extracted by the method of this paper to give 

 recoveries of 49 mg (237c), 56 mg (25%), 18 mg 

 (107f), and 6 mg (507c), respectively. Thus, the 

 official method of filtration resulted in an average 

 loss of oil and grease of 25% of the values deter- 

 mined by the proposed method. 



Two effluents (3 and 4) were precipitated by the 

 method in this paper but extracted by the Soxhlet 

 method and gave 16 and 5% low values, respec- 

 tively. In addition, contamination of the oil frac- 

 tion with Celite and fiber is apparent in the EPA 

 Soxhlet method and oil and grease values are 

 estimated to be 5-10 mg lower than reported. 



Discussion 



Different precipitation techniques were used in 

 developing this method and gave valid results for 

 specific waste effluents. For freshwater-processed 

 shrimp, Celite, alum (200 ppm), and Magnafloc 

 835A (2 ppm) resulted in complete precipitation in 

 about 15 min. The alum technique also worked on 

 waste effluents from saltwater-processed shrimp 

 and on snow crab, but precipitation was slower and 

 filtration was more difficult. In general, the hexa- 

 metaphosphate precipitation is the preferred 

 technique because it resulted in a more firm, dense 

 floe that filtered more rapidly than the alum 

 system. In addition, the soluble proteins along 

 with their oil content are recovered in the hexa- 

 metaphosphate precipitate and included in the 

 analysis. The soluble proteins generally are not 

 recovered with the alum system or by the EPA 

 method. Presumably, any reagent can be used for 

 precipitation provided there is no carry-over into 

 the oil fraction. Sulfuric acid was used to develop 

 this method, but it occasionally resulted in a dark 

 oil after drying. Consequently, the use of sulfuric 

 acid was discontinued in favor of acetic acid. The 

 proposed method should be tested further in 

 comparison with the standard EPA methods for oil 

 and grease to determine its applicability to other 

 fishery waste effluents. 



Literature Cited 



American Public Health Association. 



1971. Standards for the examination of water and 

 wastewater. 13th ed. Am. Publ. Health Assoc, Wash., D. 

 C, 874 p. 

 Environmental Protection Agency. 



1974. Oil and grease. //; Methods for chemical analysis of 

 water and wastes, p. 226-235. Environ. Prot. Agency, Off. 

 Technol. Transfer, Wash. D.C. 



Kelley, C. E., and A. W. Harmon. 



1972. Method of determining carotenoid contents of Alaska 

 pink shrimp and representative values for several shrimp 

 products. Fish. Bull, U.S. 70:111-113. 



Jeff Collins 



Pacific Utilization Research Center Kodiak 

 Utilization Research Laboratory 

 National Marine Fisheries Service, NOAA 

 P.O. Box 1638, Kodiak, AK 99615 



OCCURRENCE OF 



VOLATILE N-NITROSAMINES IN 



JAPANESE SALMON ROE 



Consumer interest and concern about food addi- 

 tives is as strong in Japan as in the United States. 

 The possibility that secondary or tertiary amines 

 and nitrites in fish roe products (sujiko) might 

 combine to produce A^'-nitrosamines, known car- 

 cinogens, has received much attention and pub- 

 licity. If the use of nitrites is curtailed in Japan, 

 American salmon canners would be hurt because 

 of loss of sales or decreased prices for roe sold to 

 Japanese processors operating in the Pacific 

 Northwest. The value of this business to the U.S. 

 salmon industry is from $10 to $15 million each 

 year. 



Investigations by Howard et al. (1970) and 

 Fazio, Howard, White, and Watts (1971) showing 

 trace quantities of A'^-nitrosodimethylamine 

 (NDMA) from samples of chub, sable, salmon, and 

 shad prompted the National Marine Fisheries 

 Service (NMFS) to be concerned about A^-nitro- 

 samines in smoked nitrite-treated fishery 

 products. This concern was shared by the National 

 Canners Association (NCA) in connection with 

 nitrite-treated salmon roe products. Various sam- 

 ples of salmon roe commercially produced in 

 canneries in the northwestern United States and 

 Alaska were obtained by the NCA for analysis of 

 volatile A^-nitrosamines. 



In addition to the analysis for nitrosamines 

 which was carried out by NMFS, samples were also 

 analyzed by NCA for residual nitrite and chloride 

 concentrations. The results of these findings are 

 presented in this report. 



Experimental 



Background 



For a number of years, Japanese companies 



683 



