Scientific Committee on Toxicity, Ecotoxicity and the Environment (CSTEE)- Opinion on the report by WS Atkins International Ltd (vol. B) "Assessment of the risks to health and to the environment of arsenic in wood preservatives and of the effects of further restrictions on its marketing and use" expressed at the 5th CSTEE plenary meeting, Brussels, 15 September 1998

The CSTEE has been asked:

a) to express an opinion on whether the degree of risk to the environment and man, as assessed by the WS Atkins Report, hereafter called 'the Report', is sufficiently justified;

b) to comment on the general quality of the Report.

The Report describes in different sections the background to the use of arsenic-based wood preservatives and the general environmental chemistry and toxicology of arsenic, and goes on to assess the environmental and health risks arising from the manufacture of CCA and the production, use and disposal of CCA-impregnated wood. In this commentary, the important assumptions, premises or conclusions are discussed, and where necessary, commented upon, for each section of the Report.

1. Introduction

In this section, the background to the use and regulation of CCA-containing wood preservatives is described. The authors indicate that they base their risk assessment on the assumption that "best current practice" is followed at all stages of the manufacture, use and disposal of CCA and CCA-treated wood. To the extent to which the technical specifications for various procedures are imposed by regulatory statutes to which compliance is compulsory, this is a reasonable position. However, it appears that some of the "best practice" guidelines assumed in the Report to be in general operation are guidelines recommended by professional or other organisations and rely on voluntary co-operation for their application. Given the well documented problems related to environmental or human exposure arising from past practices, especially at wood treatment sites, it would be important to have an indication of the extent to which such guidelines are indeed employed in practice.

Of the 3 types of CCA preservatives, discussion in the Report is focussed on type C which is the most commonly used type nowadays.

2. The environmental behaviour and effects of exposure to arsenic

Environmental chemistry:

The predominance of arsenate among the environmentally relevant species of arsenic, its limited mobility in soil, and the possibility that in certain reducing environments it may be reduced to the more mobile arsenite and further converted microbially to methylated forms are briefly described. While these premises may be largely correct, the corresponding discussion in the Report reflects only to a limited degree the complexity of the chemical and biological processes influencing the speciation, mobility and bioavailability of arsenic in such widely varying environments as marine waters, trophic estuaries, sediments and landfills, and the associated uncertainties in making reliable predictions.

Ecotoxicological effects:

a) Water:

Data on the effects of arsenate, arsenite and organic arsenic on aquatic biota are tabulated (note: it is not always indicated clearly whether the figures quoted refer to NOECs or LOECs), the most sensitive species are indicated along with corresponding PNECs derived for different types of aquatic environment. It is concluded that for low-phosphate marine waters the background arsenic concentrations are higher than the corresponding PNEC, thus leaving no room for any contamination. For other water compartments the derived PNECs are approx. 2-4 times higher than the corresponding background concentrations. In deriving these values, no "assessment factors" (i.e. safety factors to allow for uncertainties in the extrapolation process) were included. The EU "Technical Guidance Document in Support of Risk Assessment" recommends an assessment factor of 10 when long-term NOECs from at least 3 species representing 3 trophic levels are available, while allowing the possibility of adopting lower values on a case by case basis if data from field studies or model ecosystems are available. The authors of the Report correctly justify their not adopting an assessment factor on the basis that chronic data from a number of "particularly sensitive species in their most sensitive environments", at different trophic levels, are available. That this argument is more generally accepted is supported by the fact that the PNECs adopted in the Report are not lower than those suggested in other documents (1,2).

No attempt is made in the Report to derive a PNEC for ground water, and instead a "maximum loading limit" of 10 mg/l, derived by the UK Department of the Environment for landfill leachate, is adopted. Neither the meaning nor the basis of the derivation of this figure are described.

b) Air:

No PNEC is derived, but values of an "Environmental Assessment Level" developed by the UK Environment Agency quoted. Again, no justification for the use of this figure is given.

c) Terrestial compartment:

No attempt is made in the Report to address the question of a PNEC for the sediment compartment, even though available data (1) indicate that arsenic levels in the sediment or the pore water can reach values up to a few tens of mcg/kg dry matter or mcg/litre, respectively. Similarly, no soil PNEC is derived, and a "trigger value" (30 mg/kg) used by the Dutch Environmental Agency is adopted. The option of applying to soil of the PNECs for the aquatic environment (presumably in combination with soil:water partition coefficients) is rejected as inappropriate because these PNECs refer to particularly sensitive species not found in the soil. However, it is notable that where this approach (i.e. based on aquatic PNECs) was used (1), limit values were derived which are not very different from the abovementioned limit.

