United States Air and Radiation May 1993 Environmental Protection Research Triangle Park, NC 22711 EPA 453/N-93-003 Agency (MD-13) xvEPA IV IATICH I Newsletter National Air Toxics Information Clearinghouse NATICH BBS Now On-Line* by Vasu Kilaru, EPA Office of Air Quality Planning and Standards If! ThlS ISSUe... The NATICH data base can now be accessed via EPA's Technology Transfer EPA Einds Success with Negotiated Network Bulletin Board System (TTN Rulemaking... 4 BBS). The on-line service started in March and is available to any interested 1991 TRI Data Release to Include user for the Price of a Phone calL New Data 6 Clearinghouse data base was established to provide a centralized, EPA's RIHRA Program Studies accessible system to store and maintain #ncer Modeling 6 Federal, State, and local air toxics pro- gram data. It contains information on the development and implementation of air toxics programs, agency names and con- tacts, and ongoing air toxics research. State and local agencies use the Clear- inghouse information to support the complex decisions needed to regulate toxic air pollutants. The NATICH data base has been stored on the IBM mainframe at EPA's National Computer Center since 1984. To streamline and improve the accessi- bility of NATICH to its users, the Clear- inghouse staff began adding portions of the data base to the TTN BBS in March. The following information is now avail- able: agency contacts; program descriptions; (continued on page 2) Frequently Used Acronyms CAA - Clean Air Act Amend- ments of 1990 HAPs - Hazardous Air Pollutants OAQPS - Office of Air Quality Planning and Standards TRI - Toxics Release Inventory '""V 11993 Northeast States Continue to Pursue Regional Adoption of the California LEV Program* States in the Northeast continue to move forward in their efforts to adopt and implement the California low emis- sion vehicle (LEV) program throughout the region in spite of continued opposi- tion from automakers.** Maine recently became the third Northeast State to adopt the LEV standards (see Table 1), joining Massachusetts and New York. Approximately 60 percent of the vehicles in the eight-state NESCAUM*** region and 35 percent of vehicles in the entire Northeast Ozone Transport Region (OTR)**** are registered in these three States. Other OTR States are actively pursuing LEV programs through regula- tory and legislative initiatives. In October 1991, California submit- ted a waiver request to EPA for their LEV program that sets standards more stringent than Federal standards.***** (continued on page 2) Produced in / AU\[p(Dİ conjunction State and Territorial Air Pollution Program Administrators with Association of Local Air Pollution Control Officials ------- NATICH BBS (continued from page 1) ambient air concentration guidelines and standards; pollutant research; methods development; nonhealth-related impacts; permit information; source tests; ambient monitoring; emissions inventories; bibliographic citations; and ongoing research and regulatory development activities. The NATICH BBS also features com- munications options for downloading publications lists, Clean Air Act (CAA) information such as the list of Section 112 hazardous air pollutants (HAPs) and physical properties of HAPs, and current and some past issues of the NATICH Newsletter. BBS users can also access summary files from the EPA's Toxics Release Inventory** System (TRIS), which contains multimedia release infor- mation for over 300 toxic chemicals reported by industry under the Emer- gency Planning and Community Right- to-Know Act of 1986. To implement the BBS method of access, Clearinghouse staff condensed the data stored to reduce data redundan- cies and improve system performance. For example, permit information is now reported by pollutant or source category (industry type, i.e., by Standard Indus- trial Classification or SIC Code), rather than by individual permit, which elimi- nates duplicate data and summarizes information for users. Similar streamlin- ing was done for ambient air monitoring and source testing information. Additional functions and continued improvement are planned for NATICH BBS. An interactive on-line data entry capability is scheduled to be available in May. State and local agencies are encouraged to use this upcoming data entry capability to keep their information current. In the past such data was only entered once a year. The OAQPSTTN is available 24 hours per day, except for Mondays fron^ 8 a.m. to 12 p.m. Eastern Standard Time when it is down for routine maintenance and backup. Users with a computer and modem can access the TTN by calling (919) 541-5742. The following configura- tion is required for the communication software: 8 data bits, 1 stop bit, no parity, full duplex, and a terminal emulation of VT 100, VT 102, or American National Standards Institute (ANSI). For further general information about the TTN or how to