365 
mass of as yet unossified cartilage (sd Fig. 1) which represents the 
undifferentiated diapophyses of these two vertebre. No sign of dupli- 
city can be detected in the cartilage in earlier embryos than these, 
since when chondrification begins in the skeletogenous tissue it pro- 
ceeds uniformly throughout the whole tract. It is only by a loose 
figure of speech that we may speak here of a fusion of the diapophyses. 
The tract forms the external boundary of the intervertebral foramen 
through which the ninth spinal nerve issues. In the second of the 
stages figured the diapophyses have enlarged considerably, especially 
in an antero-posterior direction at the lateral border, and the foramina 
between the neural arches 9 and 10 are still large. The posterior 
border of the centra 9 and 10 can be distinguished in younger (Fig. 1), 
but not in later embryos (Fig. 2). 
A compound sacrum appears to have been normal in the extinct 
Paleobatrachus. WOLTERSTORFF (41, Teil 1, p. 46) states that 
in Paleobatrachus diluvianus the diapophyses of the seventh, 
eighth and ninth vertebre are united into a disc or plate, to the 
edge of which the ilium is attached. In other species the diapophyses 
of the seventh vertebra may be free, or, as in Paleobatrachus 
Fritschii, they may acquire a connection with those of the eighth 
vertebra in old specimens. Portis (35, p. 1892) considers the disc 
of Paleobatrachus as formed by the confluent diapophyses of 
four vertebrae, the seventh to the tenth inclusive. He also describes 
in two other extinct Anura, Ranavus and Bufavus, a compound 
sacrum situated more posteriorly than in Paleobatrachus, and 
involving no vertebre anterior to the ninth. The only other living 
Anuran, besides Pipa, which normally possesses a compound sacrum is 
Pelobates. Here also the expanded diapophyses are the united diapo- 
physes of the ninth and tenth vertebra, and the limit is marked by the 
nerve foramen on the ventral surface. As in Pipa, convincing evidence 
of the compound nature of the sacrum is forthcoming in the earlier 
stages of its formation. Bruck (7) in his work on the development of 
the vertebral column of Pelobates confines himself almost entirely to 
histological detail and does not mention the fact. Howes (24, p. 275, 
Fig. 14), in figuring the vertebral column of a Pelobates larva, 
shows the diapophyses of vertebre 9 and 10 in contact distally, but 
distinct. As a matter of fact the marginal cartilage from its first 
differentiation from embryonic connective tissue connects the neural 
arches 9 and 10. It limits the nerve foramen distally, and remains 
unbroken for life. Seeing with what readiness and conviction the 
compound nature of the sacrum in Pelobates can be demonstrated, 
it is strange that ApoLPHI, who devoted so much labour to the in- 
