198 MAYNIE R. CURTIS. 



case the reversal of the direction of the upward moving of the 

 hard-shelled egg after it had met and passed its successor might 

 result in a double egg with a yolk in the pointed end and a hard- 

 shelled egg in the blunt end. 



Parker (12) cites cases of "soft-shelled" eggs in the body 

 cavity described by Davaine (5) and Landois (10). Several 

 other cases are described by von Durski (6). A few cases of 

 membrane-covered or normal hard-shelled eggs in the body cavity 

 of apparently normal birds have been observed by the authors." 

 A previous description (3) of a bird which backed all her eggs into 

 the body cavity due to the fact that the oviduct ended blindly 

 at the posterior end of the isthmus has been briefly summarized 

 on page 196. It has also been shown by Pearl and Curtis (15) 

 that eggs are found in the body cavities of birds killed some 

 months after their oviducts have been ligated in the isthmus 

 or shell gland. If the ligature is in the isthmus the eggs are 

 membrane-covered but if all or part of the shell gland lies above 

 the ligature some or all of the eggs have shells. Since the egg 

 membrane and shell are formed in the caudal portion of the 

 oviduct and since in all the above cases where eggs were found 

 in the body cavity the funnel mouth was the only opening of the 

 oviduct into the body cavity, the egg must have been returned 

 up the duct and out through the funnel mouth after having passed 

 as far as the isthmus or shell gland. In two cases the authors 

 have observed hard-shelled eggs with no secretion around them 

 in the upper portion of the albumen-secreting region. Speci- 

 mens No. 8 and 9 described above were also hard-shelled eggs 

 found in the albumen-secreting region of the duct. In these 

 four cases also the eggs must have been returned from the shell 

 gland. 



So far as we know an egg has never been observed moving up 

 the duct nor have any movements of the duct been observed 

 which w r ould tend to force an egg in that direction. It has been 

 generally assumed (Parker 12, von Durski, 6, Hargitt 8, Patter- 

 son 13, and Pearl and Curtis 15) that the backward movement 

 of the egg is due to antiperistalsis. In the present paper no 

 assumption has been made as to the nature of the muscular 

 action. It seems possible that this may take some other form 