access it, call the TTN hotline, (919) 541-5384. T1h> Clearinghouse staff welcomes comments and suggestions regarding the NATICH BBS or any aspect of its operations. Contact Vasu Kilaru, U.S. EPA, OAQPS, MD-13, Research Triangle Park, North Carolina 27711, or call (919) 541-0850. *See related articles in the March 1992 and March 1993 Newsletters. * *See related article in this issue. line specifications. The court held that States were not legally bound to adopt California's gasoline specifications in order to adopt the LEV standards. However, the court agreed with the automakers that New York's decision not to adopt the California reformulated gasoline specifications as part of the LEV program violated the CAA because it would necessitate the production of a "third vehicle," which is neither a "Federal" car nor a "California" car. The court upheld the automaker's argument that catalytic converters on LEVs in New York would degrade more rapidly than those in California, because commercial gasoline sold in the Northeast has a higher sulfur content than that sold in California (sulfur restrictions are more stringent in California than in the Northeast States). The automakers cited potential warranty and recall implicatio resulting from the use of different fuel formulations. They argued that because (continued on page 3) California LEV Program (continued from page 1) In January 1993, EPA issued an uncon- ditional waiver for California's LEV pro- gram. Significantly, the EPA refused a request by the automakers to make the granting of the waiver conditional on the commercial availability of California's Phase 2 reformulated gasoline. Auto- makers had raised concerns about possi- ble warranty and recall implications asso- ciated with the effects of differing fuel formulations (i.e., the California refor- mulated, Federal reformulated, or indus- try average fuels) on engine or emission control equipment. EPA's decision to issue the unconditional waiver is particu- larly important to Northeast States who are adopting the LEV standards and have opted to participate in the Federal refor- mulated gasoline program. The Federal gas program is likely to be somewhat less stringent than California's. However, the Northeast States view adoption of the LEV standards in conjunction with Federal reformulated gasoline as a signif- icant means to control auto emissions and improve air quality. In July 1992, the automakers filed a Federal law suit with six specific counts against New York challenging several components of the LEV regulations pro- posed by the Department of Environ- mental Conservation (NY DEC). In January 1993, the court found in favor of the automakers on four counts and in favor of the NY DEC on two counts. Two key issues in the suit were the legal relationship between fuels and emission standards and the legality of the zero emission vehicle (ZEV) man- date. In regard to the first issue, the automakers argued that New York's decision not to adopt California's refor- mulated gasoline program in conjunc- tion with the LEV standards violated the CAA. The court found in favor of New York on this count citing the fact that California's waiver request per- tained only to the LEV standards and did not address the reformulated gaso- 2 ------- California LEV Program (continued from page 2) Table 1. Chronology of California LEV Program Adoption Date Milestone September 28, 1990 California adopts LEV standards October 4, 1991 California submits request to EPA for Waiver of Federal Preemption October 29, 1991 OTR States sign Memorandum of Understanding to adopt LEV standards throughout the Northeast January 31, 1992 Massachusetts becomes first State to join California in adopting LEV program March 23,1992 New York adopts LEV regulations July 9, 1992 MVMA files suit against New York's LEV regulations in U.S. District Court in Binghampton, New York January 8, 1993 EPA grants California an unconditional Waiver of Federal Preemption January 22, 1993 U.S. District Court in Binghampton, New York, issues decision on New York LEV suit January 27, 1993 Maine adopts LEV regulations February 9, 1993 New York files Notice of Motion for Reargument and Stay of LEV decision March 10, 1993 Maine is served with papers in LEV suit brought by the AAMA and AIAM April 9, 1993 AAMA, AIAM, and the Massachusetts State Automobile Dealers Association file a Federal suit in Massachusetts to block implementation of the LEV standards tthe need to replace key emission con- >1 components during the useful life of the vehicle, they would have to install the catalytic converters differently in vehicles bound for New York than those intended for California, consequently, requiring the production of a third vehicle. On the second issue, the legality of the ZEV mandate, New York's regula- tions are identical to California's. Both require that 2 percent of new vehicles produced for sale in their respective States in 1998 must certify as ZEVs. 