Human health effects:

In general the discussion of arsenic toxicology in this Section of the Report is unacceptably limited, and only a few mechanistic aspects, selected on criteria which are not clear, are mentioned in passing. Such selective reference to one or two mechanistic aspects, while adding little to the value of the document, can lead to erroneous impressions. For example, the possibility that arsenic genotoxicity may be mediated by indirect processes, and its implications for low-dose extrapolation (a important issue of much current discussion - e.g. refs. 3,4) is covered in the Report only by reference to a personal communication!

The toxic end-points identified in the Report as the most sensitive ones on which to base human health risk assessment are a) lung cancer by inhalation and b) non-cancerous skin lesions by ingestion. Surprisingly, no mention is made of the well-documented induction in humans of skin cancer (and possibly other cancers) after oral ingestion. For the assessment of cancer risk by inhalation, linear extrapolation with no threshold is accepted and the unit risk derived by WHO (1997) adopted. For exposure by the oral route, a TDI of 2 mcg/kg (WHO; 1989) is adopted for non-carcinogenic effects.

As elsewhere in the Report, "criteria" values developed by various organisations are adopted without adequate justification, such as for example an "environmental assessment level" for air of 0.2 mcg/m3 (long term) and 6 mcg/m3 (short term) of the UK Environment Agency.

3. Risks from the manufacture of CCA

The production of arsenic acid and the formulation of CCA are described very briefly. Only very limited information on emissions during CCA manufacture appear to be available and for this reason only a qualitative assessment based on the assumption of best practice is made. It is concluded that environmental emissions are likely to be small and that lung cancer risks are within acceptable levels.

4. Risks from the impregnation of wood with CCA

The various stages in the process of wood impregnation with CCA are described and potential sources of environmental releases identified. It is concluded that, if best practice is followed, releases resulting from liquid discharges or solid wastes should be negligible. As regards air emissions, for which some data are given, it is concluded that they have negligible effects in the wider environment, while the occupational lung cancer risks they would give rise to are within tolerable limits (however, no indication or justification of the latter is given). No other type of occupational exposure should occur if best practice is followed. Reference is also made to an epidemiological study on 1042 wood impregnation workers which showed no increase in cancer incidence.

There is extensive documentation of past substantial soil and groundwater contamination at wood treatment sites, much of it well documented in reports from the USEPA Supefund Programme (in the Report a personal communication is given as documentation!). There is also evidence in the published literature (not mentioned in the Report) that contamination of the soil and vegetation can extend to the area beyond the immediate boundaries of such sites, something that has been attributed to wind erosion, percolation, surface drainage as well as on-site incineration of wood waste (5,6). Moreover, a number of studies (including studies published since 1990, again not mentioned in the Report) have reported urinary arsenic concentrations to be substantially elevated (up to ca. tenfold as compared to controls) in wood impregnation workers (7,8). All these data demonstrate the potential for significant environmental and human contamination at wood treatment plants and underline the importance of discussing the extent to which "best practice" is currently applied in practice.

5. Risks from the use of CCA impregnated wood

Discussion in this section is focussed primarily on arsenic leaching from wood to soil and water. Various factors affecting leaching rate are outlined, of which exposure to acid waters (e.g. acid rain or acidified lakes) is identified as the most severe.

Contamination of soil:

Data are presented in the Report on arsenic concentrations in the soil near CCA-treated poles which indicate that, while these concentrations can be well above the criterion of 30 mg/kg adopted earlier in the Report, the tight binding of arsenate to soil components results in their dropping back to background 30-40 cm away. A recent study on arsenic levels in soil under wooden decks supports this conclusion (9). It is concluded that releases from wood are limited to very confined localities and do not represent a serious risk for the wider environment. However, no discussion of effects on biota within the limited contaminated areas is included.

In addition to situations of soil contamination described in the Report, high arsenic levels are also found close to arsenic-treated wood in sandpits in children's' playgrounds (10). While they also show a rapid decrease with distance away from the wood (10), the possibility of ingestion or inhalation by small children of dispersed sand particles containing high arsenic concentrations should be noted.

