DETERIORATIVE CHANGES 353 



crease in the proportion of types of Pseudomonas (P. fragi, P. putida and 

 similar species) commonly found on all spoiling ''animal protein" foods 

 held under refrigeration, i.e., meat, turkeys, etc. 



Penetration of Tissues. The degree of penetration of the surface bac- 

 teria into the tissue of the fish is not well established. Three possibilities 

 exist: movement inwards from cut surfaces such as belly flaps and from 

 gills — possibly mainly via the major blood vessels, movement outwards 

 from the gut or gut cavity, and movement inwards from the skin surface. 



The small amount of experimental evidence available indicates that 

 while all three possibilities may be realized to some extent, general pene- 

 tration from the surface inwards is most significant. Some evidence 

 strongly suggests that significant penetration through the belly wall may 

 also take place. However, movement of bacteria into tissues is appar- 

 ently a slow process and multiplication within tissues is not very rapid. 

 Even when the fish is clearly spoiling, only small numbers of bacteria 

 can be found in the deep muscle. Consequently, the most widely accepted 

 view is that most of the primary activity of the spoilage bacteria occurs 

 in the surface layers of the fish, and the total effect is largely due to sec- 

 ondary diffusion of bacterial enzymes and products into the deep tissues. 



Penetration of bacteria into cut fish and fillets is certainly more rapid 

 than in the case of whole fish but is still sufficiently slow to justify the 

 application of the above theory to these products also. It is possible that 

 the reason for the primary surface action of the spoilage flora is its obli- 

 gate aerobic nature. 



Bacterial Metabolites. While comparative experiments with sterile and 

 contaminated fish muscle samples clearly confirm that bacteria are re- 

 sponsible for the gross changes in fish associated with spoilage, the actual 

 relationship between bacterial activity and chemically detectable changes 

 is by no means clear. This may, in fact, be a reflection of our inadequate 

 understanding of the chemistry of spoilage. Even in the apparently clear- 

 cut situation of trimethylamine oxide (TMO) reduction, which has been 

 attributed to the activity of bacterial triaminases, there must exist some 

 doubt since (a) the total percentage of bacteria in the spoilage flora able 

 to reduce TMO is very small and (b) there is no evidence of an increase of 

 TMO reducing bacteria during spoilage. The predominant Pseudomonas 

 types constituting the spoilage flora do attack peptides and similar sub- 

 protein N-containing substances very vigorously, yielding amounts of 

 NH3 and other amine substances. However, there is some question con- 

 cerning their ability to attack native fish protein. Typical end products 

 of protein degradation are significantly absent from spoiling fish, appear- 

 ing only after putrefaction has set in. Moreover, there is no apparent 

 increase in the proportion of casein and gelatin digesting bacteria present 

 in the spoilage flora during storage of fish in ice. 



