48 BULLETIN 150, U. S. DEPARTMENT OF AGRICULTURE! 



expenditure for equipment and economy in operation are claimed for 

 the process by its exploiters. From a priori considerations there 

 appears no reason why the process should not fulfill its promised 

 performance, though it does appear a little doubtful whether the oils 

 can be liberated sufficiently by maceration and washing without 

 bringing the material to such a fine state of subdivision that a great 

 deal would be lost in the water drawn off with the oil, or too long 

 a time would be required to permit the solids to separate by settling. 

 At present, this apparatus has received no trial in the actual commer- 

 cial rendering of salmon cuttings, and a positive opinion concerning 

 it is not justified. 



A comparison of the reports of analyses of salmon and menhaden 

 scrap, respectively, as reported in this paper on page 33, will show 

 that the amount of oil remaining in the salmon scrap is much 

 higher than that in the menhaden. While this may be due to the 

 difference in the respective methods of drying the two (an explana- 

 tion further suggested by the lower oil content of the two samples of 

 salmon scrap dried in hot-air driers and involving the supposition 

 that oils are volatilized in drying) , it also may be due to the fact 

 that the oils are not so easily recovered from salmon as from men- 

 haden. This constitutes an additional reason why some method, if 

 feasible, should be adopted whereby a more complete recovery of the 

 oil is possible. The limits of the press easily are reached. 



With the abandonment of the press, the adoption of a system in- 

 volving the use of an extractive recommends itself. The extraction 

 of the oils with gasoline theoretically should be quantitative, and the 

 exploiters of processes based on the use of this extractive claim a 

 very high efficiency. The method consists of cooking the material to 

 be rendered in closed retorts with steam. At the end of the cooking 

 the water in the material is evaporated under vacuum. When the 

 evaporation is complete, the dry residue is washed thoroughly with 

 gasoline, which removes all but about 1 per cent (more accurately, 

 1 per cent of the weight of the dry scrap, according to the claims 

 made for the process) of the oils present. The gasoline extract is 

 drawn off from the scrap and distilled. The oil remains as a resid- 

 uum, and the evaporated gasoline is condensed and recovered. It 

 is reported that there is but a slight loss in gasoline. An additional 

 advantage of the method is that all of the nitrogenous constituents 

 of the fish are saved, while in the other methods there is an indefinite 

 loss due to the solubility of certain of these in the water drawn off of 

 or expressed from the cooked fish and thrown away. A further 

 modification of the system, known as the Cobbwell system, is based on 

 cooking in oil the material to be rendered, the oil being obtained 

 from previous extractions. After cooking, the excess of oil may 

 be drawn off, when the remainder is extracted with gasoline. 



