48 Ovaries of the Gar, Lepidosteus 



In Europe a number of workers, chiefly German, have studied 

 the composition of roe and caviar. The earlier literature is 

 given in Atwater's monograph on the food fishes of America 

 (1888). More recent papers are those of Buttenberg (1900-02), 

 Rimini (1904), Farnstein (1903-04), Albu and Neuberg (1906), 

 and Weitzel (1916). Tangl and Farkas (1904) followed the changes 

 in composition of the fish egg with development of the embryo. 

 Solberg (1906) analyzed codfish roe. Konig and Grossfeld (1913) 

 made a very complete study of the use of fish roe as food, taking 

 up in considerable detail the proteins, fats, and extractives. 



Material and Methods. 



Our samples of gar ovary were obtained from twelve specimens 

 t)f Lepidosteus platystomus and one sample of Lepidosteus osseus. 

 We did not secure samples of the larger southern species, the 

 -alligator gar, Lepidosteus tristoechus. 



The method followed is essentially that perfected by Janney 

 (1916) and used by C. H. Greene (1919) in his study of the 

 extractives of the muscle of the king salmon. The samples of 

 tissue for analysis were in all cases taken fresh, generally while 

 still physiologically alive. Samples of the ovaries were placed 

 in weighed glass-stoppered bottles, weighed at once, and trans- 

 ferred to casseroles and extraction with hot alcohol was begun at 

 once, or they were covered with 95 per cent alcohol and sealed 

 with hard .paraffin for transportation. 



Samples for the determination of water were taken at the 

 same time and weighed at once, then dried to constant weight 

 at 105C. Water determinations were generally made in duplicate. 



The analysis of the samples was carried out in the following 

 manner: the sample and the alcohol covering it were transferred 

 quantitatively to a porcelain casserole of about 300 cc. capacity. 

 The alcohol was brought to boiling to complete the coagulation 

 of the proteins. It was then poured off into a wide mouthed 

 short necked flask, and placed on the water bath. Subsequent 

 alcohol and water extracts were placed in this flask, which was 

 kept on the water bath. If the sample were very fat, two or 

 three preliminary extractions with ether were made. The sample 

 was next extracted with boiling water in 50 to 100 cc. portions, 

 at least eight extractions being made. The residue was then 



