Tobacco Documents Online

~ASTER SUMM~Y FOR D&W SUBJECTIVE DOCUMENT REVIEW As of March 31, 1989 P~ 1 1 1 1 1 1 2 3 3 DescriDtlon ACUTE/CHRONIC INHALATION EXPOSURE TESTS ADDICT~0~ ADDITIVE-FI~EE CIGARS ADDITIVES GUID~NCE PANEL M~ETINGS (AGPI ADVERTISING & M_ARK~TING OF OTHER COMPANIES' BRANDS ALTERNATIVE CAUSATION AMA/ERF A~ALYSIS & COM~NTARY ~"BBA") ARIES FILTEB ASBESTOS ASSURANCES OF SAFETY & CONCESSIONS AWARENESS OF HAZARDS Q~ SIDESTREAM SMOKE B&W AWARENESS AND ACTION B&W/BAT AFFILIATES B&W/BAT M~ETINGS B&W/BAT P~ELATIONSHIP B&W KENTUCKY DERBY FLOAT B&W'S TESTING OF OTHER COMPANIES' BRANDS

5 5 5 5 5 6 6 6 6 6 7 7 7 7 7 7 7 8 8 8 8 8 8 9 9 B&W STOCKS B&W v. CBS FILES BARCLAY CONTROVERSY BAT BIOLOGICAL RZSEARCH BAT G~OUP ~&D CONFERENCES BAT ~TING BAT SPECIAL CONFERZNCES AND COMMITTEES BATUS BOARD OF DIP~ECTORS' MEETINGS BENOWITZ PAPER BIOLOGICAL TESTING COM~ITTEE MEETING MINUTES BLACKMAN LETTER BRAND HISTORIES FOR OTHER COMPANIES' BRANDS B P~AN D MANAGERS BRAND-SPECIFIC ADVERTISING & MARKETING BUDGETS BUSINESS BUILDING TEST CHAIRMAN' S ADVISORY COMMITTEE (CAC) CHARTERED PATENT AGENTS CIGARETTE ADVERTISING . CODE. I~C, CIGARETTE MOUTHPIECE ¢I GARETTE SALES qLEMENTS. EARL ¢~3SING OF LOUISVILLE PLANT COMPENSATION & TOLEKAN¢~ -2-

9 9 9 I0 i0 I0 12 12 12 12 12 12 13 13 13 13 13 13 14 15 15 15 15 15 16 COMPONENTS RESEARCH CONCEPT-PRODUCT TESTING CONFIDENTIAL RESEARCH REPORTS CONFIDENTIALITY AGREEMENTS CONFIRMATION OF CAUSATION CONSUMER DYNAMICS CONSUMPTIQ~ CONTROLLED STORE MICROMA_RKET TE~T CORPORATE MEETINGS CORPORATE PRODUCT DEVELOPMENT COMMITTEE MEETING COTININE {NICOTINE METABOLISM) RESEARCH COUNCIL FOR BURLEY TOBAC~Q COUPONS DISTRIBUTED IN GROCerY STORES CTR SPECIAL PROJECTS DEFENSIVE DOCUMENTS DESIGN OF T/N LEGEND DILUTION OF WAKNINGS DISEASES & CONDITIONS DISTINCTION BETWEEN "IC; B&W INVOLVEMENT" A/~D "ID; INDEPENDENCE/COMPANIES" DISTINCTION BETWEEN "ID; INDEPENDENCE/COMPA/~I ES" AND "IE; INDEPENDENCE/LAWYERS" DISTINCTION BETWEEN "KA: ADEOUACY" AND "KC; EFFECT" DISTINCTION BETWEEN MUTAGENESIS & CELLULAR CHANGES DISTINCTION BETWEEN TOBACCO TREATMENT AND TOBACCO DOCUMENT RETENTION/DESTRUCTIQN

16 16 16 17 17 17 18 18 18 18 19 19 20 2O 20 20 21 21 21 21 21 22 22 DONATIONS BY B&W TO CHA/~ITABLE E%rENTS & ORGANIZA- DuPONT TAS"fE PANEL ELLIS PATENT/PROJECT ARIEL EMPLOYEE INDOC'YRINATION EPIDEMIOLOGICAL S%~JDIES ETS ADVISORY COMMITTEE ~NATIONAL ETS ADVISORY EXCISE TAXES--EFFECTIlrE DATE 2/16/88 EXECUTIVE INCENTIVE PLAN ~EIP) FAILURE TO PUBLISH FILE FOLDER BEGINNING/ENDING PAGES COR~GE SHEETS) FILTER ADDITIlrE$ FILTER MA~rUFACTURING FILTER RESEARCH PE~QRMED COMPANIES ' BRA/~DS FIR.E-SAFE CIGARETTES FOP~EIGN I.~GUAGE DOC~$M"ENTS FTC STAFF REPORT FTC v. CARTER GAS PHASE CQMPONENT$ RESEARCH H)~NDWR I TTEN NOTES HEARING TRANSCRIPTS, CONGRESSIONAL RECORDS BY B&W ON OTHER STATEMENTS TO CONGRZSS, -4-

22 22 22 22 22 23 23 23 23 23 24 24 24 24 24 24 25 25 25 26 26 26 26 26 26 27 27 HISTOPATHOLOGICAL STUDIES HOFFMAN PAPER HONORARIA PROGRAM HUMAN CODES HUNTER COMMITTEE IMPROPER FOCUS OF CTR R~ESEARCH INDUSTRY COMPETITION INDUSTRy CQOpERATION INDUSTRY PERFORMANCE INDUSTRY RESEARCH REVIEW GROUP C1980-1982?~ INDUSTRY REVIEW OF GOVERNMENT RESEARCH INFOTA~ & QTHER INFORMATION-GATHERING SERVICES INHALATION STUDIES INSTITUTIONAL GRANT INTERNATIONAL CODES -- EFFECTIVE FEB. 24. 1988 INTERNATIONAL MA/LK~TING OF BARCLAY CIGARETTES INTERNATIONAL PROCEEDINGS & WARNINGS INTERNATIONAL TAR/NICOTINE DISCLOSURE IN V~VO v. IN V~TRO v. E~ VIVO JOINT FUNDING OF RESEARCH LAW DEPARTMENT ~UDGET LAWYER INVOLVEMENT LEAGUE TABLES [~EGISLATION ON ~ODUCTS LIABILITY --5--

27 27 27 27 28 29 29 29 29 29 29 3O 3O 30 31 31 31 31 32 32 32 32 32 32 33 33 LEGISLATION ON SMOKIN$ ~N THE WORKPLACE LOBBYING BY TI MANIPULATION OF RESEARCH/DATA MANUFACTURING COST ANALYSES OF THE ARIES FILTER MAR/CET I NG AGENCIES MARKETING POLICY CAH~ MAXWELL SEMINARS MOCK TRIAL DOCIrMENTS MOORGATE SECRECY PROTOCQL ~RFIT (MULTIPLE RISK FACTOR INTERVENTION TRIAL) N-BLOCTIN -- ADDITIVE FOR A "SAFER" CIGARETTE NON-CIGARETTE PRODUCTS {NCPs) PATENTS -- TRANSMITTAL MEMOS PESTICIDES/HERBICIDES PHARMACOLOGY CODES PHARM.ACQLOGY/~ SMOKING BE}i~VIOR POSITION PAPERS PRE-1964 PERSONNEL/POSITIONS ~NFOR~ITION PREMATURE DEATH PRICE SUPPORT PROGRAM ~OR TOBACCO --6-- & CA~

33 33 33 33 33 34 34 34 35 35 35 35 35 35 36 36 36 36 36 36 37 37 37 37 38 38 PROCEEDINGS RE WARNINGS PROCEEDINGS TO AMEND CIGARETTE LABELING ACTS PRODUCT DEVELOPMENT PRODUCT LIABILITY PRODUCT TEST SHOWING TAR. NICOTINE & MOISTUBE PRODUCT TESTING PRO-INDUSTRY PUBLIC RELATIONS DOCUMENTS BY TRADE ASSOCIATIONS PROJECT JANUS AND PROJECT EM~ PROJECT H~PPO PROJECT IvoRy PROJECT JANUS AND PROJECT ~ PROJECT LIMIT PROJECT MADHATTEB PROJECT PEGASUS PROJECT SHIP (SMOKE HARSHNESS IMPROVEMENT PROJECT) PSEUDO-PRODUCT TESTS PUBLIC DOCLDfENT$ RD&E BUDGET CODE -- EFFECTIVE DATE 3/2/88 RD&E MEETINGS RD&E PROJECTS CHRONOLOGY ~D&E REPORTS CHECKED OUT BY LAW DEPT. ~ECIPE$ AND SPECIFICATIONS ~EFERENC~S TO CRITICAL ~X3CUMENTS --7--

38 39 39 39 40 40 4O 40 40 41 41 41 41 41 42 42 42 43 43 43 43 43 44 44 44 REOUESTS FOR AND RESPONSES TO REQUESTS FOR PRODUC- TION OF SAMPLE CIGARETTES -- 3/10/88 FINAL DECISION RESEARCH DOCUMENTS RESEARCH FOCUS CODE -- EFFECTIV~ DATE 7/~9/~ RESEARCH FUNDING SOURCE RESEARCH TECKNIOUE$ RESEARCH TECKNIOUES ~"F" CODES) RESEARCH THAT WAS NEVER FOS,~DW~D UP UPON RESUMES OF R~SEA2CHERS ROPER STUDIES ROYAL COLLEGE OF PHYSICIANS "SAFER" CIGARETTES SENSITIVITY OF RESEARCH RE~ORTS SIGNIFICANCE SIGNIFICANCE--B&W/BAT RELATIONSHI~ SIGNIFICANCE -- FILE FOLDER PAGES SIGNIFICANCE -- LOBBYING SIGNIFICA/~CE -- SMOKING AND HEALTH RESEARCH SIGNIFICANCE -- SPECIAL ACCOVNT 4 SIGNIFICANCE -- TARGET MARKETS SMOKELESS TOBACCO ~ESEARCH TOBACCO COUNCIL, INC. SMOKE MODIFICATION $MOK-ER pERSONALITY STUDIES SMOKING INCIDENC~ SNOW W~ITE CONT~_OVERSY A$ APPLIED TO COUNCIL/SMOKELESS

44 44 44 45 45 45 45 47 48 48 48 49 49 49 49 5O 5O 5O 5O 5O 5O 51 51 51 51 51 SOCIAL SMOKING PROJECT SPECIAL ACCOUNTS STATE TAX HEARINGS STUDIES ON WHY PEOPLE SMOKE SUBSTANCES SURGEON GENERAL' S SWITCHING STUDIES TAR & NICOTINE LEGEND TARGET MA~RK~TS THEMATIC APPERCEPTION TEST (TAT) THORNTON PATENT TI -- ADVXRTISING TI -- B&W INVOLVEMENT TI -- COMMITTEE OF COUNSEL MEETINGS TI -- EXECUTIVE COMMITTEE TI -- INTELLIGENCE GATHERING TI -- LAWYER INVOLVEP[ENT TI -- PROPOSALS WRITTEN TO TI COMMITTEE OF COU-NS~L ASKING FOR FUNDING 0F ~LESEARC~ TI OUOTES BOOKS TOBACCO ACTION NETWORK {TAN~ TOBACCO COLLEGe/TObACCO ~ORUM (B&W) TOBACCO COLL~GE WASHINGTON ORIENTATION TOBACCO ~STITUTE TESTING LABORATOBy

52 52 52 52 52 52 53 53 53 53 54 54 56 56 56 56 56 56 56 57 TOBAC~Q PROCESSING BUDGETS TOBACCO WOR/(ING GROUP TOXICITY RESEARCH TRADE SECp, ET ¢0DE -- EFFECTIVE DATE 3/Z/88 TRAINING SEMINAR TRIP REPORTS TI~%NSM~TTAL NQTES & MEMORA24DA TP~AV~EL ITINERA~RIE$ & EXPENSE ACCOUNTS UNIV~ERS~T¥ OF KENTI~CKY TOBACCO & HEALTH RESEARCH trS, ~ONS UNRESO LV~ED QUESTIONS USE OF "QA" A~D "R" CODES FOR T_RCs TAC AND VdC V~RBAND DER CIGA/~.ETTENINDUSTRI~ VITALITY TESTS W.D, & H.O. W~LLS WHA.~TON STUDY WOODROSE PANEL WYATT TRAVELING PATHOLOGY 57 WYATT. WILSO~ 57 YEAM~, ADDISON 57 ~OUTH SMOKIN$ \laeu\vl.doc FELLOWSHIP I~ ENVIRONMENTAL -i0-

