Effect of Variable Intake on Body Composition 119 



6. Rats fed Friskies by way of control. See Table I for 

 composition of this ration. 



7. A high protein, high electrolyte group fed to measure 

 food consumption. 



8. A similar group to No. 7 but fed the low protein, low 

 electrolyte diet. 



At the end of the allotted period of growth (20-25 days) the 

 animals were etherized and 2 ml. of blood removed for 

 analysis either by heart puncture or tail incision. Killing was 

 accomplished by further ether exposure. The dead weight 

 was obtained and the abdominal cavity, thorax and skull 

 opened with heavy shears.* The whole body was then dried 

 in an oven at 85°-95° C. until a constant weight was reached 

 (4-5 days). During the drying process, the carcass was 

 further broken up with heavy shears. The disintegrated 

 carcass was extracted repeatedly with a cold mixture of equal 

 parts ethyl and petroleum ether and re-dried to constant 

 weight. The dried extracted carcass was then homogenized in 

 a Waring Blendor with 5 volumes of anhydrous acetone and 

 the solvent evaporated off and the material re-dried. This 

 process produces a fine homogeneous powder suitable for 

 quantitative analysis. The powder was stored in a desiccator. 



B. Chemical Methods 



Water. Calculated from weight loss after desiccation. 



Fat. During the course of the analytical work, the fat 

 extraction method used as applied to tissues by Hastings and 

 Eichelberger (1937) was examined for completeness of fat 

 extraction when applied to whole carcass. Powdered carcass 

 was exhaustively extracted in the Soxhlet apparatus serially 

 using ether, alcohol and chloroform. This process increased 

 the degree of fat extraction to the extent of 1-5-4 g. per 

 animal. Analysis of the material subjected to such extraction 



* Intestinal contents were not removed. Analysis of the total gastro- 

 intestinal tract and contents of similarly fed animals for water and fat-free 

 solids indicated that their inclusion does not appreciably alter the interpreta- 

 tion of the data. 



