a three-year term. The policy contained an endorsement which extended coverage to insure against all risks of physical loss to the property. During the policy term, the land beneath the Snapp residence began to move laterally due to unstable landfill and heavy rainfall. The trial court found the property insurer was liable only to the extent of the damages sustained before the expiration of the policy period.

In reversing, the Snapp court made reference to a specific finding that the “ ‘movement is still active and is without definite prospect of stabilization.’ ” (Snapp, supra, 206 Cal.App.2d at p. 831 .) The court continued: “To permit the insurer to terminate its liability while the fortuitous peril which materialized during the term of the policy was still active would not be in accord either with applicable precedents or with the common understanding of the nature and purpose of insurance; it would allow an injustice to be worked upon the insured by defeating the very substance of the protection for which his premiums were paid. [¶] Once the contingent event insured against has occurred during the period covered, the liability of the carrier becomes contractual rather than potential only, and the sole issue remaining is the extent of its obligation, and it is immaterial that this may not be fully ascertained at the end of the policy period.” (Id. at pp. 831-832, italics in original.)

We do not endorse that aspect of the California Union court’s holding that both insurers in that case were jointly and severally liable for the full amount of damage occurring during the successive policy period. (California Union, supra, 145 Cal.App.3d at p. 478 .) Allocation of the cost of indemnification once several insurers have been found liable to indemnify the insured for all or some portion of a continuing injury or progressively deteriorating property damage requires application of principles of contract law to the express terms and limitations of the various policies of insurance on the risk. (See Keene Corp. v. Ins. Co. of North America (D.C. Cir. 1981) 667 F.2d 1034, 1051 [ 215 App.D.C. 156 ] (Keene); Forty-Eight Insulations, supra, 633 F.2d at p. 1225.)

Although postdating Gruol, supra, 11 Wn.App. 632 [ 524 P.2d 427 ], Keene has generally been regarded as one of the first cases to adopt a continuous injury trigger of coverage analysis, at least in the context of claims of progressively deteriorating bodily injury in asbestos cases. (See, e.g., Aspinwall, The Applicability of General Liability Insurance to Hazardous Waste Disposal (1984) 57 So.Cal.L.Rev. 745, 755.) As the court in California Union recognized, although Keene was an asbestos case, the basic rationale of that decision is instructive on the question of what trigger of potential coverage should be applied in the context of continuous or progressively deteriorating property damage.

The Fireman’s Fund opinion does set forth the standard CGL policy “insuring clause,” which provides that: “The company will pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of ‘bodily injury’ or ‘property damage’ to which this insurance applies, caused by an ‘occurrence.’ ” (Fireman’s Fund, supra, 223 Cal.App.3d at p. 1628 .) Citation to this portion of the standard policy language hardly serves to support the Fireman’s Fund court’s analysis.