MASTER SUM~RY FOR B&W SUBJECTIVE DOCUMENT REVIEW As of March 31, 1989 ACUTE/CHRONIC INHALATION EXPOSU~$ T$STS If you cannot determine whether exposure was more or less than one day, make an educated guess and record the reasoning for your assumption in the comments. ADDICTION Addiction has become an increasingly important issue in tobacco litigation. In light of these developments, be careful to code all documents that mention words like "depen- dency," "habit" or "withdrawal" in connection with cigarette smoking to "DA; Addiction." The document does not have to include the word "addiction" to get the "DA" code. ADDITIVE-FREE CIGARETTES Project IVORY was to be an additive-free new product. If you find a document about an additive-free product, increase the significance and show the document to your supervisor. ADDITIVES (See also FILTER ADDITIVES.) When an additive is in code, please refer to the Additives Code Analogues list to code the appropriate substance. As you locate documents which "break" additive codes, please send such information to Atlanta to assist in updating the Additives Code Analogues list. Documents discussing additives testing will receive a "i" or "2" in significance. ADDITIV~S GUIDANC~ PANEL MEETINGS Documents concerning the role of the AGP should be coded to "ECBA; Guidance Panel." If a document includes the minutes of an AGP meeting it should also get a "BBC" (Scientific Meetings) code. If the document discusses the AGP's role in the testing of flavorings it would be double coded to "ECBA" and "ECBB; Testing Flavors." Send all AGP meeting minutes to Atlanta. ADVE~T$SING & MIR~ETING OF OTHER COMPANIES' BRANDS Do ~ot circle "AGB: Other Companies' Brands,, if the document merely mentions the brand(s) of another company, • ~., a taste test between Viceroy and Merit. The document

must actually discuss or relate to the advertising or marketing of that other brand. When B&W is tracking other companies' marketing and advertising efforts (e.~., P.O.P. displays, coupons), code to "AGB; Other Companies' Brands." Do not code all the indivi- dual activities of the other company. ALTERNATIVE CAUSATION Research that supports alternative causation is defen- sive, but you may often want to qz/estion in your comments the funding source for the research. Fully code documents relating to substances that are not in cigarettes, but which cause diseases. Do not circle the other substances code unless B&W was considering somehow using the substance in cigarettes. Include a comment such as, "Substance not in cigs, but B&W may be considering its use" (if applicable) or mention the substance in your comments as being another factor that causes or contributes to a disease. B&W began funding the AMA in 1964. Although the relationship officially ended in 1968, B&W continued to give money to the AMA until 1972 or 1973. AMES TEST When a document concedes that the Ames test is a viable test for mutagenesis and extrapolates the results to apply to humans, code to "MJ; Concessions." ANALYSIS & ?Q~M.ENTARY The analysis or commentary must have been performed by or on behalf of B&W or BAT and must relate to non-B&W and non-BAT work, research or publications. Code to "BBA: Analysis & Commentary" if a document discusses TAC (Peter Lee) reviewing government research. ARIES FILTER Research on the Aries filter or substantive discussions about the filter should be coded to "EBCC; Ventilated/ Diluted." -2-

Always circle "1008; Asbestos" whenever the document mentions asbestos. Continue to code documents discussing asbestos litigation to "M; LITIGATION ISSUES" and any other codes that apply. Mention in your comments that the document concerns asbestos litigation, synergy, asbestosis, etc. ASSURANCES OF SAFETY & CONCESSIONS An advertisement can be an assurance of safety, but not a concession. An internal document discussing an attempt to convince people that cigarettes are safe would be an assur- ance of safety and a concession, because it implies that the company feels that cigarettes are not safe, but is trying to persuade people otherwise. An internal document stating that cigarettes are safe is probably not a concession, but is an assurance of safety and should receive a significance of "i." A concession implies that the company or industry agrees with the negative claims about cigarettes. An assurance of safety is positive with regard to cigarettes. Documents regarding audits of Brown & Williamson's departments or activities should be coded to "C; B&W CORPORATE" in addltion to any other codes that may apply. AWARENESS OF HAZARDS OF $~D~STREAM SMOKE Double-code these documents to "B; AWA/~ENESS OF POTENTIAL HEALTH HAZARDS" (or one of the subcategories) and to "EJ; Environmental Tobacco Smoke." Link the two codes in your comments. H&W AWARENESS AND ACTION With the exception of Brown & Williamson's tracking of anti-smoking activities and research, only non-Brown & Williamson and non-BAT documents should be coded to "BB; B&W Awareness & Action." ~&W/BAT A~FILIATES Documents regarding these affiliates can be coded to "CC; B&W/BAT Relationship." Clarify in your comments that the involved entity is an affiliate. W.D. & H.O. Wills is sometimes a BAT affiliate and sometimes a Philip Morris affiliate. If Wills seems to be cooperating with BAT, assume that it is acting as a BAT affiliate. -3-

~&W/BAT MEETINGS Meetings between B&W and BATCO should be coded to "CCB; B&W/BAT Meetings," even if you are not positive that scien- tific issues were discussed. Code B&W/BAT RD&E meetings to "BBC; Attendance at Scientific Meetings or Conferences" (in its revised and expanded definition) and to "CCB; B&W/BAT Meetings." ~&W/DAT RELATIONSHip (See also B&W/BAT AFFILIATES, BAT BIOLOGICAL P~ESE~CH and SIGNIFICANCE--B&W/BAT RELATIONSHIP.) This code applies specifically to the relationship between B&W and BAT (BATCo) . Documents regarding BATUS will more than likely be coded to "CB; History and Organization" or "C; B&W CORPORATE." An othel-wise innocuous document which evidences the relationship could still receive a significance of "3." If CDC masters several documents together and none of those documents is an "extra," and if one of those documents evidences the B&W/BAT relationship, code all documents within the Master Id range to "CC; B&W/BAT Relationship." Do check to make sure that the documents should be actually mastered together. (Sometimes CDC will incorrectly master an entire file together.) If the master document is covered by one c~ding form (i.e., one large document with "extras"), only code "CC" to the pieces which evidence (on their face) the relationship. Of course, in this case, the large document will be coded to "CC" because it includes the extras. Correspondence between Jenson & Son (BAT's patent agent) and Kane, Dalsimer & Kane (B&W's patent counsel) should be coded to "CC; B&W/BAT Relationship." ~&W k~ENTUCKy DERBY FLOAT Documents discussing the float should be coded to "A; ADVERTISING/M~KETING" and "C; B&W CORPORATE." B&w'$ TESTING OF OTHER COMpAN~$' ~RANDS When B&W is conducting product tests on other companies' brands, code to the product testing ("AAC") code that reflects the methodology that B&W is employing for its tests. RD&E analysis of these cigarettes should be coded to "EH; Quality Control." -4-

~&w STOCKS Documents discussing B&W stocks should be coded to "C; B&W CORPORATE." B&W v. CBS FILES These files generally have a cover page that identifies the box and file numbers, describes the file contents and provides a date range. If this cover page is orange, it should receive a "O" significance code. If it is the white copy that follows the orange copy, it should receive an "NB" and should be coded to "AL; B&W v. Jacobsen." The date should be 820000/E, rather than the date range shown on the form. In addition, assign any codes that reflect information contained in the general description. BARCLAY CONTROVERSy Assign a significance of "2" if the document discusses the effects of the controversy, advertising of the filter, effects on sales, litigation issues or the status of cases. Laboratory research will probably receive a "3" in signifi- cance. Code documents that discuss the FTC's involvement in the controversy to "AM: Barclay Controversy" and whatever other codes apply. Do not code to "KEC; FTC v. B&W." Documents relatlng to the Barclay controversy will often be coded to "EFA; Compensation/Modification." ~AT BIOLOGICAL RESEARCH Code to "CC; B&W/BAT Relationship" when a copy of the report is distributed to B&W personnel and possibly to "MAB; [Confirmation of Causation] By BAT" when results or discus- sion warrants it. Do mot automatically code to "MBB; [Failure to Conduct Research] by BAT" and "MJ: Defensive Documents" unless the document specifically discusses (i) the need to conduct research due to the legal climate or (2) the responsibility of the company to test its products. DAT GROUP R&D CONFERENCES We now have the proceedings from the meetings held in 1962, 1967, 1968, 1969, 1970, 1972, 1974-1976, 1978-1984. Please continue to collect any of these documents that you may find and send a copy to Allen Creighton. -5-

B~T ~RR~T~G When BAT brands and their marketing are discussed, code to "AGC; BAT Brands" and any marketing codes that apply to those brands. In your comments, be sure to link which marketing codes apply to BAT brands and which apply to B&W brands. If there are B&W brands, be sure to indicate whether they are domestic or international. BAT SPECIAL CONFERENCES AND COMMITTEES King & Spalding has started a collection of documents which concern special BAT conferences and/or committees. The following three areas are of special interest: ao BAT Group Research & Development CoD~erences: These are held annually and started in 1962. We have the proceedings of the meetings held in 1967, 1968, 1969, 1970, 1976 and 1980. ¢~airman's Advisory Compittee ICAC): These conferences will probably be headed "Research Conference" and have extensive references to CAC and CAC companies. We have proceedings from these conferences for 1976, 1981 and 1983. These annual conferences are usually held in different countries each year. Co Bioloqical Resea?ch Committee: These meetings are normally held annually at GR&DC. We do not have any of the proceedings from these meetings. Please send a copy of any of these proceedings that you find to Allen Creighton. BATUS BOARD OF D~RECTORS' M~ET~NG$ Code documents discussing these meetings to "CE; Meet- ings." Always identify BATUS in your comments. BENOWITZ PAPER Dr. Neal Benowitz published cotinine research (e._~, Project TEMPER). One of his papers was entitled, "Smokers of Low Yield Cigarettes Do Not Consume Less Nicotine." BIOLOGICAL TESTING COMMITTEE MEET;NG MINUTES Send all these minutes to Atlanta. -6-

B LACKMAN LETTER This document should receive an "NB" because lawyers were involved, and it should receive a higher significance. The 9/25/84 letter is from Blackman to Owen, Boxall, Williams and Gaisch, and it transmits a note from G.A. Reed entitled "Summary of a Series of Smoke Component Retention Studies." The note concludes that it is futile to estimate tar consumption from measurements of nicotine/cotinine, NO or CO in blood, urine or breath. There is a memo from Lance Reynolds attacking the Blackman letter. BRAND HI$TOR~S FOR OTH$~ COMPAN~S' B~L~NDS Code to "AGB; Other Companies' Brands." BRAN~ P~ANAG~RS Documents that list brand managers for Brown & Williamson brands should be coded to "CB; History & Organiza- tion" and whichever brand marketing codes apply ( .~., "AC; Viceroy Marketing"). BRAND-SPECIFIC ADVERTiSiNG & MARKETING BUPGETS Code these documents to "AI; Marketing/Advertising Budget" and whichever brand Advertising/Marketing codes apply. DUSINESS BU~%DING T~ST The purpose of this test is to determine profitability/ share/image impact of heavy spending. These tests should generally be coded to "AAB; Marketplace Studies." CAC This acronym can stand for the Cigarette Advertising Code, the Chairman's Advisory Committee or the Cigarette Associated Companies. The latter two are related to BAT (BATI?). They are usually referenced in the context of research conferences and biological testing. If you see "CAC companies," the Chairman's Advisory Committee is probably the entity involved. CHAirMAN'S ADVISORY COMMIq~EE (CAC) This committee is part of BATUS, but B&W and BAT people attend the meetings. The committee tends to focus on S&H issues in society (e.__~, fire-safe cigarette, litigation). -7-

Code documents discussing CAC meetings to "CE; Meetings" and identify the Chairman's Advisory Committee in your comments. CHARTERED PATENT AGENTS Chartered patent agents and trademark registration agents should be treated as attorneys. These documents should receive either an "NA" or an "NB." CHEMO$OL Documents that discuss the use of chemosol (an additive in the late 1960s thought to reduce B(a)P delivery) should be coded to "EE; 'Safer' Cigarette." CIGARETTE ADVERTISING CODE~ INC. Documents regarding administrative and business activities of this organizatlon should be coded to "~HB; Cigarette Advertising Code" even if the documents do not specifically relate to marketing policy. Assigning this one code uniformly to Cigarette Advertising Code, Inc. documents will facilitate their later retrleval. CIGARETT~ MOUTHPI$CE Research regarding a cigarette mouthpiece should be coded to "EBA; Design Variables" and "EBC; Filter Research." Research regarding a ventilated mouthpiece should be coded to "EBA; Design Variables" and "EBCC; Ventilated/Diluted." CIGARSTT~ SALES Marketing studies showing industry-wide cigarette sales figures which also separately break out sales figures for Brown & Williamson should be coded to "AAE; Industry Performance" and to "CF: Sales/Profits/Pricing/Market Share." If Brown & Williamson's sales are not broken out, code only to "AAE: Industry Performance." CLEMENTS, ~ARL According to Larry Amos, Clements acted as a B&W attorney, but did not receive a law degree. References to Clements should be coded as if he were an attorney. CLOSING OF [j3UISVILL$ PLANT Documents discussing the closing of this plant should be coded to "CB: History & Organization." -8-