Contamination of water:

A number of experimental studies of the rate of arsenic leaching from wood are reviewed. It is noted that, while these studies reach variant conclusions regarding the absolute leaching rates, they all agree that leaching drops rapidly with time. It is pointed out that for low-phosphate marine waters the calculated PNEC does not permit any release of arsenic without additional ecotoxicological risk. For freshwater, calculations based on a worst-case scenario concludes that leaching from wood used in mooring facilities or marina decks would be unlikely to result in enough arsenic release to exceed the PNEC. The possible contribution of other sources of arsenic contamination (e.g. airborne deposition) is not discussed, but it is likely to be small. Despite this limitation and the presence of a number of arithmetical errors in the calculations, the conclusion reached in the Report is fundamentally correct and is broadly supported by the results of a recent experimental and field study (11). On the other hand, it is recognised that special situations may exist in particular European regions where the combination of the amount of wood employed and volume of water affected are such that higher levels of contamination may arise.

Data not discussed in the Report demonstrate increased arsenic concentrations in sediments close to CCA-treated wooden piles (12) as well as in tissues of species growing close to such wood (13). At least one study has demonstrated trophic transfer of such contamination and has suggested that it may have biological effects (14). While limited, these observations do suggest possible localised ecotoxicological risks that need to be considered.

Human Exposure:

Limited data on exposure derived from inhalation of wood dust during woodcutting and drilling and from dermal contact of children in playgrounds are considered. It is concluded that, in a worst case scenario, the latter situation can result in arsenic intake that alone can exceed the total TDI for children while even under less extreme conditions it may make up a substantial proportion of the TDI. It is therefore concluded that such exposure may constitute a health risk for children. Surprisingly this conclusion is not carried through to the Conclusions section of the Report. An attempt to downplay this risk by comparing it to the (unavoidable) risk from exposure to uncontaminated soil cannot be considered valid. Finally, to the risks for children via dermal exposure, one could add here the risks, already mentioned above, of additional exposure via ingestion or inhalation of sand particles containing high concentrations of arsenic.

6. Risks from the disposal of CCA impregnated wood

Three modes of disposal of CCA impregnated wood are considered, i.e. a) controlled incineration, b) controlled disposal (of wood or incinerator ashes) to a landfill and c) uncontrolled dumping or burning.

The process of incineration is discussed and the partitioning of arsenic among its various products (bottom ash, fly ash, atmospheric emissions) calculated. As elsewhere in the Report, the assumption of current best practice is made, meaning that incinerators operate according to current EU specifications. This assumption seems justified in so far as that by the time currently produced wood will need to be disposed of it is likely that all incinerators will be operating according to current specifications. Based on this assumption it is concluded, as a worst case scenario, that arsenic released to the atmosphere during controlled incineration will incur an estimated risk of lung cancer of 1.6 per 106, which exceeds slightly that of non-exposed persons.

Controlled disposal of incinerator ash is assumed to take place according to specified procedures, with bottom ash being disposed to monofil compartments and fly ash to "total containment" sites. The extent to which these practices are indeed upheld in practice is not discussed. The role of various factors affecting the leaching behaviour of arsenic within landfills is discussed, including landfill composition, pH and redox conditions, and the difficulty of making reliable predictions, especially about long-term behaviour, indicated. Based on limited experimental and field data, the conclusion is drawn that, at least over a short to medium time horizon, no unacceptable risks for the environment or man are likely. This conclusion is in concordance with all the limited information currently available. Such paucity of information, in combination with the complexity of the processes affecting the speciation and mobility of arsenic under the conditions of landfills, introduces a high degree of uncertainty into any predictions of the long-term behaviour of arsenic-containing waste.

Uncontrolled burning of CCA-treated wood in homes or in open grounds is discussed in the Report and it is concluded that air emissions from such practices can give rise to increased estimated risks of lung cancer. The CSTEE notes that indeed there is direct literature documentation of acute health effects in families burning such wood in their home, the significance of this possible source of human exposure lying in the difficulty of recognising CCA-treated old wood. Ash left over from uncontrolled burning of wood in open grounds is considered in the Report to generate unacceptably high local soil contamination and, if the same site is used repeatedly, it may present a risk of ground water contamination. Finally, the Report points out that the uncontrolled disposal of CCA-treated wood in open sites may give rise to unacceptable localised soil contamination, although the risk of ground water contamination is considered unlikely.

