182 THE VOYAGE OF H.M.S. CHALLENGER. 



length in one s^iecimen, tind 7^ in another. In the latter the greatest diameter of the 

 glandular portion of the stomach measured 2^ inches, while the gizzard measured l^ 

 inches from before backwards, and 1 inch from side to side. The external constriction 

 between the glandular and muscular portions of the stomach was well defined in both 

 specimens, the stomach externally closely resembling in form that of Eudyptes chrysocome 

 (PL XIII. fig. 3). The duodenum comes oif from the anterior wall of the gizzard. 

 In one of the specimens examined, there was a small csecal pouch, which projected 

 to the right immediately above the p)doric aperture. This pouch, which admitted the 

 point of the finger, was not recognisable in the other specimen of Spheniscus magellanicus, 

 nor indeed in any other species which I examined. Probably, therefore, it was an 

 individual peculiarity. The proventricular gland in one specimen formed almost a 

 complete belt, which on the right wall of the viscus measured 3 inches, while on 

 the left wall it did not exceed 2 inches in breadth. Between the two extremities of the 

 band was a space measuring ^th of an inch in breadth, which, however, was not, as 

 in Spheniscus demersus, entirely devoid of glands. Here the two extremities of the 

 proventricular gland were united by means of a small number of glandular foUicles, 

 which were irregularly disposed, and placed at a greater distance from one another than 

 is the case with those which form the greater portion of the patch. In a second specimen 

 the proventricular gland on the right wall of the stomach measured 3 inches, while on 

 the left it measured only 1 inch in breadth. In this specimen the gland formed a com- 

 plete belt, which completely surrounded the gastric cavity. The posterior border of the 

 gland in both specimens was straight, the diminution in its breadth on the left wall of 

 the stomach taking place at the expense of the anterior border, which sloped obHquely 

 from before backwards, and from right to left. In the second specimen, as in the first, 

 the glandular follicles were not so closely aggregated on the left as on the right waU of 

 the stomach. In this specimen the difference in the aggregation of the follicles was less 

 observable than in the first, and hence in it, as above remarked, the proventricular gland 

 really presented a completely zonular character. The pyloric orifice is defended by 

 several folds of mucous membrane. In other respects the organ agrees with that of 

 Eudyptes. 



That the gastric gland of Spheniscus magellanicus really differs in form from that of 

 Spheniscus demersus, I have con^dnced myself by the examination of several additional 

 specimens of both species. In the one it is triangular in form, and does not form a com- 

 plete zone, while in the other the belt of foUicles is complete. Whether, however, this 

 character is of sufficient weight to justify us in regarding these birds as specifically distinct, 

 seems to me doubtful, and I shall devote a few lines to the discussion of this question in 

 a future paragraph. (See p. 228.) 



In one specimen of Spheniscus magellanicus, the stomach was filled with a grey 

 pulpy mass, the composition of which it was impossible to determine. In a second the 