CoM~ENTS Please place "other substances," "other diseases" and "other cancers" at the beginning of the comments field to make it easier for the lawyers to find them and to make the comments more clear. Try to limit the number of significance "I" documents that do not receive comments. For those significance "i" documents that you do comment on, make sure your comment describes why the document is significant and does not simply contain the names of substances and diseases. It is not necessary to explain all codes in the comments field, e.~., lust say "BAT knew X was bad but did not conduct research" rather than "I coded to Failure to Conduct because BAT knew X was bad but did not conduct research." COMPENSATION & TOLEPANC~ Documents coded to "EFA; Compensation/Modification" will often be coded to "EDB; Tolerance" as well. Documents relating to the Barclay controversy are often coded to "EFA." COMPONENTS RESEARCH Code the following document types to "EA: Components Research" or an appropriate subcategory thereof: I. Whole smoke analysis: 2. analysis of tobacco leaf or stem; 3. analysis of specific constituent(s); and 4 o effects of smoke or smoke components on biological systems. If a substance that ends up in the particulate phase of smoke is being identified or measured in tobacco leaf, code to "EA; Components Research." Do not code to "EAA" unless the substance is being measured or identified in the smoke. CONCEPT-PRODUCT TESTING Whenever you have a concept test or an ad test in conjunction with product testing, code to "AACB; Concept- Product" testing. co~c~ss~oNs A document written by, cr perhaps quoting, an industry representative stating that an experlmental model relates to -9-

a human model or stating that the results are applicable to humans should be considered a concession, as well as a confirmation of causation. Concessions are limited to documents in which an industry representative's opinion or conclusion is stated. An advertisement can be an assurance of safety, but not a concession. An internal document discussing an attempt to convince people that cigarettes are safe would be an assur- ance of safety and a concession, because it implies that the company feels that cigarettes are not safe, but is trying to persuade people otherwlse. An internal document stating that cigarettes are safe is probably not a concession, but is an assurance of safety and should receive a significance of "i." A concession implies that the company or industry agrees with the negative claims about cigarettes. An assurance of safety is positlve with regard to cigarettes. If such a document concerns biological responses or conditions that are not diseases and extrapolates results to the development of disease, code to "MA; Confirmation of Causation" and "MI: Concessions." An example would be a company document reporting that smoke exposure produced an increase in goblet cell production, which might be a model for the development of bronchitis. CONF~DENTIA% ~ESEA~CH REPORTS The "confidential" status of a research report does not imply failure to publish or suppression of research results. Internal research reports are confidential or proprietary for competitive reasons. Research results which show real breakthroughs in the area of disease causation should probably have been published: if any such results are found in a confidential internal report, inquire in the comment whether the report was published and when appropriate, code to "MA; Confirmation of Causation" (and add appropriate comments). CONFIDENTIALITY AGREEMENTS Confidentiality agreements between B&W and other companies or individuals regarding joint research projects should be coded to "C" and whatever research codes apply. If the document does not identify the research project involved, code to "E." CONFI.~JkT~ON OF CAUSATION The "MA: Confirmation of Causation" codes are intended to capture documents suggestlng that research results confirm or supplement the alleged evldence that smoking causes -i0-

disease. The objective is to capture documents which plaintiff's counsel could use to show that anyone's, especially Brown & Williamson's or BAT's, research produced results confi~-ming that smoking causes disease. Research results should be briefly described in the comments field. These codes pertain to causation of disease, e.~., cancer, pulmonary disease and heart disease. Documents con- cerning production of hyperplasia, metaplasia or tumors via mouse skln painting, inhalation or other techniqn/es should be considered confirmation of causation. A document written by, or perhaps quoting, an industry representative stating that an experimental model relates to a human model or stating that the results are applicable to humans should be con- sidered a concession, as well as a confirmation of causation. Concessions are limited to documents in which an industry representative's opinion or conclusion is stated. Some confirmation of causation documents may have favorable points such as questlonlng the validity of the animal model or the results. These documents should be considered defensive. Confirmation of causation documents will gener- ally receive a significance of "i." The "MA: Confirmation of Causation" codes do not apply to documents discussing biological responses or conditions which are not dlseases in and of themselves. Smoke or constitutuent effects on ciliastasis, goblet cell production or ~n ~ mutagenesis will not be considered confirmation of causatlon. However, if a. company document regarding these biological responses also extrapolates results to the development of disease, then code to "MA; Confirmation of Causation" and "MI; Concessions." An example would be a document reporting that smoke exposure produced an increase in goblet cell production, which might be a model for the development of bronchltls. Documents coded to "MA: Confirmation of Causation" must be considered for coding to "ME; Failure to Publish/ Suppres- sion." Reserve coding to "ME; Failure to Publish" for final reports or other significant documents reporting major findings. "Failure to Publish" should be used conservatively in conjunction with "MA; Confirmation of Causation." Documents discussing decisions to publish research should, of course, be coded to "ME: Failure to Publish/Suppression." When the results of epidemiological studies confirm or supplement the alleged evidence that cigarette smoking causes disease, the document should receive an "MA; Confirma- tion of Causation" code. -Ii-

CONSUMER ~YNAMICS Consumer Dynamics is an assumed name that B&W sometimes used when performing marketing research (probably during or after 1980). CQNSUMPTIO~ Documents discussing consumption should be coded to either "EF; Human Smoking Behavlor" (if the focus is on individual smokers or per capita consumption) or "AAE; Industl-y Performance" (if the focus is on total nu/~ber of cigarettes sold or shipped). CONTROLLED STOR~ M][CROM~RKET T~ST Code documents discussing this test to "AA; Advertising/Marketing Research." If you find a document that describes this test, please send it to Lori Eubanks. CORPORATE M~$TING$ In addition to internal board meetings, "CE: Meetings" also includes corporate-type meetings outside of B&W that B&W personnel (usually executives or officers) attended. You would also code to "CE" a document discussing a trip by a BATI representative to B&W so he could become more familiar with the U.S. tobacco industry. Do not code documents to "CE; Meetings" which reflect corporate personnel/representa- tives' attendance at industry meetings pertaining to awareness of potential health hazards; these documents should be coded to "BBC: Attendance at Scientific Meetings or Conferences." Use a comment to identify the meetings. Code documents discussing BATUS Board meetings to "CE" and identify BATUS in your comments. CORpORAT$ PRODUCT DSVELOpMENT COM~M!TTEE MEST~NG MINUTES These minutes are not automatically considered to be proprietary unless they contain proprietary information. They should be coded to "BBC," which, in its expanded definition, contains all RD&E meetings. COTININ~ (NICOTINE M~TABOL~SM) R~SEABCH Code these documents to "EA: Components Research" and "FEB: Clinical," and clrcle "1031; cotinine" on the sub- stances list. These documents may also be coded to "FCD; Metabolism." -12-

COUNCIL FOR BURLEY TOBACCO Documents relating to this organization should receive a "QA" code. COUPONS DISTRIBUTED IN GROCERY STORES Code to "ABDB; Coupons" and to "ABDD; Retailers" because the activity took place in the retail environment. CT~ SPECIAL PROJECTS If you k~ow that a document reflects a C'fR Special Project, even though the document does not specifically refer to CTR, code to "GEA; CTR Special Projects." Include an explanation in your comments. CTR Special Projects began in 1966. Q~FENSIVE DOCUMENTS "Defensive" research reports (or abstracts) should focus on the disease causation theories and show either (I) inconclusive results (whlch support the industry's contention that research is inconclusive regardinq disease causation) or (2) no disease formation resulting from exposure to smoke or smoke constituents. These reports can be either internal or external. Documents discussing supportive views held by doctors, pathologists and other experts may be defensive, but always question in your comments the affiliation of or financial support behind these persons. A person's testimony on behalf of the industry may be defensive. If the person was funded through Special Account 4, be sure to circle "GEE; Special Account 4." If you cannot determine how the person was funded, you may want to question in your coml~ents the funding or financial backing for that person. DESIGN OF T/N LEGEND Code any documents discussing design of the T/N legend to "KB; Design." Be sure to include "T/N" in your comments field so the user knows that the document concerns the T/N legend, rather than the warning. DILUTION OF WARNINGS If a document discusses the industry diluting warnings, code to "KD; Dilution of Warnings," since B&W, as a part of -13-

the industry, is going to be held responsible for the industry's activities. Some of these documents may also be coded to "M~; Assurances of Safety." DISEASES & CONDITIONS (See also DISTINCTION BETWEEN MUTAGENESIS & CELLULAR CH~dqGES.) Code all mentions of diseases and conditions to the appropriate "D" code(s) except when mentioned in an actual warning label. When "DDC; Other Cancer" is coded, list the type of cancer in the comment. When "DG; Other Diseases or Unspecified Health Consequences" is coded, list the condition in the comment or if unspecified, write "unspecified disease" or "unspecified health consequence" in the comment. Code "death" to "DG; Other Diseases" and add "death" in the comment. Use a comment to explain any disease coded which does not allegedly relate to smoking. Code tuberculosis to "DB; COPD" (even though it is a bacterial infection). Code pneumonia to "DG; Other D~seases." Asbestosis should receive the "DB; COPD" code. References to thoracic cancer (thorax) should be coded to "DDC; Other Cancer." Code references to cancer of the windpipe, throat or vocal cords to "DDB; Laryngeal." Collect all references to laryngeal tumors, lesions cellular changes and cancer; Laura Owens is interested in the documents, especially those related to disease causation. When a document states that a substance is carcinogenic, code the substance and "DD; Cancer." Even though the document does not discuss cancer specifically, it implies that the substance is capable of causing cancer. Statements alleging that smoking is associated with lung cancer and other cancers should be coded to "DD; Cancer" and "DDA: Lung." Do not code to "DDC; Other Cancer" because the "other cancers" may include laryngeal cancer. Code documents discussing diseases or conditions of the female reproductive system to "DF; Pregnancy/Fetal." Discussions of early menopause should receive this code. Statements alleging that smoking causes premature deaths should be coded to "JB; Risks" and "DG; Other Diseases," in addltion to whatever other codes might apply. "DO; Respiratory" is for items such as changes in goblet cell formation and ci!iastasis. These two conditions do not -14-

indicate confirmation of causation. "MA; Confirmation of Causation" is rese~-ved for actual diseases. Code to "MI; Concessions" (or "MIA" or "MIB") if an industry document implies that ciliastasis or goblet cell formation can occur in humans (i.e., the document extrapolates from animals to humans). If the document does not extrapolate to h~umans, it is not a concession. Studies focusing on a genetic predisposition to cancer should be coded to "DDE; Mutagenesis." DISTINCTION BETWEEN "IC; B&W INVOLVEMENT" AND "ID: !NDEPgN- DSNC~/COMPANIES" "IC" is fairly benign. Code here items such as B&W's monitoring of CTR's actlvltles. "ID" is more sinister. Use it to show that CTR is not Independent, that the companies stacked the SAB, that the companies did not approve certain types of research or that the companies controlled the issuance of grants. DISTTNCTION BETWEEN "~D; INDEPENDENC~,/COMPANI~S" AND "IE; INDEPENDENC$/LAWYSRS" Include in-house counsel under "ID," and code outside counsel to "IE" (e.__~., Pepples' involvement will always be coded to "ID" and never to "IE."). This distinction applies only to the CTR codes; it does not apply to the TI codes. DISTINCTION BETW~$N "KA; ADEOUACy" AND "KC; ~FF~CT" "Adequacy" focuses on the warning itself. It includes studies performed on various types of warnings ( .~., arrow studies) and comparisons of warnings. An adequacy study will probably be relatively short term. "Effect" is more person-oriented (Did the warning make people stop smoking?). There will probably be quite a bit of overlap between the two codes. DISTINCTION BETWEEN MUTAGENESIS & CELDULAR CHANGES Mutagenesis is a permanent cellular change that involves DNA (genetic material). Cellular changes usually involve hyperplasia, metaplasia or goblet cell formation (cellular changes in a tissue). Mutagenesis is not always associated with cancer, but cancer results from mutagenesis. DISTINCTION BETWEEN TOBACCO ~=~=~ ........ ~--~ AND TOBACCO PROCESSING Tobacco treatment ("EBBB") should be coded when the purpose of the research or technique is to modify the nature of the smoke. Tobacco processing ("EG") should be coded when -15-

the purpose of the processing technique is to improve the smokeability or quality of the smoke. DOCUMENT RETENTION/DESTRUCT!ON Code press queries that include the statement "Reten- tion: 2 Years" to "CA; Document Retention/Destruction." Code documents discussing document retention/dest~-uction policies for non-B&W entities to "CA; Document Retention/ Destruction" and link the code to the appropriate entity in your comments. A TI document retention policy would also be coded to "H; TOBACCO INSTITUTE." DONATIONS ~Y ~&W TO CHARITABLE $VKNT$ & ORGANIZATIONS Always write "SHOWS B&W AS A GOOD CORPORATE CITIZEN" in the comments field; code to "C; B&W CORPORATE." Duolite is a chemical resin that selectively removes components such as acrolein and formaldehyde from tobacco smoke. Documents that substantively discuss the Duolite or Purite filter should be coded to "EBCB; Selective Filtra- tlon." OuPONT TASTE PANEL The DuPont facility is an "anonymous" B&W-funded center for sensol-y evaluation research which provides feedback to the Product Division of RD&E by obtaining consumer reactions to new flavors, processes and designs. This information is used in improving present products and developing new ones. A typical study, for example, attempts to determine the acceptability and optimum level of pipe tobacco-type aromatic flavors in cigarettes (as a way of improving sidestream). Another explores the advantages of a new filter material. A typical large panel contains I00 cigarette smokers recruited from the general population. Panelists come in once a day in groups of 15-30 for the duration of the study (typically two weeks). Typical small panels contain 50 panelists, divided into groups of 8 to 12. About one-half of the DuPont panels are large and the others are small. Code documents to "ECAC: Taste Panel" if a cigarette with a modified or new additive is being tested. Do not code to "ECAC" if the filter is being tested. Code to "AAC; Product Testing" or a product testing subcategory if marketing is involved. Add a comment if the testing is very preliminary and you circled a product testing code. -16-