'Hie mandate increases to 5 percent of new vehicles in 2001, and to 10 percent in 2003. The court agreed with the auto- makers that this provision in the New York regulations violated the CAA because it constituted a restriction in the sale of the three other categories of Cali- fornia vehicles (transitional LEVs, LEVs, and ultra-LEVs). Basically, automakers argued that the ZEV-mandated percent- ages would restrict the sales of these three other categories of vehicles ^fcuired under California's LEV prgram. The State of New York has filed a motion for reargument in hopes that the court will reconsider its LEV decision. Air quality officials in New York and other OTR States believe that the court decision is flawed and that it will be overturned based on a reexamination of existing information. In fact, a special commission appointed in Massachusetts to review that State's LEV regulations voted in early March not to change their regulations. However, automakers filed a Federal suit to block implementation of the LEV stan- dards in Massachusetts in April. Also, the Maine Board of Environmental Protection adopted the LEV program after the court ruling in New York. Plaintiffs opposed to Maine's LEV program have already served the State with papers to initiate a lawsuit. In addition, the legislatures in both New Jersey and Maryland have recently passed LEV bills since the New York court decision. For more information, contact Arthur Marin at NESCAUM at (617) 367-8540. *See related articles in the March 1990 and January 1992 Newsletters. * *The Motor Vehicle Manufacturers Association (MVMA) originally repre- sented both domestic and foreign auto- makers. It separated after legal action was initiated into the American Automobile Manufacturers Association (AAMA) and Association of International Automobile Manufacturers (AIAM). * * *The eight States comprising the Northeast States for Coordinated Air Use Management (NESCAUM) region are Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Rhode Island, and Vermont. *** *The OTR includes the eight NESCAUM States, Pennsylvania, Delaware, Maryland, Virginia, and the District of Columbia. **** *Section 209(b) (1) of the 1990 Clean Air Act Amendments (CAA) pro- vides that any State that adopted motor vehicle emission standards before March 30, 1966, that are equal to or more stringent than Federal standards can request and receive an EPA waiver certify- ing this. California is the only State that meets this eligibility requirement for receiving waivers. 3 ------- EPA Finds Success with Negotiated Rulemaking Although the EPA follows a strict public notice and participation strategy in the development of regulations, approxi- mately 80 percent of its resulting "major" regulations are subject to litiga- tion, and implementation may be delayed for years. To see if there was a better way to develop some of the more than 200 regulations normally in the works, the EPA established the Regulatory Negotiation Project in 1982. Its goal was to investigate the potential for using negotiation in developing better pro- posed rules and reducing the nearly inevitable uncertainty and delay in imple- mentation that accompany litigation. A regulatory negotiation (or "reg neg") is an innovative way to develop proposed rules. A federally chartered advisory committee is convened with representatives from all affected parties including industry, State and local offi- cials, environmental groups, and the EPA. The members negotiate on behalf of their constituencies and attempt to reach consensus on exact language or agreement on core conceptual principles that can be used as the basis of a pro- posed rulemaking. EPA agrees to use the consensus as the basis of its pro- posed rule. The negotiations are chaired by a neutral facilitator whose role is to keep the parties moving toward consen- sus. Once a proposed rule is published, the negotiation is completed, and the Administrative Procedures Act and other relevant statutes govern the remainder of the rulemaking process. EPA views regulatory negotiation as a win-win strategy. The Agency gains in developing a more environmentally effec- tive, innovative, and pragmatic rule, which reduces litigation costs and delay in rule implementation. The regulated industries are able to participate more fully in the development of a rule, reduc- ing uncertainty about what will be required and leading to rules that are more understandable. State and local agencies, often responsible for imple- menting programs to enforce EPA rules, are able to build in administrative needs. Environmental and public interest groups concerned about reducing pollu- tants in the environment and represent- ing the "little guy," are all able to voice their concerns and impact the propose^ regulations more fully and openly than under traditional informal notice and comment rulemaking. Since