558 MR. T. GOODEY ON THE SKELETAL [Mar. 15, 



view of a. longitudinal vertical section of this portion (fig. 16), it 

 is at once evident that in this specimen the formation of centra 

 has proceeded much farther than in the smaller and, doubtless, 

 younger specimen. The constrictions beneath the imperforate 

 basidorsals extend well towards the centre of the notochord, 

 whilst those beneath the perforate basidorsals are quite well 

 developed, being much larger and more sharply defined than the 

 corresponding constrictions in the smaller specimen. In both 

 large and small constrictions also calcification has taken place, so 

 that rounded Vs are shown in the sectional view. 



In the caudal region, then, we have well-marked, incipient, 

 cyclospondylic centra. Another point which is especially inter- 

 esting is that the calcified rings become most strongly developed 

 in that part of the notochord below which the ventral lobe of the 

 caudal fin has its greatest depth. One would expect this to be 

 the case when the occurrence of centra and calcified bands is 

 considered from the point of view of the mechanical importance 

 of this region. If the diplospondylic condition of the vertebral 

 column is concerned, as Ridewood (28) suggests, with flexibility, 

 then we have here double the number of parts capable of articu- 

 lation with one another that we should have if the monospon- 

 dylic condition obtained in its place. Since also, the greatest 

 purchase on the water is obtained by the tail where the caudal 

 tin is deepest, we should expect to find here not only flexibility 

 provided for bub also a provision for increased strength in the 

 skeletal supporting tissue. This is indeed what we actually find, 

 for, as pointed out above, the diplospondylic condition is found 

 here and the incipient centra are most strongly developed and 

 calcified over the deepest part of the caudal fin. 



Region 4. — This short region is of particular interest, because 

 it has not been figured and described before. The specimen which 

 Garman figured had lost the extreme tip of the tail. Fig. 17 

 represents it natural size, and from this it is seen that the 

 vertebral column is a gradually tapering structure which remains 

 segmented up to the end, no urostyle being present. The noto- 

 chord has but very slight growths of the arcualia round it, and in 

 two parts it is slightly segmented externally. In the dorsalia at 

 the commencement of the region can be recognized both basi- 

 dorsals and interdorsals. Very soon, however, the distinction 

 between them becomes lost and they apparently fuse with each 

 other to form small, irregularly shaped pieces, which, towards the 

 extreme tip, are rhomboidal in outline and are of varying lengths. 

 The dorsal radial elements, which are borne on the dorsal side of 

 the neural arches, are also of variable length. They do not corre- 

 spond segmentally with the dorsalia, and as we approach the 

 extreme tip of the tail they become comparatively long. The 

 hsemal spines are the most regular in shape and occurrence, 

 being, with one or two exceptions, of the same width up to the 

 end of the tail. 



The point of special interest in this region is the disposition 