Dupont smokers appear to be consumers who have been ~ to make distinctions between different cigarette characteristics. Use the taste panel code ("ECAC") when "in- house" or expert smokers are involved, primarily in the testing of additives. Code product testing by the Dupont smoker panel to the product testing code ("~%C") . Taste panel documents receive a funding source code if the focus is on product development. They should not receive a funding source code if the focus is on tobacco processing or QC. ELLIS PATENT/PROJECT ARIEL Project Ariel deals with nicotine pharmacology and research for nicotine analogues or substitutes (an alterna- tive design cigarette prototype). Key terminology to look for is: Ellis patent; Ellis/Hughes: Secret Southampton Project; Ellis smoking device: Project Ariel; the Battelle patent; Ellis' tar!ess cigarette; the Special Ellis Project; and the Secret Smoking Device. Key persons in the 1960s (June 1961-January 1968) were Ravlin, Yeaman, Blalock and Griffith. Key persons in the 1970s and 1980s were Parrack, Reynolds, Litzinger (1983) and I.W. Hughes. Herbert Schachner and David Williamson were the inventors of the device in the Ellis patent. Collect all references to Project ARIEL and Sir Charles Ellis's patent and forward them to Dan Willoughby as soon as possible. SMP$OYE$ INDOCTRINATION All references to B&W indoctrination/orientation of its employees to the "industry point of view" on smoking and health issues should be brought to the attention of your supervisor; Dan Willoughby is collecting this information. Look for any references to the tobacco college or the tobacco forum. Sp!DEMIOLOG!CAL STUDIES When the results of epidemiological studies confirm or supplement the alleged evidence that cigarette smoking -17-

causes disease, the document should receive an "MA; Confirma- tion of Causation" code. ~T$ ADVISORY COMMITTEE ~NAT~ONAL ETS ADVISORY COMMITTEE) The National ETS Advisory Committee addressed issues concerning ETS. Its members were attorneys, company repre- sentatives and TI representatives. It is not a TI or CTR committee, and therefore, references to this committee should not receive an "H" code. In some cases, these documents may be coded to "GEA; CTR Special Projects" or "GEE; Special Account 4." Code to "EJ; Environmental Tobacco Smoke" and "NB" or possibly "NA." ETS CLAIMS When a B&W or other industry employee claims that ETS is not harmful, code "MHA: B&W - Assurances of Safety" or "M~; Assurances of Safety," in addition to "EJ; Environmental Tobacco Smoke." EXCISE TAX~$--~[FECTIVg DATE 2/16/88 The proposal for an excise tax code has been approved. The new code reads as follows: CFA Excise Taxes: Documents discussing or concerning the taxes that federal or state governments levy on cigarette products. Tariffs are not excise taxes and should not receive the "CFA" code. EXECUTIVE INCENTIVE pLAN (EIP) In 1983, TI's Executive Committee met and decided that TIPAC needed more money. The executives returned to Kheir companies to try to get eligible employees to contribute to the company's PAC and TIPAC. In the case of B&W, employees were told that a percentage of their contribution to BACPAC would be earmarked for TIPAC. These documents should be coded to both "HB; Lobbying" and either "L: LOBBYING" or "LA; B&W Lobbying." [AILUR$ TO ~UBLISH The fact that a research report is confidential does not imply that the company failed to publish. If a confidential report seems to indicate confirmation of causation, include a ccmment such as, "IS THIS CONFIRMATION OF CAUSATION & DID WE FAIL TO PUBLISH THIS?" -18-

Documents coded to "MA; Confirmation of Causation" must be considered for coding to "ME; Failure to Publish/Suppres- sion." Reserve coding to "ME" for final reports or other significant documents reporting major findings. "ME; Failure to Publish" should be used conservatively in conjunction with "MA; Confirmation of Causation." Documents discussing decisions to publish research should, of course, be coded to "ME; Failure to Publish/Suppression." FILE FOLDER BEGINNING/ENDING PAGES ~ORA.~GE SHEETS~ All orange file folder pages should receive a signifi- cance of "O." If a duplicate white file folder page follows, ~t may be coded to "NB; Other Legal Reference" and receive a significance of "3" if an attorney's name appears on the folder. Check the CDC printout to b__e sure that a "File Folder Name" has been entered for the beginning page; if it has not, please annotate the printout in red to include this informa- tion, i.e., "FILE FOLDER NAME: (full title of file)." The file folder name will then probably have to be added to the printout for each document w~thin that file. With regard to documents that CDC objectively coded during the winter of 1988, there are no file titles. Do not write in file names for these documents, except in the case when CDC did not code the file folder beginning page at all. If CDC missed this page, write the objective codes on your printout and add the file folder name to each document in that file, including the file folder ending page. FILTER ADDITIVES Do not code filter additives as other substances or as additives unless the filter additive ends up in cigarette smoke. Code as an other substance and as an additive if you do not know whether it ends up in the smoke. A change in the taste or smell of the smoke is an indication that the substance is probably going through. Polyethyleneimine (PEI) is a filter additive and was used in the Duolite filter. PEI selectively removes HCN and aldehydes. However, do not code "EBCB; Selective Filtration" unless a document discusses PEI in the context of selective filtration. Acids that are added to filters should generally not be coded on the substances llst because acids are used to remove bases, such as nicotine, from the smoke. -19-

According to a patent found in Houston, zinc, cadmium and mercury can pass through the filter when they are used as filter additives. Since these substances then can be inhaled by the smoker, code them to "EC; Additives" and to "1049; Metals" when they are used in the filter, in addition to whatever other codes apply. FILTER MANUFACTURING A document that relates solely to the manufacturing of filters would be coded to "EG; Tobacco Processing," not "EBC; Filter Research". FILTER RESEARCH PERFORMED By ~&w ON OTHER COMPANIES' BRANDS Code this research to "EBC; Filter Research" or to one of the filter research subcategories. [IR~-SAFE CIGARETTES Until March ii, the procedure was to code these docu- ments to "EE; Safer Cigarette" and put "FIRE-SAFE" in the comments. If the document discussed trying to develop a fire-safe cigarette, we coded to "EI; Product Development" and "EE; Safer Cigarette." In the comments, we linked the codes. A similar procedure applied to "KN; Other Proceed- ings" and "JB; Risks." A self-extinguishlng cigarette code was introduced on March II, 1988. The new code reads as follows: EIC Se!f-Extinqu~sbing Ciga?ette: Any aspect of development, production or quality control of self-extinguishing cigarettes. Do not double code to this category. For example, if you have proceedings about a fire-safe cigarette, code to "EIC" only. If you have a document solely concerning lobbying with regard to fire-safe cigarettes, code to "EIC: Self- Extinguishing Cigarette" and assign a significance of "3." The only items you will double-code with "EIC" are issues involved in the current litigation (e.__~g_~, B&W/BAT relation- ship or industry conspiracy); in these cases, the signifi- cance level may increase. FOREIGN ~ANGUAGE DOCUMENTS When you come across a document written in a foreign language (with no apparent English translation]: i) put a yellow sticker on the document, 2) record the document range -20-

and language-type on a community list, 3) assign a signifi- cance of "2" (unless enough information is available to give it a higher significance) and 4) make a comment in the comments field about the language or what you believe the document pertains to if there is incomplete information in the objective data. If you are able to gleen any information from the document, circle the appropriate codes. Try to give the proper lawyer code by using the CDC forms or looking for a lawyer's name in the document. Free was a tobacco-free cigarette designed by Interna- tional Brands, Inc. Corn shucks and cocoa shells, among other substances, were to be the filler. For financial reasons, the project was not pursued. Documents discussing Free will generally be coded to "EBBA; Tobacco S%~bstitutes." FTC STAFF REPORT The Staff report of May 1981 should be coded to "KEA; FTC v. Carter." FTC v. CARTER On May 12, 1976, the FTC subpoenaed the six major cigarette manufacturers, 20 ad agencies, TI and CTR. In January 1979, the Federal District Court enforced the subpoenas against all respondents except TI and CTR. GAS pHA$~ COMPONENTS R$S~A~CH Code to "EAB; Gas-Phase" a study demonstrating that carbon monoxide levels increase as a cigarette is smoked. GOR~ Gio Gori was with the Tobacco Working Group (TWG) when he published parameters for a safer type cigarette. These new parameters included lowered tar, nicotine, carbon monoxide, etc. He alienated the government because it felt that he should have stated that all cigarettes were bad, but Gori stated that a "safer cigarette" could be made. He then went to the Franklin Institute where he completed several research projects for CTR and worked separately for B&W. After leaving the Franklin Institute, he worked for B&W, primarily in the area of additives. Several BAT research projects were named by using the letters in Gori's name (e.c., IGOR, VIGOR, etc.). -21-

HANOWRITTEN NOTES Pay particular attention to handwritten documents. These papers are typically rough drafts which have not been carefully worded, and they often contain sensitive informa- tion. It is also easy for a plaintiff's attorney to pick these out of stacks of typewritten documents. HEARING TRANSCRIPTS, STATeMeNTS TO CONGRESS, CONGRESSIONAL If B&W kept a statement, it is important enough to fully code. All transcripts and Congressional records should be fully coded. A statement by a B&W spokesperson will definitely receive a "i" in significance. A statement by a TI representative will probably be a "I." HISTOPATHOLOGICAL STUDIES Histopathological studles can be "FEA; Autopsy" or "FEB; Clinical." If you cannot tell whether the tissue is from a living person (e._~., bronchoscopy) or a corpse, code to "FE; Human." HOF~.MAN PAP~E Hoffman wanted to show that a cigarette with a special filter (Barclay?) had higher deliveries when smoked as smokers actually smoke. It appears that someone convinced Hoffman to not publish his paper. BAT/Hamburg and B&W (Gio Gori and Lance Reynolds) were involved. Hoffman may have sent his report or an abstract of his report to the Bureau of Alcohol, Tobacco & Firearms. This matter occurred around 1983. Documents relating to this paper should receive a higher significance. HOnORARiA pROGRAM The Honoraria Program was to fund senators to speak at BACPAC meetings. It was administered through the TI State Activities budget. Documents discussing the program should receive a lobbying code and a significance of "i." HUBER Pay attention to research performed by Dr. Huber. He was charged with mismanagement of funds and fabrication of research data. HL~IN CODES (See RESEARCH TECHNIQUES ("F" CODES) AS APPLIED TO HI/MANS.) -22-

HUNTER COMMITTEE References to the work of this committee should be coded to "EC; Additives', or "ECBB; Testing Flavors" and "ECBC; Testing Others." Do not code to "ECC; HHS/HEW Disclosures." ~ IMPROPER FOCUS OF CTR RESEARCH Documents evidencing an improper research focus will generally be double-coded to "IG; Research Focus" and "MB; Failure to Conduct Research." INDUSTRY COMPETITIQN Documents coded to "MK; Industry Competition" should discuss smoklng and health issues in relation to research. Specifically look for any statement that research was performed in order to get a competitive edge or anything that implies that research was secondary to marketing. A document that suggests that a product should be designed to (subcon- sclously) cause a smoker to adjust his smoking habits (e.q., compensate) to obtain greater delivery of nicotine is an example. Another example would be a document that discusses not removing an allegedly carcinogenic substance from the product because removal would adversely affect the product's taste, thus giving other unaltered (better tasting) products a competitive edge. Do not code product improvement for purposes of normal competition here. Unfortunate statements not related to smoking and health should not be coded to "MK." SNDUSTRY COOPERATION These documents must show evidence of the industry working together and must contain statements that could be misconstrued as conspiracy. Some TI meeting minutes may warrant the "ML: Industry Cooperation" code, but be careful not to put all TI meetings here. TI is the industry's lobbylng arm, so the companies are expected to meet and work together. INDUSTRY PERFORMANCE Code reports on and reviews of the international cigarette industry to "AAE; Industry Performance and "p: INTERNATIONAL.', Add clarification in your comments field. Large quantities of SOM data for a non-B&W brand should be coded to "AAE; Industry Performance." -23-

