104 
JANE I. llOBEkTSON. 
therefore, finally come in contact with one another round the 
left margin of the constriction that demarcates the proximal 
from the middle segments Finally, (in the adult horizontal 
position of the heart) the free inner borders of the first and 
third septa incline towards one another in the lumen of the 
bulbus, and the former becomes concave on its dorsal and the 
latter on its ventral surface (Text-fig. 22, B.E. 1 and B. R. 3.). 
In the transverse part of the bulbus, the transverse con- 
tinuation of septum 1 maintains this concavity, which, how- 
ever, in this region is directed posteriorly (PL 5, figs. 1, 2, 
and 4, Sjy. V.t.). 
The chief interest in tracing the development of the bulbus 
Text-fig. 22. 
Transverse section through the distal segment of the bulhns 
cordis at Stage 37. B. B. 1. Right hnlhns ridge (spiral valve). 
B. B. 3. Left bnlbns ridge (left longitudinal valve.) 
cordis in Lepidosiren is the information so obtained con- 
cerning the formation of the spiial valve. If the foregoing- 
observations are correct, then — at least in Lepidosiren — 
the bulbus valve owes its spiral form to the process of kinking 
and asymmetrical expansion of an elongated but originally 
straight bulbus, and not to any twisting or counter-twisting of 
that segment of the heart. Boas (2) comments upon the 
difficulty of believing that any twisting of the bulbus (conus) 
occurs in Cera tod us, but having only an adult specimen at 
his disposal for examination cannot suggest any other more 
satisfactory explanation for the presence of a spiral valve. 
It is of interest to note how closely the above account of the 
development of the bulbus cordis and its associated endo- 
thelial structures agrees with that given by Langer (16) and 
Creil (9) for amphibians and reptiles. 
