FISHERY BULLETIN: VOL. 71. NO. 3 



some consumers can be expected to eat dis- 

 proportionately large amounts of species which 

 they prefer and which in some cases will con- 

 tain more than this average. This interim level 

 is still in effect. 



The application of the 0.5 mercury interim 

 guideline has led to serious operational and 

 economic hardship for some segments of the 

 fishing industry. The swordfish industry has 

 been virtually eliminated, at least for the pres- 

 ent. The tuna and halibut industries are 

 operating voluntary self-regulation schemes 

 involving extensive testing, elaborate and ex- 

 pensive handling procedures, and product loss. 

 Press reports on other species have alarmed 

 consumers and adversely affected sales. It is 

 therefore worth examining this present guide- 

 line level to see if evidence exists that would 

 provide a basis for upward modification which, 

 while ensuring adequate consumer protection, 

 would alleviate the burden on the fishing 

 industry. 



Apart from reexamination of the assump- 

 tions, such as the proportion of total mercury 

 present as methylmercui-y in fish, two key 

 figures can be reviewed which together estab- 

 lish the basis for the 0.5-ppm level. The first 

 of these, the so-called acceptable daily intake 

 (ADD of 30 jUg methylmercury/day, has been 

 reviewed recently by the World Health Organi- 

 zation (WHO)/Food and Agriculture Organi- 

 zation of the United Nations (FAO) Expert 

 Committee of Food Additives, but no informa- 

 tion has thus far emerged which will permit 

 a modification of this figure. The second figure 

 is the high average consumption level of 60 g 

 fish daily. 



In order to examine the second assumption 

 more carefully, a computer program was written 

 to apply the results of a survey of fish con- 

 sumption patterns of over 1,500 families to 

 estimate the distribution of individual intakes 

 of any microconstituent for which the levels 

 in different species of fish are known. This 

 was then applied to predict the effect of alterna- 

 tive regulatory guidelines (maximum permitted 

 levels) on reducing the mercury intake of 

 consumers. The same program is directly appli- 

 cable to estimating the distribution of individual 

 intakes of any microconstituent, whether nutri- 



tious or potentially toxic, provided that the 

 levels are known in all the species of fish 

 consumed. 



The following report details the procedures 

 used and discusses the results as applied to 

 methylmercury. 



MARKET FACTS INC. SURVEY 



Commencing February 1969, Market Facts 

 Inc. of Chicago, 111., under contract with the 

 National Marine Fisheries Services (NMFS), 

 carried out a 1 yr survey of fish consumption 

 patterns of 1,586 U.S. households with a total 

 of 4,864 persons. The participants were selected 

 at random from a large panel designed to 

 parallel census data for the United States with 

 respect to population density and degree of 

 urbanization, geographic region, household in- 

 come, and age of panel members, providing a 

 reasonable cross section of other demographic 

 characteristics. 



The head of each household completed a diary 

 of fish purchases twice monthly for 12 mo. These 

 diaries reported purchases of fish and shellfish 

 products by item, weight and cost, numbers 

 of fish meals eaten away from home by item, 

 and the number of meals consumed at home 

 prepared from sport fish by species. Purchases 

 of meat, meat products, and meat substitute 

 foods were also recorded. 



This survey was undertaken to provide infor- 

 mation for economic analysis. Results have 

 been reviewed by Nash (1971) and by Miller 

 and Nash (1971). 



APPLICATION OF THE SURVEY 

 TO INTAKE ESTIMATES 



General Procedure 



In order to organize the data in the Market 

 Facts survey to enable intakes to be estimated, 

 they were first transformed into a data base 

 which lists for each family, estimates of the 

 total weights of each kind^ of fish consumed 



3 The term kind of fish is used here rather than species 

 since, although most of the kinds of fish in the data 

 base are identifiable species in the biological sense, some 

 refer to common names representing mixed species 



616 