Include under this code any documents that substan- tively discuss market share of other tobacco companies. Also include Maxwell documents discussing stock performance and documents discussing Maxwell seminars. INDUSTRY RESEARCH REVIEW GROUP (1980-19827) Pepples was chairman of this group in 1982. The group received research funding requests for CTR Special Projects and Special Account 4 projects. Documents involving this group will generally be coded to "GEA; CTR Special Projects" or "GEE; Special Account 4," "IC; B&W Involvement" when B&W is involved, "ID; Independence/Companies" when in-house counsel is involved, "IE; Independence/Lawyers" when outside lawyers are involved and possibly "ML; Industry Coopera- tlOn." ~NDUSTRY RSVlEW OF GOVERNMENT RESEARCH Code to "BBA; Analysis & Commentary" if a document discusses TAC (Peter Lee) reviewing government research. ~NFOTAB & OTHSR INFORMATION-GATHERING SERVICES If the text of an Infotab document relates to interna- tional non-RD&E issues, assign the "P" code. If Infotab is monitoring domestic information only (e._~., a summary of TI's activities in regard to the Surgeon General's report), do not assign the "P" code. In other words, the "P" code is based on the material being gathered, not the group collecting the material. INHALATION STUDIES In addition to coding to the appropriate "inhalation" codes, remember to code to the appropriate "In Vivo" codes because an inhalation test is also in v~vo. This will aid the attorneys in their search for specific research docu- ments; they may not be aware that inhalation work is in v__~y_Q. INS,~TU~ONAL GPANT An institutional grant can be made to either a univer- sity or an institution. INTERBAT As of April 1985, the flow of information to the U.S.A. was restricted. B&W could input as much information as it wanted. Look for any documents that describe what type of material was restricted. -24-

If a document discusses the use of STABFIL (Science/ Technology Abstract Bulletin and Filter Data Base), it should be coded to "CCA; INTERBAT." STABFIL was par~ of INTERBAT. INTERNATIONAL CODES -- EFFECTIVE FEB. 24. 1988 The international codes are: AGC ~AT ~?aDds: Documents that relate to the adver- tising and marketing of BAT cigarettes; and INTERNATIONAL: Documents discussing B&W's activi- ties in producing and marketing its or other companies' brands for an international market. Quality control on foreign-marketed cigarettes and product development for a foreign country will receive the "P" code, in addition to the appropriate QC or product development codes. Whenever you code to "P," include a comment on which portions of the document or which codes relate to international activities. Documents from or regarding a foreign country will not necessarily receive an international code (i.e., do not code all BAT documents to "P" or "AGC"). The "P" code can De used for B&W's international marketing activities (and production work for international marketing), another company's international marketing activities, the league tables, international regulations, international lobbying and international proceedings. It should oD_Q_q be used for research (other than production work) or for funding sources. INTERNATIONAL MARKETING OF ~A~CLAY CIGARETTES Even though in several European countries Barclay was manufactured by BAT, rather than by B&W, for the sake of consistency and simpler searching, double-code documents discussing the international marketing of Barclay cigarettes to "AF; Barclay Advertising/Marketing" and "P; INTERNATIONAL." INTERNATIONAL PROCEEDINGS & W~RN%NGS Documents discussing failure of the Swedish warnings should be coded to "K; WARNINGS/PROCEEDINGS" and "P; INTERNATIONAL." Proceedings regarding the Swedish warnings should be coded to "K; WARNINGS/PROCEEDINGS," "K]q; Other Proceedings" and "P; INTERNATIONAL." -25-

INTERNATIONAL TAR/N~¢QTINE DISCLOSURE If B&W decides to voluntarily show amounts of tar and nicotine on packs marketed internationally, code to "AHA; B&W Policy" and "P; INTERNATIONAL" rather than to "KL; T&N Disclosure Proceedings." The "KL" code is limited to a specific time frame and a specific set of activities. IN vlvo v. ~N V~TRO v. EX VIVO According to distinctions drawn by the industry, i~ v~tro research usually pertains to exposing a DOD-~ammal (usually a living microscopic animal or clam) to a test substance to ascertain any changes in its living state. In vivo research usually pertains to exposing a ~ammal to a test substance to ascertain any changes in its living state. Ex vivo research pertains to tissue or organs removed from a llvlng animal for exposure to a test substance. This test is usually done on large, mammalian animals. When trying to distinguish between in vivo and ~ vivo, base your decision upon the time of exposure. For example, if a dog was exposed to smoke and then killed, the research would be i~ vivo because the dog was alive at the time of exposure. JACOBS, SD Jacobs was a lobbyist around 1966 and 1967. He sometimes worked for B&W and at other times worked for TI. When a document reflects Jacobs' involvement in lobbying, try to deterznine the company on whose behalf he was lobbying. jOiNT FUNDING OF RESEARCH If B&W and another organization are funding research together, circle both funding source codes and write, "JOINT RESEARCH WITH " in your comments. LAW DEPARTMENT bUDGET A budget for B&W's Law Department should be coded to "C; CORPORATE." LAWV=R !.~£0 LVZMENT "NA" can contain material from a lawyer acting in his legal capacity and material to a lawyer asking for legal advice. "NB" pertains to documents which are copies to a lawyer for his information and monitoring and to documents reflecting a lawyer acting In a non-legal/business capacity. "NC; No Legal Reference" pertains to documents which bear no lawyer names, contaln no legal advice and were not prepared by or for a lawyer in antlclpatlon of litigation. -26-

When you have a transmittal memo from an attorney transmitting another document (including agendas and public documents), the attachment will receive either an "NA" or "NB," even though CDC may already have the attachment mastered together with the transmittal memo. The attachment is a communication from an attorney. When a document includes an attorney's name and you know that the document is not privileged, assign an "NB" and include a brief comment stating that the document is not privileged (e.~., documents from attorneys representing other manufacturers in the Barclay controversy). LEAGUE TABLES These tables relate to T/N tracking and are compiled in Europe. They are used primarily by the governments in determining the amount of tax to levy upon various cigarette brands. These documents will be coded to "P; INTERNATIONAL" and to "EHB; T&N Tracking." ~EGISLAT~ON ON PRODUCTS ~IAB~L!TY Documents discussing products liability legislation (e.~., tort reform) should be coded to "M," as well as to "~N; Other Proceedings." LEGIS~LATION ON SMOKING IN ~ = .H_ WORKPiAC~ & CAB REGUlATiONS Code these documents to "EJ: Environmental Tobacco Smoke" and "KN; Other Proceedlngs." The document may require a lobbying code if the industry is opposing the regulation or legislation. (See PROJECT LIMIT) LITIGATION Documents discussing relevant litigation should receive the "M: LITIGATION ISSUES" code rather than the "K~: Other Proceedings" code. LOBBYING All lobbying documents will receive a significance of "I." Even documents discusslng indirect lobbying will be coded to the lobbying code and will receive a significance of -27-

The tracking of legislative activity is not necessarily lobbying. Look for direct contact by B&W or BATUS with Congressmen. If BAT is making a lobbying-type presentation to members of Parliament, code to "L; LOBBYING" and "P; INTERNATIONAL." The document will receive a significance of "I." Clarify your codes in the comments. Lobbying is a voluntary actlvity. When someone for the tobacco company is required to testify before a governmental body, that activity is not lobbying. Lobbying is less formal and is similar to PR. Lobbying includes letter"writing campaigns and other methods of influencing Congressmen. Err in favor of assigning a proceedings code if you cannot tell whether a document relates to lobbying or to a proceeding (i.e., if you cannot tell whether someone is speaking voluntarily or involuntarily). If a document discusses or contains testimony of an industry representatlve speaking at a proceeding, it will receive a significance of "l" unless the issues being discussed are not related to smoking and health (e.~., fire-safe cigarettes). (Exclse taxes are related to smoking and health, because taxes influence the price of cigarettes. The price affects the ability of young people to buy cigarettes.) Lobbying to health organizations (e._~., ACS) can receive a lobbying code, but be sure to explain in your comments that the lobbying is not directed toward a govern- mental body. Industry grass-roots activities such as what occurred in the Dade County and San Francisco public smoking referendums should receive the appropriate lobbying code(s) and a significance of "i." Due to some confusion as to the specific definition of lobbying, not all these documents have received a lobbying code in the past. Susan Yeager will include information about how these other documents can be found in the user subjective search manual that she is preparing. ~OBBY[NG BY TI TI lobbying does not have to be double coded to "LA; B&W Lobbying." You may need zo use "HA; B&W Involvement," depending upon the specific document. -28-

MANIPULATION OF RESEARCH/DATA Any documents suggesting that the cigarette companies made recommendations to research organizations about how research projects should be performed (i.e., procedural recommendations) should be coded to "MD; Manipulation of Research/Data" or a subcategory of "MD." ~MANUFACTUR!NG COST ANALYSES OF THE ARIES FILTER Code these documents to "EG; Tobacco Processing." MARKETING AGENCIES Documents that list marketing agencies for Brown & Wllliamson brands should receive whichever brand marketing codes apply. MAP KETING PO[~I. CY Do not use this code simply because a document expresses dlssatisfaction with an agency's lack of adherence to B&W's standard procedure. A policy is generally a straight- forward, before-the-fact type of statement. If you have a question about a document, ask your supervisor. MA~ELL SEMINARS Code documents discussing Maxwell seminars to "AAE; Industry Performance." Indicate attendance of non-industry consultants in your comments field. MEETINGS Code documents discussing RD&E meetings to "BBC; Attendance at Scientific Meetings or Research Conferences." All B&W RD&E formal meetings, even meetings of the Product Development Committee and the Leaf Availability Committee, will receive this code. Be sure to code to "BBC: Attendance at Scientific Meetings or Conferences" when B&W personnel go to a research site to view research or research results. Be careful when distinguishing between "BBC; Attendance at Scientific Meetings or Conferences" and "CE; Meetings." Consider the topics that are discussed at the meeting and the prlncipals involved. For example, Corporate Product Development Committee meetings are attended by representa- tives from Marketing, RD&E and Law. Topics discussed include product development, product testing and manufacturing. Even though this group is a formal, organized committee, its meetings should be coded to "BBC" because the primary focus -29-

of the meeting is RD&E-related; corporate issues are not addressed. "CE" should be used for Board meetings, Corporate Planning Committee meetings, Executive Committee meetings and any meetings of an organized committee relating to corpor&te matters. Meetings of formal marketing committees should also be coded to "CE." Informal meetings, such as those between marketing personnel and ad agencies, would not be coded to "CE." "CE; Meetings" includes corporate-type meetings outside of B&W that B&W personnel (usually executives or officers) attended. You would also code to "CE" a document discussing a trip by a BATI representative to B&W so he could become more familiar with the U.S. tobacco industry. Code documents discussing BATUS Board meetings to "CE; Meetings"; be sure to identify BATUS in your comments. Do ~ code documents to "CE; Meetings" which reflect corporate personnel/representatives' attendance at industry meetings pertaining to awareness of potential health hazards; these documents should be coded to "BBC; Attendance at Scientific Meetings or Conferences." Use a comment to identify the meetings. Meetings between B&W and BATCO should be coded to "CCB; B&W/BAT Meetings," even If you are not positive that scien- tific issues were discussed. TRIAL DOCUMENTS Documents regarding mock trials should be coded to LITIGATION ISSUES" (if litigation issues are discussed) and "MLB; Litigation Defense" (if various members of the industry participated). MOORGATE SECR$¢Y pROTOCOL This protocol may be the agreement that B&W had with companies whose brands B&W exported. Do not automatically code documents stamped with a reference to this protocol as proprietary. M~FIT (..~ULTTpLE RISK FACTOR INTIRVENTTON TRIAL) MIRFIT was a ten-year study published in the 1982 J__~. The researchers randomly assigned 12,000 men at high risk for heart disease into either a special intervention (SI) group or a usual care (UC) group. The SI group w~s found to have a reduction in risk factors, but the mortality rate was not statistically different between the two groups. The results did not show a beneficlal effect on coronary heart disease or total mortality of intervention. The researchers manipulated -]0-

data to explain this result and said that those men who had stopped smoking had less risk of heart disease. Burson Marsteller (a PR or ad agency) designed an ad campaign relating to MRFIT for TI in late 1982 and early 1983. The industry, including B&W, was concerned about this proposed campaign. The campaign was not actually run. N-BLOCTIN -- ADDITIVE FOR A "SAFER" CIGARETTE N-bloctin is an additive which apparently lowers the risk of smoking-associated lung cancer by neutralizing the action of nitrosamines, a major group of carcinogens found in both mainstream and sidestream smoke. A company called C.A. Blockers, run by L. Douglas Keeney, has announced its intention to produce a cigarette with N-bloctin. Mr. Keeney is assisted by Mr. Parrack; both are former B&W employees. Any reference to N-bloctin should be coded to "EC; Additives," "EE; Safer Cigarette," "EBBB; Tobacco Treatment" and "1084; Other Substances," along with any other applicable codes, and given a high significance. Bring any such documents to the attention of your supervisor. NON-CrGARZTT$ PRODUCTS (NCPs) When you encounter a document dealing with NCPs (pipe tobacco, chewing tobacco, snuff, roll-your-own), circle "EK; Cigars/Smokeless Tobacco" and any other codes that apply to Zhat document. Explain in the comments field which of the codes apply to the NCP only (if your document has both cigarette and NCP information) or state that all the codes apply to NCPs only (if your document contains no information on cigarettes). The "AG; Other Brands Advertising/Marketing" sub- categories can be used for the marketing of non-cigarette products, together with "EK; Cigars/Smokeless Tobacco." Be sure to link the two codes in your comments. PANORA.~ All Panorama documents should receive a significance of PATENTS Patents should not recelve a research funding source code. -31-