establishing the program, nearly a dozen regulatory negotiations have been completed with a strong suc- cess record. Only three did not achieve final consensus, and for the two rules where litigation occurred in spite of the negotiation, the cases were decided in EPA's favor. Table 1 summarizes the sta- tus of these negotiated rules. It is not feasible for the Agency to conduct negotiations for all preproposal rules under development. EPA's Regu- latory Negotiation Project, directed by Chris Kirtz, screens rules to identify those that are good candidates for nego- tiation. Criteria include an identifiable and limited set of issues, a reasonably limited and discernable balance of inter- est groups, and a data base sufficient for meaningful discussion of issues. A nego- tiation may be initiated if the Agency believes it is to mutual advantage and that the interested parties can negotiati- (continued on page 6) Table 1. EPA Negotiated Rulemaking Status as of April 1993 Rules Standard/EPA Office Negotiation Status Rulemaking Status Notes Nonconformance Penalties for Heavy-Duty Trucks CAA/Office of Mobile Sources (OMS) First met 6/84 Consensus reached 10/84 Proposed 3/85 Promulgated 8/85 Only 13 comments on proposal; all supportive. No litigation. Emergency Pesticide Exemptions Federal Insecticide, Fungicide & Rodenticide Act (FIFRA)/Office of Pesticide Programs (OPP) First met 8/84 Consensus reached 1/85 Proposed 4/85 Promulgated 1/86 19 comments received. No litigation to date. Farmworker Protection Standards FIFRA/OPP First met 11/85 Committee unable to achieve consensus but continued informally. Proposed 7/88 Promulgated 8/92 100's of comments received and reviewed before pro- mulgation. Woodburning Stoves New Source Performance Standard/OAQPS First met 3/86 Consensus reached 8/86 Proposed 2/87 Promulgated 2/88 50 comments received. 1 No litigation to date. (continued on page 5) 4 ------- Negotiated Rulemaking (continued from page 4) Table 1. EPA Negotiated Rulemaking Status as of April 1993 (continued) Rules Standard/EPA Office Negotiation Status Rulemaking Status Notes Resource Conservation and Recovery Ad (RCRA) Minor Permits Modification RCRA/Office of Solid Waste First met 9/8(i 20 of 21 members concurred 2/87 Proposed 9/87 Promulgated 9/88 55 comments received. No litigation to date. Underground Injection of I iazardous Wastes Hazardous and Solid Waste Amendments of 1984/Office of Drinking Water First met 9/8(i Good agreement but not total consensus in 3/87 Proposed 8/87 Promulgated 7/88 30 comments received. Litigation did not involve the use of reg neg and was decided in EPA's favor. Asbestos in Schools Asbestos Hazard Emergency Response Act of 1986/Office of Toxic Substances (OTS) First met 2/87 20 to 24 members came to agreement 2/87 Proposed 4/87 Promulgated 10/87 130 comments received. Litigation by nonsignatories. Ruled in EPA's favor. Fugitive Emissions from Equipment I-eaks National Emission Standard for Hazardous Air Pollutants (N ESHAP) /OAQ PS First met 9/89 Consensus reached 10/90 Proposed 12/92 Promulgation expected Included in Hazardous Organic NESHAP proposal; one of the first maximum achievable control tech- nology (MACT) standards. I^ead-Acid Battery Recycling Toxic Substances Control Act/OTS First met 1/91 EPA suspended negotiations 5/91 to develop additional cost and risk estimates. Rulemaking suspended EPA decided risk/benefits gain insufficient to regulate at this time. Oxygenated and Reformulated Fuels CAA/OMS First met 3/91 Consensus reached 8/91 EPA issued rules to use more ethanol in 10/92 President Bush ordered EPA to reconsider rules to provide greater use of ethanol in 9/92; further action anticipated. Coke Oven Batteries MACT Standards NESHAP/OAQPS First met 2/92 Consensus reached 10/92 Proposed 12/92 A smaller group within the committee drafted rule and preamble. Disinfection By-Products Safe Drinking Water Act/Office of Water First met 11/92 Completion expected 9/93 Pending outcome of negotiation Possible risk trade-offs, complex technological issues are unique. Architectural and Industrial Maintenance Coatings CAA/OAQPS F irst met 10/92 Completion expected 9/93 Pending outcome of negotiation First consumer product rule/negotiation under § 183(e). Revision of Hazardous Waste Manifest RCRA/Office of Policy Planning and Evaluation First met 10/92 Completion expected 9/93 Pending outcome of negotiation Negotiation result of ASTSWMO3 petition to standardize varying State manifests. "Association of State and Territorial Solid Waste Management Officials. 5 ------- Negotiated Rulemaking (continued from page 4) in good faith to resolve the issues with- out compromising fundamental or stated goals, principles, or values. The public is encouraged to partici- pate by suggesting candidate rules for regulatory negotiation. Candidate rules recommended by the public will be eval- uated by the same criteria. For information regarding any of the specific negotiations summarized in Table 1 or for general information, rec- ommendations, or comments, contact Chris Kirtz, Director, Regulatory Negotiation Project (PM-223Y), U.S. EPA, 401 M Street, S.W., Washington, D.C. 20460, or call (202) 260-7565. 