FISHERY BULLETIN: VOL. 78, NO. 2 



trolled the degree of retention of TMMD in the 

 aqueous phase or its release to the toluene phase. 

 The use of 1 ml 3.7% FA and 4 ml 15.9 fxg DMA- 

 N/ml results in a large excess of FA, about 500 

 times over that required (1 FA to 2 DMA). Conse- 

 quently, in the absence of added FA where only 

 the stoichiometric amount of FA was present from 

 TMMD, an equilibrium was established in the 

 KOH systems that favored the formation of DMA 

 and its rapid extraction by toluene. A different 

 equilibrium was formed in the K2CO3 system that 

 favored the release of TMMD and extraction by 

 toluene. 



Extraction of TMA 



The extraction of TMA under various conditions 

 of base, temperature, and FA was examined. A 

 standard solution of TMA  HCl in 5% TCA was 

 prepared (15.9 AAgTMA-N/ml, i.e., 5.3 mg TMA-N/ 

 100 g). This concentration was chosen as it is near 

 the point of unacceptable quality for fish. In the 

 carbonate method (Table 5), the extraction of 

 TMA was highly dependent upon temperature 

 and would result in a lack of precision unless the 

 temperature was controlled as suggested by Hashi- 

 moto and Okaichi (1957). Absorbancies were not 

 as dependent upon temperature in the 25% KOH 

 method as with K2CO3 and were nearly indepen- 

 dent of temperature with 45% KOH. The slightly 

 lower absorbancies with FA present than if not 

 present might be caused by an impurity of DMA or 

 an interference from FA even though FA would 

 not be expected to react with a tertiary amine. As 

 stated in the section on DMA, FA might best be 

 left out in the carbonate method, i.e., only 10% less 

 TMA was extracted than was extracted in the 45% 

 KOH method. 



To determine the conditions for maximum ex- 

 tractions of TMA, the same multiple extractions 



Table 5. — Trimethylamine hydrochloride: the absorbancies of 

 picrates in the trimethylamine test as affected by the three bases 

 used and temperature of extraction. Samples were extracted for 

 60 s with vigorous hand shaking using 4 ml of 15.9 /xg TMA-N/ml, 

 with formaldehyde ( + ) and without formaldehyde (0). 



Table 6. — Trimethylamine hydrochloride: the absorbancies of 

 picrates in multiple extractions in the trimethylamine test as 

 affected by the three bases used. Samples were extracted for 15 

 min at room temperature using 4 ml 15.9 /ng TMA-N/ml, with 

 formaldehyde ( +) and without formaldehyde (0). 



were done as with DMA. In the 45% KOH test 

 (Table 6), 97% of the TMA was removed in the first 

 extraction and the remainder was removed in the 

 second extraction. The first, second, and third 

 extractions removed 80, 17, and 3% with 25% KOH 

 and removed 72, 21, and 6% with 50% K2CO3. 

 Standard curves are assumed to compensate for 

 constant experimental errors such as slightly less 

 than 100% extraction of TMA, but the reliability of 

 the data would be questionable with the low 

 recoveries reported here for 25% KOH and 50% 

 K2CO3. Neither do standard curves compensate 

 for variable errors such as the observed strong 

 dependence on temperature of the extraction of 

 TMA in the 25% KOH and 50% K2CO3 methods 

 (Table 5). 



Comparative Analyses Using Fish Flesh 



Walleye pollock, Theragra chalcogramma, were 

 held in slush-ice for 9 d and filleted. Twelve 

 separate TCA extractions were made on a compos- 

 ite sample of the ground flesh. Each extract was 

 analyzed in duplicate by each of the three TMA 

 methods and the cold method. Portions of the 

 extracts were neutralized and analyzed for DMA 

 by Dowden's 1938 method, modified slightly by 

 increasing the time of extraction to 15 min on the 

 modified mechanical shaker. 



All methods (Table 7) resulted in similar stan- 

 dard deviations but the TMA values were higher 

 in the K2CO3 method than in the KOH methods 

 and the cold method of extraction gave the lowest 

 value. The absorbancy data at 22° C of Table 2 can 

 be used to approximate the degree of contribution 

 of DMA to the TMA values in Table 7. The flesh 

 contained 2.25 mg DMA-N/100 g (6.75 /^g/ml) 

 and would contribute different amounts to the 

 TMA value according to the method of analysis 

 employed. The data of Table 2 for K2CO3 (0.327 A 

 at 22° C using 15.9 ^ig DMA-N/ml) are equivalent 



470 