Drafts of patents should receive an "NB" because attorneys authored the drafts. PATENTS -- TRANSMITTAL MEMOS For transmittal memos and miscellaneous documents that refer to a patent by number, assign the most relevant code that was assigned to the patent. It is unnecessary to repeat every] minor code that was assigned to the referenced patent (e._~., a filter research patent may have a brief section relating to processing; it is not necessary to use the processing code on every piece of correspondence relating to that patent). P~STICIDES/HERBI¢IDE$ All pesticides or herbicides that are mentioned in a document should be listed in the comments field as other substances. Be sure to make it clear that these substances are pesticides or herbicides. PHAP~MACOLOGy CODES Since the pharmacology codes are oriented more toward animals, double code the human codes ("FE; Hu~nan" or "FEB; Clinical") when the pharmacology codes apply to hu/~ans. Most human work will be more behavorial than phar~nacological. Be sure to circle any in vivo and/or animal inhalation codes that apply, as well as the pharmacology code(s), when the research was performed on animals. pHARMACOSOGY/H.t/MIN SMOKING BEHAVIOR Studies evidencing human aggression should not be coded to "ED; Pharmacology"; instead, they should be coded to "EFG; Smoker Psychology." Note: There is no need to code "FE: Human" when a "Human Smoking Behavior" code is used. pOSiT;ON PAPERS If you have evidence that B&W approved an industry position paper, code the paper to "BC; B&W Position Papers." PRE-1964 pERSONNEL/POSITIONS !NFO<~%TION If you find documents showing positions held by Brown & Wi!liamson personnel for years prior to 1964, please send that information to Houston. -32-

PREMATURE DEATH Statements alleging that smoking causes premature deaths should be coded to "JB; Risks" and "DG; Other Diseases," in addition to whatever other codes might apply. PRICE SUPPORT PROGRAM FOR TOBACCO Code these documents to "EG; Tobacco Processing" and write "PRICE SUPPORT PROGRAM" in your comments. PRISM PANEL This panel is a method in which trained panelists rate several cigarettes monadically along a number of attributes, e.~., overall intensity, draw, flavor. The attribute data is then analyzed and organized into maps showing how the cigarettes compare with each other in various "flavor spaces." BAT calls its similar procedure DELTA. Documents relating to these panels will generally be coded to "AAC; Product Testing," in addition to any other codes that might apply. PROCEEDINGS If a document discusses or contains testimony of an industry representative speaking at a proceeding, it will receive a significance of "i" unless the issues being discussed are not related to smoking and health (e.~., fire-safe cigarettes). (Excise taxes are related to smoking and health, because taxes influence the price of cigarettes. The price affects the ability of young people to buy cigarettes.) PROCeeDiNGS ~ WARNINGS There is no need to code to "K; WARNINGS/PROCEEDINGS" when you code to one of the warnings proceedings codes (e._~_g_~, "KF; Hatch-Packwood, Waxman"). Be judicious when coding to "A; ADVERTISING/MARKETING"; use "A" when the focus is on advertising, rather than on the warning within the ad. PROCeeDINGS TO AMEND CIGARSTTE LABELING ACTS If the proceeding is not specifically mentioned on the taxonomy, code to "K~; Other Proceedings," and reference in the comments field the specific act that is being discussed. PRODUCT DEVeLOPMeNT Research is probably not product development if researchers are working solely on the tobacco (probably -33-

filler modification or tobacco processing) or only on the filter. If they are researching the filter add the product, then it is product development. Ad copy and marketing- related information is not product development unless the document addresses what is new about the product. PRODUCT LIABILITY Documents discussing product liability should be coded to "M; LITIGATION ISSUES" even if the product involved is not tobacco. Many of the legal issues will still apply to cigarette litigation. ~©~UCT TEST SHOWING TAR~ N%COT~NE & MOISTtrRE Code these documents to "EH: Quality Control" if they are a general analysis of the product. Do not code them to "EHB; T&N Tracking" unless the focus of the document is on monitoring of tar, nicotine or other substances. pRODUCT TESTING RD&E has a Product Division that does product testing and works hand-in-hand wlth its marketing research counter- part. When comparisons of establ~shed brands are conducted to ascertain consumer preferences, code to the appropriate "AAC; Product Testing" code, but not to "EI: Product Development" unless the document shows an intent to modify £he established brand. When comparisons of established brands are conducted to ascertain chemical or physical differences and similarities, code to either "EH; Quality Control" or "EI; Product Development," depending upon the reason for the comparison. When comparisons of new brands (to each other or to an established brand) are conducted to ascertain consumer preferences, code to the appropriate "AAC: Product Testing" code and to "EI; Product Development." If a product test is for the purpose of product improve- ment, it should receive the "EI: Product Development" code. If it is for the purpose of monitoring, it should be coded to "EH; Quality Control." The tests will often have the words "product improvement" or "monitor" written on them. Product tests will generally not receive a funding source code or a brand marketing code. Concept-product and ad-product tests, however, w~ll receive a brand-marketing -34-

code, because a concept or an advertisement is being tested in conjunction with the product test. p~O-~NDUSTRY PUBLIC RELATIONS ~OCUMSNTS BY TRADE ASSOCIATIONS (Decision was changed on June 3, when new codes were introduced.) When there was no generic trade association code (prior to June 3), we coded pro-industry public relations documents to "HC; PR/Publications." We indicated in the comments that a trade organization other than TI wrote the document. pEOJECT G Project G took place around 1979-1981 and involves some prelimlnary work wlth physicians. Project G is probably a forerunner to LIMIT. Send any documents concerning this pro3ect to Allen Creighton or Dan Willoughby. ?ROJ~CT HIPPO (See also PROJECT MADHATTER.) Send documents discussing Project HIPPO to Dan Willoughby. PROJ,C ~ ~VORY Project IVORY related to an additive-free cigarette. Send documents discussing Project IVORY to Dan Willoughby. PROJECT JANL~S AND PROJECT ~MN These two projects have become increasingly important to the present tobacco litigation. Any document containing a substantive discussion of these projects should get a significance of "i" and should be sent to Dan Willoughby. Project LIMIT was a plan to market a low tar, low gas cigarette to pha.~macists and physicians. In order to prevent smokers from compensating, the cigarette was not lower in nicotine. Send any documents discussing this project to Allen Creighton or Dan Willoughby. Other possible names were CARE, MINIM, FACET, SELECT and BAI~NCE. The work took place probably between 1979 and 1981. Look for Dugans Farley Communications Associates' report "A Medical ?rogram." Contact Dan willoughby if you see this report. -35-

PPOJECT MADHATTE~ Sir Charles Ellis performed nicotine research in Project MADHATTER and in Project HIPPO. MiADHATTER was a precursor to project ARIEL (the Ellis patent). The MADHATTER project occurred in the early 1960s. Collect all references to Project MADHATTER and forward them to Dan Willoughby as soon as possible. PPOJECT PEGASUS Pegasus was an increased manpower concept (larger sales organization in order to improve sales/share trends). PPOJECT RIO All documents discussing Project RIO should be sent to Dan WilloughDy. PROJECT SH~ [SMOKE HARSHNESS IMPROVEMENT PROJECT) Project SHIP (1984-85) was a joint effort between B&W and BAT to investigate the exlstence and effects of smoke Irritation and methods to reduce the irritation. The project involved chemical analysis and biological activity ratlngs of smoke from various tobacco blends. Parameters studied for irrltancy were aerosol particle size, aerosol generation temperatures, smoke generation, pyrolysis, filtration, microbiological toxicity and human taste analysis. PSEUDO-PRODUCT T~STS Pseudo-product tests should be coded to "AA~H; Packing Research" and to "AAC; Product Testing." The test is actually of the pack, but the participants believe it is a product test. pUBLIC DOCUMENTS Do not fully code all public documents, but do give full coding to any critical aspects of those documents. Assign significance rating as you normally would. Give full coding if the documents reference Brown & Williamson, quote Brown & Williamson employees or contain marginalia. Code more significant or comprehensive public documents that relate to the smoking and health issue to "BB; B&W Awareness & Action." Be sure to always begin your comments field with "PUBLIC DOC." Newsclips are not very important documents. They do not necessarily reflect company pos=tions, and their presence in B&W files does not Imp!y that B&W adopts or dlsputes -36-

statements appearing in such articles. Capture the major topics discussed, and do not over-code. These documents will receive a significance of "3" unless the newsclips either quote B&W employees or centain marginalia that may make the documents more significant. RD&E BUDGET CODE -- EFFECTIVE DATE 3/~/88 The new code is: GBA RD&E Budget: Documents discussing proposed or actual budgets and/or expendi- tures for RD&E. Do not use this category in place of a funding source. This code is for department, project or overall budgets. It includes any capltal expenditures over $5,000 (e.g., money for a new smoking machine or gas chromatograph). It does not include funds for equlpment maintenance. MEETINGS Code documents discussing these meetings to "BBC; Attendance at Sclentific Meetings or Research Conferences." All B&W RD&E formal meetings, even meetings of the Product Development Com/~ittee and the Leaf Availability Com/~ittee, will receive thls code. "CE: Meetings" is to be reserved for corporate meetings. Be sure to code to "BBC: Attendance at Scientific Meetings or Conferences" when B&W personnel go to a research site to view research or research results. RD&E PROJECTS CHRONOLOGY To facilitate the development of an RD&E projects data base, collect and record on the chronology forms all relevant information on research projects performed by the industry. These forms will then be sent to Atlanta. Do not list the information in the CTR status reports, which contain information on hundreds of studles; copy these reports so they can be sent to Atlanta. RD&$ REPORTS CHECKED OUT BY SAW DEPT= RD&E reports from B&w's libra.~y were sent to or checked out by the Law Dept. Assign an "NB" code to these documents, and include a short comment such as "RPT CHECKED OUT BY LAW DEPT." -37-

~C~pES AND SPECIFICATIONS Documents which reveal recipe or formula information should be coded to "O: TP~ADE SECRET" as well as to any other codes which apply ( .~., the "EC; Additives" codes) ; write "RECIPE" in the comment. Recipes or fo~-mulas are a listing of those raw ingred- ients or in-process materials wl~Ich are used in making the product. A recipe usually specifies the quantities of ingredients by weight or percentage and may include manufac- turing directions. A document which reveals any of this information should be coded as a trade secret. A document which reflects the specification for a product is merely outlining the parameters for judging the quallty of the finished product. The parameters may be expressed in absolute terms, percentages or as target values/ranges as a measure of the physical or chemical nature of the product. Specifications, although they reflect the end product of a recipe, are not recipes and therefore do not reveal trade secrets; do not code "O; TRADE SECRET" when specifications for a product are stated. A document stating specifications may receive the "EC: Additives" code(s REFERENCES TO CRITICAL DOCU~ENTS An innocuous document that cross-references a critical document will not receive a "I" in significance. Explain the relationship between the two documents in your comments if it is not evident from the CDC printout. REQUESTS FOR AND R~SPON$~$ To REQUESTS FOR P~O~VCTION OF S~MP~E C~GAR~T~S -- 3/10/88 FINAL DvCTS!ON Prior to March 10, we coded these documents to "NC; No Legal Reference," "GB; B&W Alone," "EI; Product Development" and any marketing codes that applied. We used the Product Development code for a product proposed as an experiment and the Tobacco Processlng code if the recipe was for a real brand. Recipes received a significance of "2." On March i0, it was decided that some of these docu- ments are "EH; Quality Control" and some are "EI; Product Development." The document that requests the samples should help you determine whether it is QC or product development. All samples that are produced for RD&E will receive a funding source code. Those samples produced for marketing do not get a funding code. Include any substances that are targeted. (Targeted substances will have a number next to them in parentheses.) These documents will normally receive a -38-

slgnificance of "3," but marginalia may increase the signifi- cance. ~ESEARCH DOCUMENTS Pay special attention to research that took place before the warning. Send 1950s research performed by or on behalf of Brown & Williamson or BAT to Dan Willoughby. P~SEARCH FOCUS ¢OD~ -- EFFECTIVE DATE 7/~9/88 E Research Focus: Documents relating to scientific or research oriented work not covered by the subcategories below (e.__~., ash studies, description of scientific instruments, etc.) When a document is coded to this category please Include a description of the research in the comments field. RKSEARCH FUNDING SOQRCE Assign a funding source to all scientific "research" projects that have an identifiable funding source. This code should not be used for marketing or lobbylng. With regard to literature revlews, one funding source will be the company funding the review. The other funding source(s) will be for the actual research pro~ects that are belng reviewed. If a document discusses research funded by Brown & Williamson and the document identifies the amount of money spent on that research, code to "GB; B&W Alone" and "GBA; B&W RD&E Budget." If you can tell from the document that B&W funded the research, but you cannot tell how much B&W spent, then code to "GB." Code RD&E budgets and capital expendi- tures to "GBA" only. "GBA" is not to be used for Qc or tobacco processlng. When a document shows B&W financing a portion of a grant, code to "GB; B&W Alone" (if a specific dollar amount is not given) or "GB; B&W Alone" and "GBA; RD&E Budget" (if a specific dollar amount is given) and to whichever "G" codes also apply. If necessary, ~nclude a comment stating that B&W is financing the pro~ect accordlng to its market share. Documents that state how much the industry has spent on research should be coded to "GE; Tobacco Industry." If the TAC, TRC or VdC is funding research, code to "GE; Tobacco Industry" and "R; INDUSTRy RESEARCH ORGANIZA- TIONS." -39-

