used, with small amounts of added carbohydrate, 

 to modify fish flavor and odor. 



The use of ultrafiltration in the processing- of 

 soluble hydrolysates is another interesting ap- 

 proach which might improve product character- 

 istics. Rapid advances are being made in mem- 

 brane technology and ultrafiltration membranes 

 are probably available which could remove ex- 

 cess salt and possibly some undesirable nonpro- 

 tein nitrogen while retaining peptides and most 

 free amino acids. The concentration of the hy- 

 drolysate aflfected by ultrafiltration would also 

 reduce the evaporator or dryer requirements. 



The hygroscopic nature and low bulk density 

 of the spray-dried soluble powders are physical 

 properties which could stand improvement. 

 These problems would be avoided, however, if 

 the hydrolysates were marketed as concentrated 

 pastes. 



The partially soluble FPC that was prepared 

 from alewife press cake and described in this 

 report had very poor organoleptic characteris- 

 tics. These characteristics can undoubtedly be 

 improved by certain changes in the process and 

 operating techniques without significantly af- 

 fecting the low cost and good nutritive value of 

 the product. A truly bland product probably 

 will not be obtained, however, without the addi- 

 tional use of chemical solvents. 



There is presently a demand for a soluble pro- 

 tein concentrate which could be used in carbon- 

 ated beverages. No product has yet been de- 

 veloped which meets all the requirements. A 

 product that is soluble, nutritious, and relatively 

 bland in taste would command a premium price. 

 Considerable improvement in the flavor of sol- 

 uble hydrolysates of whole fish would be required 

 for this application. Enzymatic solubilization 

 of a solvent extracted FPC may be the means 

 for filling this need if the price is right. Steam 

 stripping of the FPC would be neither necessary 

 nor desirable. 



There are certain nonfood uses to which en- 

 zjonatically prepared FPC might be applied in 

 the near future. The financial risks of construct- 

 ing an FPC plant are reduced if the plant can 

 be used primarily for animal feed production 

 until an FPC market is developed. The use of 

 soluble FPC as a partial milk replacement for 

 calf feeding should be investigated. Another 

 possible use is in microbiological culture media. 

 Significant amounts of casein hydrolysates, rel- 



atively high priced, are presently used for this 

 purpose. 



Biological procedures would off"er advantages 

 over a solvent extraction process for either ship- 

 board processing or small-scale village industry 

 operations. The problems of solvent transpor- 

 tation, storage, and recovery would be eliminat- 

 ed. On board ship the freshly caught fish could 

 be comminuted and by proper control of pH and 

 temperature be preserved while undergoing par- 

 tial hydrolysis. Pumping of the slurry to tanks 

 on shore for final processing would be easy. A 

 simple biological process with minimal equip- 

 ment costs and suitable controls to insure prod- 

 uct safety would be appropriate for village in- 

 dustry operations. 



In summary, a variety of biological procedures 

 are available which offer interesting alternatives 

 to FPC production by solvent extraction. Ad- 

 vantages in terms of functional properties and/ 

 or costs of FPC products are possible in many 

 cases. A process has been developed for the 

 preparation of a totally water-soluble FPC with 

 good nutritive value through the enzymatic hy- 

 drolysis of whole fish. A partially soluble FPC 

 prepared from fish press cake has also been de- 

 scribed and preliminary cost estimates have been 

 made for both processes. The principal factor 

 limiting food applications of these and other 

 biologically prepared products are their taste 

 characteristics. Flavor improvement should be 

 the primary goal of additional research and de- 

 velopment. 



SUMMARY 



Fish protein concentrates (FPC) may be pre- 

 pared by various biological, as well as chemical, 

 procedures. Both microbial cultures and iso- 

 lated enzymes have been used for this purpose. 

 This report summarizes research that has been 

 carried out by the National Marine Fisheries 

 Service (formerly Bureau of Commercial Fish- 

 eries) on biological methods for FPC pro- 

 duction. The effects of various commercially 

 available proteolytic enzymes and processing 

 conditions on yields and characteristics of water- 

 soluble fish protein hydrolysates is presented. 

 Soluble products tended to be deficient in either 

 tryptophan or histidine depending on the pH of 

 hydrolysis. Hydrolysis of raw fish with an al- 

 kaline protease of Bacillus suhtilis at pH 8.5 or 



29 