1991 TRI Data Release to Include New Data by William Wallace, EPA Office of Pollution Prevention and Toxics EPA's Office of Pollution Prevention and Toxics (OPIT) has scheduled release of the 1991 data in its Toxics Release Inventory (TRI) data base in late May. The 1991 release includes new data on source reduction and recycling activity from facilities required to file Form R for Reporting Year (RY) 1991 Oanuary 1 to December 31, 1991). This is the fifth year that information on routine and accidental releases of toxic chemicals to the environment has been available. The Pollution Prevention Act of 1990 requires eight new types of data to be reported for RY 1991 on the TRI Form R, under Part II, Section 8, "Source Reduction and Recycling Activities." All facilities subject to the mandatory TRI reporting requirements must include the following information on a facility-by- facility basis for each toxic chemical: quantities entering waste streams prior to treatment, recycling, or dis- posal or that are otherwise released to the environment, the percentage change from the previous year, and similar estimates for the next 2 years; quantities recycled, similar estimates for the next 2 years, and the recy- cling process used; the percentage change in the quantity recycled from the previous year to the current reporting year; source reduction practices; techniques used to identify source reduction opportunities; a production index (i.e., the ratio of production in the reporting year to production in the previous year); amounts of chemicals released through catastrophic events, remedial actions, or other one-time events not associated with production processes; and the amount of chemical that is treated and the percentage change from the previous year. The first reports containing this expanded information were due to EPA and the States by July 1, 1992. At the close of business on April 8, 1993, the Agency's mainframe TRI data base was frozen (no additional changes were per- mitted) for one week to produce com- puter tape extracts of the data. These extracts were provided to organizations including the National Library of Medicine (NLM), Government Printing Office (GPO), and National Technical Information Service (NTIS) for creation of publicly available data products, such as compact disks or computer diskettes with data available on a State-by-State basis, and reports including the com- bined national report, the 1991 Toxics Release Inventory Public Data Release (EPA 700S-93-002). All RY 1991 Form R report data received through February 19, 1993, are included in the public data release. Data quality assurance activities continued until the freeze date and included resolv- ing outstanding Notices of Non-Compli- ance, correcting any data entry errors, investigating invalid codes and other pos- sible data discrepancies, and entering Form R revisions received from facilities. For further information, contact Steve Newburg-Rinn or Ruby Boyd at U.S. EPA, OPPT, TS-793, 401 M Street, S.W., Washington, D.C. 20460, or call (202) 260-3757. EPA's RIHRA Program* Studies Cancer Modeling by Marc J. Mass, EPA Health Effects Research Laboratory This is the second article in a series that focuses on current activities supported by RIHRA to advance the scientific basis for risk assessments for all regulatory programs. In 1986, EPA issued Carcinogen Risk Assessment Guidelines that supported the use of the linearized multistage (LMS) model as essentially the default mathematical procedure for performing high-dose to low-dose extrapolations in carcinogen risk assessments. However, the 1992 draft revisions to these guide- lines give substantial latitude to use other models. The updated guidelines will apparently stress that the mecha- nism of carcinogenesis should be consid- ered when determining the quantitative methods to use when extrapolating from high to low doses. The revised guide- lines also allow the use of models other than the LMS when sufficient information regarding a carcinogen's mechanism of action permits mathematical description that is concordant with the biological events in the carcinogenesis process. EPA's Research to Improve Health Risk Assessments (RIHRA) program hcg been aware of the Agency's desire to (continued on page 7) 6 ------- RIHRA Program (continued from page 6) plore the utility of low-dose extrapola- n models other than the LMS model for carcinogen risk assessments, espe- cially biologically based dose-response (BBDR) models. Non-BBDR models such as the LMS (a modification of the Armitage-Doll model**) use as variables only the dose and tumor incidence. Therefore, such