Documents showing evidence that tobacco companies, such as B&W, made payments directly to the American Medical Association (AMA) for research should be given a "GB" or "GE" code according to who is doing the funding. Do not code "G" for AMA funding. RESEARCH TECHNIQUES The "F" codes should be used when a document includes references to the effects of smoke or smoke constituents on a biological system. Leaves/plants are not biological. receive an "E" code, not "F." Work on plants will RESEARCH TECHNIQUES ("F" CODES) AS APPLIED TO HUMANS Human research should be coded to the appropriate "FE; Human" code. Usually the other "F" codes do not apply to humans. However, if a document discusses a human clinical study on metabolism (e.____g~., nicotine metabolism), code also to "FCD; Metabolism." If a document discusses a human clinical study on topical application of a substance (e.~., nicotine), code also to "FCB; Topical Application." Do not code human clinical studies to "FC; In Vivo" merely because they are "in life" or in vivo. RESEARCH THAT wAS NEVER FOLLOWED UP UPON Use "ME; Failure to Publish/Suppression" for actual promising research that was never pursued (e._~., research report on safer cigarette with marginalia stating that the cigarette will not be marketed). RESUMES OF RESEARCHERS Code diseases and substances from resumes of persons who have done smoking & health research. Other~ise, resumes will generally receive only a significance of "3" and an "NC" code. ROPgR STUDI~S Fully code these studies. Be sure to circle all applicable "EF; Human Smoking Behavior" codes. In addition, code to "HD: Polls and Surveys" a Roper report that was commissioned by TI. -40-

RQYAL COLLEGE OF PHYSICIANS The college can be abbreviated as "RCP." Be sure to code RCP reports, as well as SGRs, to "BB; B&W Awareness & Action." "SAFER" CIGARETTES It is important to look for infor~mation on why certain products were not manufactured. Look for statements that "safer" cigarettes tasted bad, especially in taste panel documents. Vanguard and Fact had lower biological activity, but B&W discontinued these brands. Look for documents that explain why the brands were discontinued. Also look for early taste panel tests with negative results. Documents showing that B&W is aware of another organiza- tlon's research on a safer cigarette should be coded to "EE" (e._~., documents in B&W's files about the USDA safer cigarette studies). Documents that discuss the aerosol cigarette should be given the "EE" code, as well as the "EIB; Aerosol Cigarette" code. SENSITIVITY OF RESEARCH If you do not have the technical or scientific knowl- edge to be able to understand a research report, include a comment such as, "REQUIRES TECH REV TO PROPERLY ASSESS SENSITIVITY." SIGNIFICANCE (Some of this information is clarified in the SIGNIFICANCE subcategories.) Documents will receive a higher significance if they discuss: Additive-free cigarettes, additives testing, the Blackman letter, important aspects of the Barclay contro- versy, the Hoffman paper, safer cigarette manufacturing, B&W/BAT S&H research sharing, the Snow White controversy, allegedly hazardous substances or substances that B&W took off the market and young adults or youth. S~GNIF~CANC~--B&W/BAT RELATIONSHIP All documents that evidence the sharing of smoking and health type research between B&W and BAT should receive at least a "2" in significance. Other lesser types of research such as minor product development work or tobacco processing would receive a significance of "3." Shared research that -41-

appears to be proprietary in nature should get an "0" code and a "CC" code. SIGNIFICANCE -- FILE FOLDER PAGES File folder beginning and ending pages will receive a significance of "O." S~GNIFICANCE -- All lobbying documents will receive a significance of "i." Even documents discussing indirect lobbying will be coded to the lobbying code and will receive a significance of SIGNifiCANCE -- SMOKING AND HEALTH RESEARCH Research with a stated purpose related to health should receive at least a "2" in significance, while research for taste or flavor purposes will most likely ge~ a "3." Three basic types of research deal with smoking and health. The first type of research involves the detection or quantification of smoke components. Research reports covering this topic may have a higher significance depending on the substances involved and the focus of the paper. In general, the substances list may be used as a guide to sensitive substances. Studies that focus on these substances or other components thought to be potentially harmful should get a "2" or higher. The second type of research includes studies to reduce cernain components in smoke. The significance of the document will be controlled, in part, by the importance of the substances that are researched. Most documents discuss- ing research for reduction of smoke components will get a "2" or higher. As the basic research moves toward product development and test marketing, the significance will decrease depending on the component involved. Documents showing an attempt to reduce certain highly sensitive substances, e.c[., benzo[a]pyrene, CO, PAH, HCN, etc., demand a higher significance. The third type of research concerns biological testing. Most documents concern±ng biological testing of whole smoke, components of smoke or additives will receive at least a "2." If any health claims or warnings are made about these substances, the document will probably be a "I." -42-

~;$NIFICANCE -- SPECIAL ACCOUNT 4 Special Account 4 documents will receive a significance of "I" because of lawyer involvement with scientists. ~IGNIFICANCE -- TARGET M~ARKET$ Unless there is a slur or an otherwise significant issue contained in a document concerning or discussing a minority target market, apply normal significance to these documents. The same applies to documents discussing the military target market. If a document uses the word "young" to describe a target market, assign a significance of "i" or "2," depending upon the context. Code "young adult" to a "2" and "youth" to a "I." If the document targets persons who are 18 to 21 (i.e., 18 through 20), assign a significance of "2." Assign a significance of "I" if the document targets persons under 18. Circle the youth code whenever persons under 21 are tar- geted. You may further clarify in the comments field the age of the target audience. SIMON GAME The Simon game was used as a VICEROY and KOOL promotion. Code discussions of the game to "ABD; Promotions" and identify the game in your comments. Marketing research relating to the game should be coded to "AA; Advertising/ Marketing Research" as well. Be sure to assign any brand- specific marketing codes that apply. SMOKeLeSS TOBACCO RESEARCH COUNCIl/SMOKEleSS TOBACCO COUNCIL. This organization has many activities which are similar in purpose to CTR. It solicits applications and funds research in the area of smokeless tobacco products. It also performs many activities similar to TI, such as lobbying and public relations work. The president is Michael Kerrigan (former vice-president of TI). This organization should get an "R" code. Documents relating to the Smokeless Tobacco Research Council are generally not a "QA." The "QA" role is often played by the Smokeless Tobacco Council, Inc. Kerrington is the president of the Smokeless Tobacco Council. SMOKE MODIFICATION Under "EB; Smoke Modification," there must be an attempt to manipulate smoke. Also code to "EA; Components -43-

Research" if the focus is to identify, quantify and manipu- late. SMOKER PERSONALITY STUDIES Smoker personality studies should be coded to "AABE; Segmentation Study," in addition to any other codes that apply. $MOKING INCIDENCE Code documents about smoking incidence to "AABH; Demographics." SNCW WHITE CONTROVERSy These documents should be coded to "AN; Cinema Adver- tlslng" and should receive a significance of "2." SOCIAL SMOKING PROJECT Code to "EJ: Environmental Tobacco Smoke" and whatever other codes apply. SPECIA~ ACCOgNTS "GEE; Special Account 4" includes all documents on special accounts. Speclal Account 4 was actually one of several Special Accounts, maintained by law fir~s and administered by lawyers. These documents should always be coded "NA" for legal discussion or recommendation, and they should always receive a significance of "i." Special Account documents may include funding for research by expert witnesses in preparation for testimony directly related to a particular case, preparation for testimony in congressional hearings, or other research which was administered by lawyers ra~her than by the Scientific Advisory Board of CTR. When a Special Account 4 person is testifying at a hearing, be sure to code to "GEE; Special Account 4." Do not code "one-liner" statements by Special Account 4 recipients unless they dlscuss their or other scientists' research. SPINDLETOP Brown & Wil!iamson made direct grants to Spindletop Research, Inc. in 1966 and 196S. The 1966 project was entitled "Analysis and Control of the Products of the Pyrolysis of Tobacco." General research was performed in -44-

1968. Brown & Williamson gave $22,000 to Spindletop in 1966 and 1968. STATE TAX HEARINGS These hearings should be coded to "K~; Other Proceed- ings." Be aware that there are three Sterlings: Theodore, Elia and David. All three did work for the industry with regard to the sick building syndrome. Ted Sterling wrote "Does Smoking Kill Workers or Does Working Kill Smokers?" and another article on the "sick building syndrome." He once chaired a round-table discussion entitled, "Attributable Risk Due to Smoking, Occupation and Other Relevant Variables." Both CTR and NIH funded his work. Often, it is impossible to determine who is funding some of his projects. Unless you can ascertain a specific funding source, assign a funding source of "GEE; Special Account 4" and "GA; Government." According to B&W's marketing department, such studies performed by B&W may be competitively sensitive. They often mention addiction or motivatlon, so assign a significance of "I" to these studies. SUBSTANCES The Other Substances code is meant to capture flavors, additives or constituents not already pre-printed on the coding form. Other substances need to be written in the comments field; however, use your best judgment on what to write in. Chemical reagents or solvents used in laboratory procedures are not important in the litigation and should not be mentioned. Examples might be ethanol, acetic acid or hydrochloric acid. Filter materials such as polystyrene or cellulose acetate also do not need to be mentioned unless the document focuses on possible health effects of burning the filter. Filter additives that might end up in the smoke should be coded to "1084; Other Substances." Chemicals used in manufacturing may or may not need to be mentioned, depending on whether or not B&W is concerned that these chemicals may leave residues on the tobacco. One example may be freon used in the tobacco expansion process. Always look for smoking & health implications when COnsidering whether to mention other substances. Make a list of other substances -45-

and send additions to Atlanta weekly so that the list can be updated. If the document mentions "other substances," but never identifies those substances, do not circle "1084; Other Substances" on your form. When you do circle "1084; Other Substances," identify the substances in the comments field. If between one and ten other substances are mentioned in the document, list all substances in your comments. If the list of other substances is lengthy, list the substances that you judge to be most signiflcant or that are found in highest concentrations and add "X MORE SUBSTANCES LISTED" to the comment. Note in your comments if no sensitive substances are mentioned in the long list. Make sure that the sUdD- stances mentioned in the document are not solvents or plasticizers. Angelica Root is supposedly carcinogenic. Code to "1084; Other Substances," increase the significance and identify the substance In your comments. Cocoa and chocolate are considered as carcinogenic in England, so increase the significance. Only list substances that are in tobacco, in cigarette smoke, added to tobacco or clgaret%e paper, me~abolltes of smoke constituents, products of burned filters or fllter materials, chemical residues or allegedly synergistlcally hazardous. When a substance is encountered in a document, refer to the Chemical Substance list to deter-mine whether the sub- stance is allegedly hazardous; this may affect the signifi- cance of the document. If you encounter a document which describes a substance as hazardous and this information is not reflected on the Chemlcal Substance llst, please send a copy of the document to Atlanta so that the list can be updated. Research on carboxyhemoglobin is gas-phase research. Alkaloids have been asslgned a substance number of 1085 (as of 3/11/88) . "1002: Acids IVolatlle)" should read "1002; Acids" [May 6, 1988 decision]. Amino Acids should be coded to "1084; Other Substances," no_~t "1002; Acids." Amino acids found in tobacco include: Threonine, asparagine, pro!ine, glutamic acid, aspartlc acld, leucine, isoleucine, serine, tyrosine and lyslne. -46-