Conclusions

Based on the discussion described above, the following unacceptable risks are identified in the Report:

a) effects on aquatic organisms in low-phosphate marine waters, arising from arsenic leaching from CCA-treated wood;

b) marginally increased lung cancer risks from controlled incineration of CCA-treated wood;

c) significantly increased lung cancer risks from uncontrolled use of CCA-treated wood for home heating;

d) significant effects on the environment from uncontrolled burning and disposal of CCA-treated wood.

e) potential risks from the unpredictable long-term leaching behaviour of arsenic in special waste landfills;

In addition, a risk for children arising from skin contact with CCA-treated wood in playgrounds, is identified in the body of the Report, but it is not included in the Conclusions Section.

The CSTEE considers that overall the Report identifies correctly most of the main risks associated with arsenic-based wood preservation, which are derived primarily from the disposal of CCA-treated wood. In addition to the risks identified in the Report; the CSTEE notes that further risks may exist:

a) for children through the ingestion or inhalation of sand particles in playground sandpits, and

b) for soil, aquatic or sediment-inhabiting biota growing in close proximity to CCA-treated wood.

Finally, a potential risk not discussed in the Report, is that of skin cancer by ingestion of arsenic, known to be significant in areas where arsenic levels in drinking water are naturally high.

The CSTEE wishes to underline that a major source of concern regarding the use of arsenic-containing wood preservatives relates to the high degree of uncertainty regarding the speciation of arsenic during its long-term storage in landfills (the major points of arsenic accumulation), making reliable quantitative predictions about its migration and bioavailability extremely difficult. This is a serious knowledge gap which the CSTEE recommends should be addressed by further research. In the meantime, it would be advisable to exercise caution by limiting the use of arsenic-based wood preservation to those situations where it is absolutely necessary.

Comments on the quality of the WS Atkins report

With the reservation regarding the absence of an adequate discussion of the prevalence in real life of "best practice", and with the addition of the relatively minor risks mentioned in the previous paragraph, it is considered that the methodological approach employed in the Report, and the conclusions arrived at, are broadly valid. However, in arriving at their conclusions, the Report's authors should have given a more thorough discussion of the environmental chemistry and toxicology of arsenic and presented more adequate documentation (specific examples of this weakness are mentioned above). While it is recognised that a detailed review of the chemistry and toxicology of arsenic is not possible in a focussed risk assessment document such as the Report, a more systematic discussion of these matters, plus a list of the bibliography consulted should have been included in order to enhance confidence that all relevant information had been taken into account.

Although much of the information on the subject of wood preservation admittedly is available only in the form of technical reports (many of which are indeed referred to in the Report), there is also substantial information in the open literature to which no reference in made (some examples are included in the references quoted in this Opinion): It is characteristic that, of the 65 or so references in the Report, only 9 are papers published in the open literature. Thirty are technical reports, 5 are papers from meeting proceedings (presumably not subjected to peer review) and there are even 9 references to personal communications (!), some of which concerning issues of major importance.

Inadequate documentation is a general weakness of the Report, seen also in the way in which various numbers are adopted as "quality criteria" values without adequate justification. These include:

a "target value" for arsenic in surface waters of 5 mcg/l, taken from a report to RIVM (1);

a "loading limit" for arsenic in landfill leachate of 10 mg/l, taken from UK DOE (this seems a high value whose basis ought to be clarified);

a "trigger value" for arsenic in the soil of 30 mg/kg, taken from the Dutch Environment Agency;

an "environmental assessment level" for arsenic in the air of 0.2 mcg/m3 long term / 6 mcg/m3 short term, taken from UK Environment Agency;

The paper by Assmuth (1996), referred to in the Report, lists a number of additional quality criteria derived by national and international bodies. What is the basis for selecting the above particular values and not others mentioned in this paper or other sources? Were they the most conservative? The most scientifically valid? In any case, what is the basis for their derivation and, most importantly, what risks (if any) are associated with them? Although most of the conclusions of the Report do not in the final analysis depend critically on the particular quality criteria values adopted, their use in the Report without adequate justification constitutes a source of confusion.

