474 DISCOVERY REPORTS 



percentage of fat is much higher than in the Blue whale and Bartenwal results given 

 by Backhaus and Schreibe respectively. This may point to a fundamental difference in 

 the composition of the milk of northern rorquals from southern ones, or be due to 

 physiological differences of the individual whales from which the samples were taken, 

 or again to errors incidental to sampling or analysis. It has been previously pointed out 

 in this note that the addition of formalin inhibits the breaking up of the protein masses 

 which occlude the fat globules and so tends to produce a low milk-fat result for a 

 formalin-preserved sample. Also Backhaus and Schreibe state that their samples were 

 (i) of a reddish tinge, (ii) slightly red colour, which points to a dilution of the milk 

 sample by blood. However, it is obviously impossible to comment at length on such 

 a small number of analyses. 



One very striking difference occurs in the milk-sugar figures of the Blue whale of 

 Backhaus and the Bartenwal of Schreibe. In the former the percentage of milk-sugar 

 is 5*63, whilst in the latter it is 0-38, which latter figure points to the supposition that 

 the sample was not taken from a freshly killed whale, as stated by Schreibe, and decom- 

 position of the milk-sugar has occurred, giving a low figure for this estimation and 

 consequently too high a figure for the water content. 



The milk-fat is described by Backhaus as non-solid at ordinary temperatures and 

 water-clear in colour. Schreibe describes it as yellowish, whilst I should describe it as 

 decidedly solid at ordinary temperatures and whitish in colour with a very faint yellow 

 tinge. At South Georgia, some whale milk was shaken up and the resulting "butter" 

 resembled soft lard in appearance and colour. 



The following is a brief resume of the methods used at South Georgia in the analysis 

 of whale milk. 



Specific gravity. By specific gravity bottle at 15*5° C. 



Total solids. 5 c.c. of the well-mixed sample were weighed out into a porcelain dish 

 and I c.c. of acetone added. The milk was evaporated to dryness on a steam bath and 

 was then dried to constant weight in a steam oven. 



*&' 



Ash. The residue from the total solids estimation was gently ignited and the ash 

 cooled and weighed. 



Fat content. Werner-Schmid process. About 10 c.c. of well-mixed milk were 

 weighed out into a flat-bottomed graduated tube fitted with a cork. 10 c.c. of con- 

 centrated hydrochloric acid were added and the tube and its contents were heated in a 

 water bath at about 60° C. for ten minutes, with constant shaking. The tube was then 

 rapidly cooled by immersion under running cold water. 30 c.c. of alcohol-free ether 

 were then added. The cork was then inserted and the tube shaken vigorously for two 

 minutes. When separation into two layers had occurred the ether layer was blown over 

 into a weighed flask. The ether extraction was repeated three times, with 20 c.c. of ether 

 each time. The ether was then distilled off on a hot-water bath and the flask containing 

 the milk-fat was dried in a steam oven until constant in weight. The fat was then re- 