non-BBI)R models do not consider other factors that may dras- tically affect tumor incidence over time such as the number of cells susceptible to carcinogenic insult at the time of expo- sure, the rate at which carcinogen-induced changes occur in these cells, the growth rates of these altered (premalignant) cells into tumors, and the rate at which these cells die. The Moolgavkar-Knudson (MK) multistage model is a BBDR model that considers the additional parameters above. Researchers believe it may be capable of making more accurate risk extrapolations than the LMS model at low doses. The MK model has been used to predict tumor incidences in ^>man populations where there are Ranges in the age-related incidences of cancer, particularly in tumor types where hormonal fluctuations cause cell prolifer- ation rates to change over time in the tar- get organ. Non-BBDR models typically were unable to predict tumor incidences accurately in such tissues. Also, the MK model allows consideration of the effects of carcinogens that operate by genotoxic mechanisms (carcinogen-induced changes in DNA), or by nongenotoxic mechanisms (effects on cellular differen- tiation processes). EPA's Health Effects Research Laboratory (HERL) has established a cooperative research agreement with internationally known cancer researcher Dr. Henry Pitot at the University of Wisconsin's McArdle Laboratory for Cancer Research to experimentally derive the parameters needed to use the MK model with a rat liver carcinogenesis protocol. The data that the laboratory generates is being used by EPA'S Office of Health and Environmental Assess- ment (OHEA) to test how well the MK model performs high- to low-dose extrap- olations. Deriving parameters for the MK can- cer incidence model requires the use of experimental protocols in which early, carcinogen-induced changes can be read- ily recognized. Relatively few animal car- cinogenesis systems meet this criterion. A rat liver carcinogenesis protocol was chosen because it has been used produc- tively for many years to understand steps in the carcinogenesis process. In a sim- plified scheme, rat liver cells exposed to carcinogens grow into a mass called an altered hepatic focus (AHE) that is thought to eventually give rise to a tumor. By counting the number of AHFs after a dose of carcinogen, measuring the rate of growth (or recession) of the AHE, and the number of resultant tumors com- pared to the AHE, it is possible to derive some of the parameters of the MK can- cer incidence model. Will other high- to low-dose BBDR extrapolation models be useful to the Agency in performing more accurate quantitative carcinogen risk assess- ments? EPA researchers believe so and OHEA scientists have already compared BBDR and non-BBDR high- to low-dose extrapolation models using experimental data from a rat liver carcinogenesis study using dinitrotoluene. They showed clearly that the use of different extrapola- tion models would affect unit risk esti- mates for dinitrotoluene exposure. Such advances made in BBDR low-dose extrapolation modeling for cancer may benefit risk assessments used in all areas of regulation. For further information on this RIHRA project, contact Marc J. Mass, Ph.D., U.S. EPA Carcinogenesis and Metabolism Branch (MD-68), HERL, Research Triangle Park, North Carolina 27711. For further information on RIHRA, contact John Vandenberg, Ph.D., RIHRA Director, at the same address except for MD-51, or call (919) 541-4527. *See related articles in the July 1991 and January 1993 Newsletters. * *The Armitage-Doll model was among the earliest quantitative risk assessment models developed, and was used in the 1950s to study the relationship of cigarette smoking and lung cancer. ------- 'Hie NA T1CH Newsletter is published six limes a year by Ihe National Air Toxics Information Clearinghouse. The Newsletter is prepared by Radian Corporation under EPA Contract Number 68-1)1-0125, Work Assignment 2-14. The EPA Editor is laurel Driver, EPA Office of Air Quality Planning and Standards, Research Triangle Park, North Carolina 27711, Telephone (919) 541-2859. The Radian Project Director is Linda Cooper, Radian Corporation, P.O. Box Box 13000, Research Triangle Park, North Carolina 27709, (919) 541-9100. The Newsletter is distributed free of charge. To report address changes, write Meredith Haley, Radian Corporation, P.O. Box 13000, Research Triangle Park, North Carolina 27709. The views expressed in the NATICH Newsletter do not necessarily reflect the views and policies of the Environmental Protection Agency. Mention of trade names or commerical products does not constitute any endorsement or recommendation for use by EPA. BULK RATE Postage and Fees Paid E.P.A. G-35 United States Environmental Protection Agency Pollutant Assessment Branch, MD-13 Research Traingle Park, NC 27711 Official Business Penalty for Private Use $300 EPA 463/N-93-003 May 1993 ------- |