Nitrogen salts (e.~., sodium nitrate) should be coded to "1053; Nitrates." Nitrogen gases should be coded to "1055; Nitrogen Oxides." Code octahydrocoumarin (a coumarin substitute) to "1032; Coumarin" and to "1084; Other Substances." Assign a significance of "i" if the document discusses discontinuing coumarin or testing of octahydrocoumarin. Synonyms for polycylic aromatic hydrocarbons -- PAH, aromatic amines, n-hydrocarbons, n-PAH, aryl nitrates. Synonyms for deer tongue -- Carolina vanilla, dog tongue, Trilis~ Odo?at~ssi~, vanilla leaf, prairie pine and blazing star. Be sure to code to "1084; Other Substances" items such as alcohols, ketones, esters, vitamins, free radicals, e~c. Code cellulose as an other substance if it is added to tobacco or used as a tobacco substitute. Do not code it as an other substance when it is simply in the filter. Do note code adhesives for filter paper as other substances or additives. Do code adhesives for rod paper since they will burn and could end up in the smoke. Polyethyleneimine (PEI) is a filter additive and was used in the Duolite filter. PEI selectively removes HCN and aldehydes. However, do not code "EBCB; Selective Filtration" unless a document dlscusses PEI in the context of selective filtration. Acids that are added to filters should generally not be coded on the substances list because acids are used to remove bases, such as nicotine, from the smoke. According to a patent found in Houston, zinc, cadmium and mercury can pass through the filter when they are used as filter additives. Since these substances then can be inhaled by the smoker, code them to "EC; Additives" and to "1049; Metals" when they are used in the filter, in addition to whatever other codes apply. SURGEON GENERAL'S REPORTS In addition to whatever other codes apply, code these reports and substantive discussions of these reports to "K; WARNINGS/PROCEEDINGS." Be sure to include "SGR" somewhere in your comments. A documenz must discuss the Surgeon General's report, rather than merely reference the report, to be coded to "K; WARNINGS/PROCEEDINGS." -47-

SWITCHING STUDIES An actual switching study will be coded to only "AABG; Switching Study" and possibly to "AABH; Demographics" and will receive a significance of "3." ~n analysis of a switching study, however, should be fully coded and will often receive a higher significance due to data on persons under 21 years of age. TAR & NICOTINE LEGEND When a document discusses the size of the legend (e._~., 8 point, I0 point), code to "KB; Design," include a comment stating that the code relates to the T/N legend (rather than the warning notice) and do not circle the tar and nicotine codes on the substances list. If a document discusses the number of milligrams of tar and nicotine that should be included in the legend, d_~o circle tar and nicotine on the substances list. For other situations relating to the tar/nicotine legend, try to determine whether or not tar and nicotine are being discussed as substances. TARGET MARKETS (See also SIGNIFICANCE--TARGET MARKETS) When you encounter a target market that is not specif- Ically listed in your taxonomy (e._~., full taste, Marlboro smokers), code to "ABE; Target Audlence" and identify the specific target audience in your comments field. Targeting of dual-concerned smokers should be coded to "ABEC; Health Conscious." Indicate in your comments field that the target market is the dual-concerned smoker. Targeting of hi-fi smokers should be coded to "ABEP; Low Tar." Do not code very general demographic breakdowns to the specific target audience, but do code to the relevant target audience when the study concerns a specific group (e._~., a black smoker study). The Raleigh "Traditional Values" target market should be coded to "ABEJ; Psychographics." Code to the appropriate target market code when Brown & Williamson's ads are directed toward a specific audience or when Brown & Williamson is studying a specific audience (e._~., black smoker study). Also code to a target market code when the company uses a specific category of persons in -48-

a product test, .~., when Brown & Williamson wants to use I00 black people in a KOOL product test. THEMATIC APPERCEPTION TEST (TAT) TATs should be coded to "AAAB; Communications Testing." THORNTON PATENT Raymond E. Thornton developed the patent for BAT and it was filed in the UK in late 1967. Thornton assigned the patent to B&W in late 1968 and it was filed in the USA on 9/3/68. After numerous amendments, the patent was issued on 10/26/71. The patent concerns a cigarette which selectively removes PAHs (part of which is believed to be B(a)P). The cigarette has a filter-type material which extends the full length of the cigarette a~nd is surrounded by tobacco. The filter material is combined with activated charcoal, activated basic alumina, ammonium nitrate or oxides of transition metals to selectively remove the PAHs. Other patents referenced in the USA patent are Geisler, Helm, Berger and Keith II. Patents were also issued in Canada and France. Per Bob Sachs, the Thornton patent may be more signifi- cant than the E!lls patent. All documents which discuss the paten~ should be coded to "CC: B&W/BAT Relationship," "EE; Safer Cigarette" and other applicable codes and should receive a significance of "!." Key persons were James J. Salerno, Jr. (KDK), Patrick M. Kelly (B&W), Larry Amos (B&W), T. Davies (BAT), R.J. Tug~ood (Jensen & Son) and J.R. Tugwood (Jenson & Son). Collect all references to this patent and foward them to Dan willoughby as soon as possible. 3 i/LSI/~RD Documents discussing this organization should receive a significance of "i." They should also receive an "NA" because the data bases were set up by lawyers. TI -- ADVERTISING Do not double code TI ads to "HC; PR/Publications" unless the ad relates to or refers to a publication. There is no need to double code a TI advertising budget to "H; TOBACCO INSTITUTE" and "HF: Advertising": simply code to the TI advertising code. TI advertising research should be coded to "HF; Adver~islng"; do not double code to "HD; Polls and Surveys." -49-

TI -- B&W INVOLVEMENT Code to "KA; B&W Involvement" if the document is from B&W to TI. In other cases, base your decision about whether to code to "HA" upon the content of the document. The • document must evidence prior, current or future involvement or action by B&W specifically (e._~, a letter from Chilcote to the Executive Committee simply referencing Executive Committee action is not coded to "HA"; if the same letter mentions Pepples as an Executive Committee member, code to "HA" and "HG."). A simple FYI or transmittal letter from Chilcote to Pepples would not necessarily receive the "HA" code. ~ -- COMMITTEE OF CQU~SEL M~TINGS The Committee of Counsel is a TI committee. Documents concerning Committee of Counsel meetings will be coded to "HG: Lawyer Involvement" and possibly to "HA; B&W Involve- ment," along with whatever other codes apply. These meetings are not the same as national counsel meetings. Often, meetings will be held at TI facilities and have nothing to do with TI. T! -- EXECUT~V~ CO-M/~.!TTEE Pepples was on this committee from 1982 through the present, so circle "NB" on these documents for those years and include a brief explanation in your comments. Tf -- INTE~%IG~NC~ GATHERZNG TI's monitoring of hearings should be coded to "HK; Intelligence Gathering." TI -- LAWYER INVOLVEMENT Code to "HG; Lawyer Involvement" if the document is from a lawyer to TI or is from a lawyer transmitting TI materials. In other cases, the document must show involve- ment or action by lawyers in order to receive the "HG" code. This code can be inferred to other documents if the project dlscussed in those documents was handled by attorneys (e.~., Project ROCKET), but be sure to make the basis for your inference clear in your comments. TI-- PROPQSA~S WRITTEN TO TI. _~.,.~.~..ce~~== OF COUNSEL ASKING FOR FUNDING O[ ~SEARCH Code to "HG: Lawyer Involvement" and whatever other codes apply. Include a comment stating that TI does not fund scientific research. -50-

TI QUOTES These books are generally defensive documents (~.e., they contain pro-industry arguments) and should receive a significance of "i." They should be coded to "H; TOBACCO INSTITUTE" in addition to any other codes that apply. TODACCO ACTION NETWORK (TAN) Because the purpose of the TAN is to lobby for the industry, any discussion of the organization's activities should be coded to "HB: TI-Lobbying." If the document discusses B&W's contributions to the TAN, also code to "LA; B&W Lobbying." If the document discusses another manufac- turer's contributions to the TAN, also code to "L; LOBBYING," and make an appropriate comment regarding the identity of the lobbying party or parties. TOBACCO COLLEGE/TOBACCO FORUM (See also EMPLOYEE INDOCTRINATION.) Treat agendas for either B&W-related activity as you would treat agendas for B&W meetings. The mere mention (vs. a substantive discussion) of a toplc to be discussed at the gathering is enough Information to code to the applicable code (e._:i., I0:00 Discuss Litlgation and B&W's Defense: code to "M; LITIGATION ISSUES" and "MLB; Litigation Defense"). Documents discussing the B&W tobacco forum or tobacco college will generally be coded to "C; B&W CORPORATE," in addition to any other codes that apply. TOBACCO COLLEGE WASHINGTON OR%ENTATION These documents typically receive a significance of "l" or "2." Congressmen are generally speakers. Circle at least one lobbying code when the document evidences an invitation to a senator, when B&W helped develop a speech for a Congressional speaker or when a senator received an honorarium (fee to a speaker). Agendas will receive a lobbying code if they show that B&W employees met with Congressmen. TOBACCO INSTSTUT$ TESTING LABORATORY Because TITL is a division of TI, documents reflecting B&W's involvement with TITL should be coded to be appropriate TI taxonomy codes. -51-

TOBACCO PROCESSING ~QQGETS Code these budgets to "EG; Tobacco Processing," rather than to "GBA; RD&E Budget." TOBACCO WORKING GROUP Since the TWG worked on various projects, double-code to "EE: "Safer Cigarette" and "EED; Tobacco Working Group" when the Group is specifically working on a safer cigarette. The TWG was formed in or around 1972. in 1975 unless an extension was approved. It was terminated ~OX%CITY RESEA~ Any research documents relating to the toxicity of smoke or smoke constituents (lethal doses - LD tests) should receive the "FCC; Toxlclty" code. If the toxicity test is performed on a microorganism, also use the appropriate "FB; IN Vitro" code. Key words to look for are "chronic dose" and "lethal dose." Toxicity studies, which are relatively short te~m (i.e., days to weeks), measure illness and/or death of test animals resulting from exposure to a substance. Abnormal ~ii effects include liver damage, convulsions, prostration, loss of appetite, lack of groomlng, abnormal protein levels in blood or urine, etc. Toxicity studies will often be coded to "DG: Other Diseases" when a disease, a condition or death is observed. TRACKING When coding a study to "AAD; Opinion Tracking" or "EHB; T&N Tracking," be sure that the study is actually tracking a trend or the amount of a substance over time and is not simply a one-time study. TRADEMARKS When a document discusses brand name trademarks, code to the relevant marketing code(s) ("A," "AC," "AD," "AE," "AF," "AGA," "AGB" or "AGC") . Code to "EK; Cigars/Smokeless Tobacco" if the trademark relates to NCPs. Assign a "P" code if the trademark is registered in a foreign country. TRADE SECRET CODE -- EFFECTIVE DATE 3/2/$8 The "O" code reads as follows: -52-

TRADE SECRET: Documents which contain competitively sensitive information, including: I) documents which outline overall specifications (e.~., flavor, casing and humectant formulas and amounts, leaf blend proportions, etc.); 2) documents which outline a unique tobacco treatment technique or cigarette manufacturing process which is not patented; or 3) documents which contain any other form of proprietary information. TRAINING SEMINAR TRIP REPORTS Trip reports for a salesperson's attendance at a sales training seminar can be coded to "A: ADVERTISING & MARKeT- ING." Trip reports for a middle management person's attendance at a management training seminar can be coded to "C; B&W CORPORATE." Trip reports for an upper management person's attendance at a training seminar can be coded to "CE; Meetings." TRANSMITTAL NOTES & MEMORANDA A transmittal should be coded to topics actually discussed on the face of the transmittal itself. Codes are not to be inferred from the document which accompanies the transmittal. The transmittal will be linked to the at=ach- ment by the Master I.D. and the CDC objective fields, such as title and inferred title. Transmittals as a general rule will have a significance of "3" unless they contain other important information which is sensitive or can be coded to a taxonomy code (e._~., a transmittal stating, "This research is bunk."). TRAVEL ITINERARIES & SXP~NS~ ACCOUNTS These documents will generally receive a significance of "O." However, if they relate to B&W/BAT meetings, or B&W's attendance at smoking & health meetings or B&W atten- dance at a large scale industry meeting, assign a signi- ficance of "3" and code to the appropriate meeting and lawyer codes. UN~VSRS!T~ O~ KENTUCKY TOBACCO & HEALTH RESEARCH INSTITUTE This institute is funded by a Kentucky state sales tax on tobacco products and is unrelated to TI or CTR. The members may be appointed by the governor. Note any indica- tion of industry influence. If the institute is funding research, code to "G; RESEARCH FUNDING SOURCE." -53-

Consider unions to be trade organizations ("Q" or "QA") if they are representing the industry ('.~, lobbying or speaking at a proceeding). Do not treat unions as trade organizations if they are representing the workers (@~_q_~, striking). UNRESOLVED OUESTIONS I. What is a headroom test? 2 o What is Project Bluegrass? (It seems to be related to centralization and corporate facilities.) 3. Are all Roper studies done on behalf of TI? 4 o What is a T.O.T. test (Total Offering Test)? (It appears to be some type of identified product test. Is it a concept-product test?) o 6 o Are specifications from suppliers of additives proprie- tary? What is the National Tobacco Education Council? 7 o 9 o Is Project T a forerunner of Project Truth? What are the TI contributions programs to scientists? - (Mentioned in a memo from Russell to Humber. Zahn managed these programs.) What is SASIB? I0. Is B&W a member of the Smokeless Tobacco Research Organization? ii. 12. 13. What is the Aspen paper? (Written by Dietrich H., probably around 1983, and presented at a cancer conference.) What is the function of the Tobacco Products Institute in Bulgaria? (This organization does have some research functions, but it may also perform TI-like activities.) In what years was Gori with the Tobacco Working Group? Was he with TWG when he made his statement that a person could smoke a tolerable number of certain cigarettes without appreclable ill effect? -54 -