Answer to the questions posed to the CSTEE

a) The Report correctly identifies the main risks associated with arsenic-based wood preservation. Additional risks, not identified in the Report, exist for children playing in sandpits with arsenic-treated wood and for soil, aquatic or sediment-inhabiting biota growing in close proximity to CCA-treated wood. Finally; the uncertainty of predictions of the long-term behaviour of arsenic in landfills are higher than implied in the Report.

b) While the risk assessment methodology employed in the Report is broadly correct, the quality of documentation and justification contained is not considered adequate: Much of the important bibliography on the environmental chemistry and toxicology of arsenic appears to have been ignored. Furthermore, the use of "best current practice" hypothesis, while largely imposed by the lack of adequate data, should have been better justified, as should have been the adoption of various environmental guidelines developed by various agencies.

References

1. Sloof W., Haring B.J.A., Hesse J.M., Janus J.A. and Thomas R. Integrated criteria document: arsenic. RIVM Report No. 71040014 (1990) (included in the Atkins Report).

2. Landner L. Speciation and biological effects of arsenic in the aquatic environment. Report no. 2/98 to the Swedish National Chemicals Inspectorate (1998) (included in the Report)

3. Abernathy C.O., Chappel W.R., Meek M.E., Gibb H. and Guo H.R., Is ingested inorganic arsenic a "threshold" carcinogen? Fundam. Appl. Toxicol. 29 (196) 168-175.

4. Rudel R., Slayton T.M. and Beck B.D., Implications of arsenic genotoxicity for dose response of carcinogenic effects. Regul. Toxicol. Pharmacol. 23 (1996), 87-105.

5. Larsen E.H., Moseholm L. and Nielsen M.M., Atmospheric deposition of trace elements around point sources and human health risk assessment. II. Uptake of arsenic and chromium by vegetables grown near a wood preservation factory. Sci. Total Environ. 126 (1992) 263-275.

6. Ndiokwere C.L., The dispersal of arsenic, chromium and copper from a wood treatment factory and their effect on soil, vegetation and crops. Int. J. Environm. Studies 24 (1985) 231-234.

7. Boudreau Y. and Rinehart R., Health Hazard Evaluation Report HETA 91-314-2179, Memphis Wood Preserving Company, Horn Lake, Mississippi. Govt. Reports Announcements & Index 16 (1994).

8. Farmer J.G. and Johnson L.R. Assessment of occupational exposure to inorganic arsenic based on urinary concentrations and speciation of arsenic. B. Ind. Med. 47 (1990) 342-348.

9. Stilwell D.E. and Gorny K.D., Contamination of soil with copper, chromium and arsenic under decks built from pressure treated wood. Bull. Environ. Contam. Toxicol. 58 (1997) 22-29.

10. Henningson B., 1995, described in Jacks G. and Bhattacharya P., Arsenic contamination in the environment due to the use of CCA-wood preservatives. Draft Report to the Swedish Chemicals Inspectorate (1997) (included in the Report)

11. Breslin V.T. and Adler-Ivanbrook L., Release of copper, chromium and arsenic from CCA-C treated lumber in estuaries. Estuarine, Coastal and Shelf Science 46 (1998) 111-125.

12. Weis J.S., Weis P. and Proctor T. Copper, chromium and arsenic in estuarine sediments adjacent to wood treated with chromated-copper-arsenate (CCA). Estuarine, Coastal and Shelf Science 36 (1993) 71-79.

13. Weis J.S. and Weis P., Transfer of contaminant from CCA treated lumber to aquatic biota. J. Exp. Mar. Biol. Ecol. 161 (1992) 189-199.

14. Weis J.S. and Weis P., Trophic transfer of contaminants from organisms living by chromated-copper-arsenate (CCA)-treated wood to their predators. J. Exp. Mar. Biol. Ecol. 168 (1993) 25-34.

