As a separate operation, on July ISth, a stady was made of the 

 relation between the size of a given seal and the paleness of its liver. 

 This was done as follows: One hundred carcasses of known length were 

 sajipled at random, (The "body length of every male fur seal killed on the 

 Prihilof Islands is measured routinely to the nearest inch.) From these 

 were selected the 10 having the palest livers. It was found that the mean 

 length of the 10 seals having the pale livers was virtually the same as 

 the mean length of the 90 seals having the darker livers (41.9 inches 

 versus ^2.0.) Thus, in tiiis sainple there was no perceptible difference 

 in size between seals having pale and those having dark livers. 



Analytical Procedure 



Each liver was ground in a meat chopper and a representative 

 fraction was stored in a screw~c^ jar at -20° F. About ^ hours before 

 analysis, the sample was thawed at 35° F. The sample was blended in a 

 Waring Blendor, and about 10 grams of the resulting material was accurately 

 weighed into a square shaking bottle of 180-millili ter capacity. A heaping 

 teaspoon of powdered pumice, exactly 50 milliliters of ethyl ether, and two 

 heaping teaspoons of anhydrous sodium sulfate were then added. In a shaking 

 device, the bottles were subjected to lUU one-inch strokes a minute for one 

 hour. They were then centrifuged and aliquots were removed for oil and 

 vitamin A deteimnations. All saii5)les were run in duplicate. 



To estimate the oil corjcentration, a lO-milliliter aliquoti por- 

 tion of the ether extract was pipetted into a tared, 50-milliliter beaker, 

 and the ether was evaporated on an air bath. Three minutes after the dis- 

 appearance of the ether, the beaker was removed and allowed to come to room 

 temperature. The wei^it of the beaker and its contents then gave the 

 remaining information necessary for the calculation of the oil concentra- 

 tion in the liver. 



To determine the vitamin A potency, a 1-milliliter aliquot por- 

 tion of the ether extract was diluted with isopropanol, and the optical 

 density of the solution was measured at 328 m ma, by means of a Beckman 

 spectrophotometer. The vitanan A potency was calculated as "spec" (spec- 

 tropho to metric) -anits using the formula 2000 x E (l percent, 1 cm, 328 m mu, 

 isopropanol) . It was recogoized that this gave only a gross estimate of 

 vitamin A, but it was thou^t adequate for the present purpose. 



Results 



The results of the analyses are presented in Tables 1 and 2. On 

 the average the livers weighed 2.^ pounds- and contained nearly h percent 

 oil and 2.U million "spec" units of vitamin A per po\md. The range in 

 potency was U,170 to 63^,000 "spec" units per gram of oil, and 0.lU6 million 

 to 26.6 million "spec" units per liver. (The potency of 63^,000 "spec" units 

 per gram of oil is the maximum thus fax encountered in our three-year study 

 of for seal livers.) The richest liver contained 182 times as nuch vitamin 

 A as the poorest. About one-half of the livers yielded an oil having a 