Decisions of the highest courts of other states which are consistent with the conclusions we reach today—rejecting a manifestation trigger and adopting the continuous injury trigger of coverage for claims of continuing or progressively deteriorating bodily injury or property damage arising under third party CGL policies—include Owens-Illinois, Inc. v. United Ins. Co. (1994) 138 N.J. 844 [ 650 A.2d 974, 990, 995 ] (New Jersey Supreme Court unanimously adopts continuous injury trigger in a case involving CGL coverage for asbestos claims relating to both bodily injury and property damage); Trustees of Tufts Univ. v. Commercial Union Ins. Co. (1993) 415 Mass. 844 [ 616 N.E.2d 68, 74 ] (Massachusetts Supreme Judicial Court unanimously rejects manifestation trigger in environmental property damage case involving claims under occurrence-based CGL policies, explaining, “Nothing in the language of the policies requires that the claimed property damage be discovered or manifested during the policy period. The inquiry instead is whether property damage, as defined in the policies, ‘occurred’ within the policy period and within the meaning of the word ‘occurrence.’ [Citation.] Indeed, the very nature of an ‘occurrence’ as opposed to a ‘claims-made’ policy is to provide coverage for property damage that occurred during the policy period whenever that liability is imposed.”); J.H. France Refractories Co. v. Allstate Ins. Co. (1993) 534 Pa. 29 [ 626 A.2d 502, 507 ] (Pennsylvania Supreme Court unanimously adopts continuous injury or “multiple” trigger for asbestos-related bodily injury claims brought under CGL policies, noting, “Rather than selecting one or another of the phases [exposure, manifestation, discovery, etc.] as the exclusive trigger of liability, it seems more accurate to regard all stages of the disease process as bodily injury sufficient to trigger the insurers’ obligation to indemnify, as all phases independently meet the policy definition of bodily injury.”); Harford County v. Harford Mutual Ins. Co. (1992) 327 Md. 418 [ 610 A.2d 286, 294-295 ] (Maryland Court of Appeals rejects manifestation trigger in environmental pollution case); see also Sentinel Ins. Co. v. First Ins. Co. (1994) 76 Hawaii 277 [ 875 P.2d 894, 917 ] (Hawaii Supreme Court unanimously adopts injury-in-fact trigger in construction defect case alleging claims for property losses under CGL policies, but explains that “where injury-in-fact occurs *687 continuously over a period covered by different insurers or policies, and actual apportionment of the injury is difficult or impossible to determine, the continuous injury trigger may be employed to equitably apportion liability among insurers.”).

Other courts which have recently applied a continuous injury trigger in environmental property damage cases for which coverage was claimed under standardized occurrence-based CGL policies include the Oregon Court of Appeals (St. Paul Fire & Marine Ins. Co. v. McCormick & Baxter Creosoting Co. (1994) 126 Ore.App. 689 [ 870 P.2d 260 , 264-265]), and the United States District Court for the District of Delaware (Harleysville Mut. Ins. Co. v. Sussex County (D. Del. 1993) 831 F. Supp. 1111, 1124 [applying Delaware law]).

One commentator has suggested that, “because it encourages all insurers to monitor risks and change appropriate premiums, the continuous trigger rule appears to be the most efficient doctrine for toxic waste cases.” (Note, Developments in the Law—Toxic Waste Litigation, supra, 99 Harv. L.Rev. at p. 1581.)

“Claims made” policies beneficially permit insurers more accurately to predict the limits of their exposure and the premium needed to accommodate the risk undertaken, resulting in lower premiums than are charged for an occurrence-based policy. (See, e.g., Pacific Employers Ins. Co. v. Superior Court (1990) 221 Cal.App.3d 1348, 1359-1360 [ 270 Cal.Rptr. 779 ].) Another name for a “claims made” policy is a “discovery” policy. (VTN Consol., Inc. v. Northbrook Ins. Co. (1979) 92 Cal.App.3d 888, 891 [ 155 Cal.Rptr. 172 ].) “Claims made” coverage arose more than 20 years ago, initially in the field of professional liability insurance, *689 because underwriters were concerned that occurrence-based coverage was adversely affecting the underwriting process. Because the injury and negligence giving rise to a malpractice claim is often not discoverable until years after the negligent act or omission, professional liability insurance carriers, in an effort to reduce their exposure to an unpredictable and lengthy “tail” of lawsuits, shifted to the “claims made” policy. (Pacific Employers Ins. Co. v. Superior Court, supra, 221 Cal.App.3d at p. 1358; see also Keeton & Widiss (1988) Insurance Law: A Guide to Fundamental Principles, Legal Doctrines, and Commercial Practices, § 5.10(d), p. 598.) The “claims made” concept was subsequently extended into the field of general liability coverage, and in 1986 ISO issued both a revised standard form occurrence-based CGL policy (now referred to as a commercial general liability policy) and a new standard form CGL “claims made” policy.

The Hawaii Supreme Court has likewise concluded that even where an injured third party has filed a lawsuit or claim against the insured, “if the insured’s liability is in any degree contingent, there is an insurable risk” within the meaning of the loss-in-progress rule, (Sentinel Ins. Co. v. First Ins. Co., supra, 875 P.2d at p. 920 , fn. omitted.)