List of documents made available to the Scientific Committee on Toxicity, Ecotoxicity and the Environment via its Secretariat to help it reach the opinion requested by the services of the Commission on the report by WS Atkins International Ltd (vol. B) "Assessment of the risks to health and to the environment of arsenic in wood preservatives and of the effects of further restrictions on its marketing and use"

CSTEE/98/6

Assessment of the Risks to Health and to the Environment of Tin Organic Compounds and of Arsenic in Certain Biocidal Products and of the Effects of Further Restrictions on their Marketing and Use

Draft Final Report

Volume B :

Assessment of the Risks to Health and to the Environment of Arsenic in Wood Preservatives and of the Effects of Further Restrictions on its Marketing and Use

February 1998 - WS Atkins International Ltd

CSTEE/98/6 - Rev. 1

Assessment of the Risks to Health and to the Environment of Tin Organic Compounds and of Arsenic in Certain Biocidal Products and of the Effects of Further Restrictions on their Marketing and Use

Final Report

Volume B :

Assessment of the Risks to Health and to the Environment of Arsenic in Wood Preservatives and of the Effects of Further Restrictions on its Marketing and Use

April 1998 - WS Atkins International Ltd

CSTEE/98/5 - Add. 1

CSTEE/98/6 - Add. 1

Annex II to contract with WS Atkins International (Tin, Arsenic)

Technical Annex description.

CSTEE/98/6 - Add. 2 (Doc. III/3508/98)

Review of Directive 89/677 on arsenic.

CSTEE/98/5 - Add. 3

CSTEE/98/6 - Add. 3

Council Directive of 21 December 1989 amending for the 8th time Directive 76/769/EEC on the approximation of the laws, regulations and administrative provisions of the Member States relating to restrictions on the marketing and use of certain dangerous substances and preparations (89/677/EEC) - OJEC No. L 398/19 of 30.12.89.

CSTEE/98/6 - Add. 4

Comments by the Netherlands on the risk assessment for arsenic in wood preservatives - E.U. - DG III Report by WS Atkins International.

CSTEE/98/6 - Add. 5

Dutch comments on EC-DGIII document on arsenic by WS Atkins, March 1998.

Netherlands' comments on the Draft final report Volume B: Assessment of the risks to health and to the environment of arsenic in wood preservatives and of the effects of further restrictions on its marketing and use, by WS Atkins, February 1998.

CSTEE/98/6 - Add. 6

Comments on the Draft Final Report Volume B : "Assessment of the risks to health and to the environment of arsenic in certain biocidal products and the effects of further restrictions on their marketing and use".

Federal Environmental Agency of Germany - Umweltbundesamt Berlin - 24/4/1998.

CSTEE/98/6 - Add. 7

Arsenic in the aquatic environment - speciation and biological effects.

Exemption Substances Project - Lars Landner - Swedish Environmental Group –KEMI report No 2/98.

CSTEE/98/6 - Add. 8

Arsenic contamination in the environment due to the use of CCA-wood preservatives.

Exemption Substances Project - October 1997 - KEMI (pre-print version).

CSTEE/98/6 - Add. 9

Compilation of Data from Nordic Reviews on Arsenic in Impregnated Wood.

COWI - Nordic Council of Ministers/Norwegian Pollution - Control Authority - April 1998.

CSTEE/98/6 - Add. 10

Consequences of usage restrictions for CCA-type wood preservatives.

Exemption Substances Project - October 1997 - KEMI (pre-print version).

CSTEE/98/6 - Add. 11

Swedish comments to the first interim report by WS Atkins October 1997 on arsenic in wood preservatives.

28 November 1997 - KEMI.

CSTEE/98/6 - Add. 12

CSTEE/98/5 - Add. 12

Review of Directive 89/677 on Arsenic and Tin (Inter Alia) "Swedish comments to First Interim Reports by WS Atkins on Assessment of the Risks to Health and to the Environment of Tin Organic Compounds and Arsenic in Certain Biocidal Products, October 1997".

28 November 1997 - KEMI.

CSTEE/98/6 - Add. 13

Swedish comments to : Interim Report Part B by WS Atkins December 1997 on "Arsenic in certain biocidal products - Effects of further Restrictions on the Marketing and Use".

KEMI - 26 January 1998.

CSTEE/98/6 - Add. 14

CSTEE/98/5 - Add. 17

Review of Directive 89/677/EEC on Arsenic and Tin (Inter Alia).

11 March 1998 - KEMI.

CSTEE/98/6 - Add. 15

Adoption by written procedure of opinion on Arsenic.

Fax from Dr. Lambré (INERIS) 19/6/1998.

CSTEE/98/6 - Add. 16

Adoption of the opinion on Arsenic.

Fax from Dr. Tarazona - 22/6/1998.