high temperature and short digestion time. A 

 French fishing cooperative at Nantes is presently 

 employing an enzjonatic procedure to convert the 

 better quality fish processing wastes to a sol- 

 uble product for use in calf-feeding rations. 



McBride, Idler, and MacLeod (1961), lique- 

 fied herring with proteolytic enzymes as well as 

 by ensilage and acid hydrolysis. Liquefaction 

 by ensilage was shown to be due to the naturally 

 occuring enzymes in the fish and not to bac- 

 terial action. The acid-hydrolysed product was 

 dark in color and the freed oil had a high free 

 fatty acid content. The most eflFective enzyme 

 treatment found was digestion of precooked her- 

 ring with pepsin at pH 2. Bromelin and Rho- 

 zyme B-6 were less eflFective. The main problem 

 with enzymatically liquefied herring was the for- 

 mation of a very stable oil-protein emulsion. 



A liquefied fish protein (LFP) has been pre- 

 pared by Higashi and co-workers in Japan for 

 possible use in infant diets. Fish were hydrol- 

 ysed with a streptomyces protease in the pres- 

 ence of a high boiling solvent such as xylene 

 or toluene. The liquefied product was steam dis- 

 tilled prior to concentration or drying. Studies 

 on the odor and peptide compositions of LFP 

 have been reported by Onishi and Higashi 

 (1968). 



Ehlert (1962) has hydrolysed mechanically 

 disintegrated fish using a dead culture of lactic 

 acid bacteria as the source of hydrolytic en- 

 zymes. It is claimed that superior oil separation 

 can be obtained and that the hydrolysate, ad- 

 justed to below pH 4, is quite stable. The Lumino 

 Feed Company of Denmark, the patent assignee, 

 is presently marketing an animal- feed ingredient 

 produced by this process. 



The relative activities of a wide variety of 

 commercially available proteolytic enzymes act- 

 ing on a fish protein substrate have been reported 

 by Hale (1969). Saha (1940) found that pep- 

 sin and trypsin digested undercooked fish more 

 completely than fish protein that was raw, fully 

 cooked, or fried. 



A fish protein concentrate has been prepared 

 in Japan by Dr. Tomiyama (1968) by autolysis 

 of sardines under acid conditions. No additional 

 enzymes were used but chloratetracycline was 

 added as a preservative. A breast milk substi- 

 tute which included 12 Sf autolysate and 52% 

 wheat flour was formulated and used successfully 

 in infant feedings. 



Duncan Law and David Crawford of the Ore- 

 gon State University Seafood Research Labora- 

 tory have liquefied fish by pumping the finely 

 comminuted material through a heat exchanger 

 during a short autolysis period. A process 

 whereby the autolysate is prepared and defatted 

 with a hexane-isopropyl alcohol mixture is de- 

 scribed in a patent by Lum (1969). 



A patent by Bedford (1957) describes the 

 hydrolysis of fish and fish material by reaction 

 with raw tuna viscera. Preliminary hydrolysis 

 at pH 5.5-6 is followed by hydrolysis at pH 8 

 to make full use of the visceral enzymes. It is 

 claimed that the addition of minor amounts of 

 NaCl and urea increase the digestion rate sub- 

 stantially. 



Ensilage of fish, consisting of hydrolysis by 

 native enzymes under sufficiently acid conditions 

 to prevent bacterial growth, has been studied for 

 many years in the Scandinavian countries. Fish 

 silage is produced industrially for animal feed- 

 ing. Work carried out at the Danish Ministry 

 of Fisheries with herring has been reviewed by 

 Hansen and SchmidtsdorflF (1968).' A novel 

 sieve plate apparatus was developed for the sep- 

 aration of suspended solids from hydrolysate and 

 oil fractions. 



Formic acid is prefered for fish ensilage be- 

 cause it makes the free amino acids less avail- 

 able to microorganisms and the product is stable 

 at a less acid pH. A silage process involving 

 serial additions of HCl, H2SO4, and formic acid 

 is the subject of a patent by a Polish company, 

 Centralne Laboratorium Przemyslu Rybnego 

 (1962). A method for preparing fish silage on 

 a small scale using a bitumen-coated metal drum 

 has been described in an Indian Fisheries Bul- 

 letin (Anonymous, 1964). 



Some information is available on the nutritive 

 value of fish protein hydrolysates. Sripathy et 

 al. (1963) studied the influence of degree of 

 hydrolysis on the nutritive value of fish hydrol- 

 ysates prepared with papain. A 17-hr hydrol- 

 ysate had a higher nutritive value than did two 

 short-term hydrolysates, but the protein efficien- 

 cy ratio (PER) was slightly below that of skim- 

 milk powder. The short-term hydrolysates had 



" Hansen, P., and W. Schmidtsdorff. 1968. Fish pro- 

 tein hydrolysate: A short review of the work carried 

 out in Denmark and suggestions for future research. 

 Presented to the International Association of Fish Meal 

 Manufacturers Annual Conference, Bremen, Sept. 1968. 



