Fat — The oven-dried samples from the moisture determination were com- 

 bined for the determination of the fat content. The dry material was 

 scraped loose, washed with ether into a large thimble, and extracted with 

 ethyl ether in a Soxhlet extractor for 8 hoiirso The solvent was evaporated 

 and the residue was dried for 30 minutes at 105° C. before being weighed,' 

 The major portion of the fat in most samples was from cream, milk, cheese, 

 butter, or cooking fat and was readily extracted, since these types of fat 

 do not oxidize appreciably during drying at this temperature© 



Protein — Triplicate samples estimated to require about 10 ml, of 0,2 N 

 hydrochloric acid each on titration were digested in 300 ml» Kjeldahl flasks 

 with 15 ml, of sulfuric acid, 7 to 8 grams of sodium sulfate, and a small 

 amount of copper sulfate as catalyst. The ammonia was collected in a 4-per- 

 cent solution of boric acid and titrated directly. The usual factor of 6»25 

 was used to calculate the percent of protein. 



Mineral matter — Duplicate samples of 5 to 15 grams were weighed rapidly 

 in tared porcelain evaporating dishes. The samples were dried at about 250'' C, 

 in a muffle furnace equipped with a heat control. The heat was increased to 

 350° to 400° C, to char and drive off volatile matter and was then held at 

 600° C, until the ash was light in color. Check analyses wei^ usually within 

 0,03 percent, except when bone fragjnents were present. 



Carbohydrate — Totail carbohydrate was calculated as the difference between 

 100 percent and the simi of the percent of moisture, protein, fat, and mineral 

 matter. This value includes the small amounts of crude fiber, organic acids, 

 aiid other miscellajaeous substances that are present. 



Food energy — The calories per 100 grams were calculated using the values 

 of 4» 9> and 4 calories per gram for protein, fat, and carbohydrate, respect- 

 ively. 



Serving portions — Information on the size of individual serving portions, 

 their food-energy value, and their content of protein, fat, and carbohydrate 

 is of practical value to dietitians or clinicians. These data are tabulated 

 in the right-hand section of table 1, 



The most commonly used estimate of a serving portion of fish, 125 grams 

 or 4 ounces, refers to an average serving of fried or broiled fillet or steak 

 cind has little significance for most of the dishes included in the ~table« 

 Fortunately, most of the dishes were prepared on a 6- serving-portion basis, 

 so that the size of an average portion was easily determined by dividing net 

 weight of the cooked dish by 6, Serving portions were estimated on the basis 

 of similsu? recipes for the few dishes for which data on net cooked weight were 

 lackingo 



Table 1 includes several samples of canapes and hors d'oeuvres which differ 

 from the main-dish-of-the-meal type of dish in that they are more of a "snack" 

 or extra. For this type, the serving portion was based on the unit weight per 

 snack rather than on a 6-portion basis. Although the portions are small, usuailly 

 20 to 40 grams, this type is frequently very high in energy value, so that even 

 these small portions contribute significantly to the dieto 



-4- 



