study of Source Factors of Solnbles by Analysis of Variance 



Area Effects on Conqjosition of Solubles 



The remainder of the statistical analysis was devoted to a study of 

 the effect of areas, processes, and month of sample collection on Tarious 

 characteristics of the condensed solubles. The statistical analysis in- 

 dicates that the number of samples from any single area of plant location 

 is too small to justify any definite conclusions as to the effect of area. 

 This conclusion is particularly true for areas $ and 6, in the Gulf of 

 Mexico, In the east Gulf (area 5) for example, 3 of the U samples came 

 from different storage tanks at the same plant on the same date, and all 

 of the west Gulf samples were collected at the same time, U of them from 

 2 plants. This may result in what should be considered as a processing 

 factor being listed as an area effect. 



As an example of other interrelations of sample history and solubles 

 properties, the plants in the South Atlantic and Gulf areas, have diffi- 

 culty "cooking" their solubles to the desired 50-percent solids concentra- 

 tion, table 6, This may be related to the high air and water temperatures 

 in these areas, since these high temperatures result in "soft" fish as a 

 normal condition, and some of the fish are in pretty poor condition before 

 they are pumped from the boat hold. Even when the fish are in fairly good 

 condition when landed, the press water is often held in temporary storage 

 or in settling tanks until partial protein breakdown occurs. Some of the 

 plants in these areas use a gravity separation process and this makes it 

 much easier to separate the oil. It also reduces the viscosity of the 

 finished solubles so that the stickwater can be handled in evaporators 

 of the air-stream type, all of which were located in the Southern areas. 

 Another effect is to in'irease the ammonia content as a result of this pro- 

 tein breakdown (table 6), On the other hand, there is one atypical plant 

 in each of the Gulf of Mexico areas. These two plants process stickwater 

 almost directly from the centrifuges, and it will be noted in the table 

 on page 9 of the Research Report that these plants, code nos, EG 2 and WG 2, 

 produced condensed solubles with a lower ammonia content, higher fat content 

 and higher viscosity as compared with other samples from these areas. There 

 is no explanation that can even be suggested to account for the apparent 

 area differences in the total ash content. 



Effect of Process 



The section headed "Process Means" needs an explanation of the code 

 numbers, since only two processes, the multistage vacuum evaporators and 

 air-stream evaporators, are involved. The vacuum-evaporated samples are 

 further divided into (1) sarnies drawn from storage tanks and (2) samples 

 taken directly from the process line from evaporator to storage. Only four 

 samples were obtained from air-stream evaporators, and all were from storage 

 tanks. 



- 11- 



