CATTDAL DIPLOSPONDTLT OF SHAKKa. 51 



The next double vertebra (6 • 7) is a repetition of (4* 5) except 

 tbat the three neural spines occupy the full length of the two 

 centra, and that the haemal arch on centrum 7 is completed by a 

 haemal spine (lis). The double vertebra (8 • 9) shows a further 

 departure. Both haemal arches are complete ; and between the 

 two narrow crural plates an intercrural, uunotched on its anterior 

 and posterior borders by nerve foramina, has been introduced. 

 The middle of the three neural spines is thus again exceptionally 

 placed, since it lies immediately over a plate, and not over a 

 boundary line as it did in (4 • 5) and (6 " 7). The double vertebra 

 (10 • 11) is the first of the normal diplospondylous series, and 

 differs from (8 • 9) only in the greater breadth of the crural 

 plates. A feature of special interest in vertebra (8 • 9) is that 

 the intercalated intercrural has only appeared on the left side. 

 As seen from the right side, this double vertebra is an exact 

 counterpart of (6 • 7). 



To summarize the above description : — The transition is effected 

 by steps taken in the following order — the broadening of the 

 crural plate and the introduction of an additional neural spine ; 

 the division of the centrum and crural plate, and the doubling 

 of the haemal process ; the intercalation of an additional inter- 

 crural between the two contiguous crurals. The great advantage 

 attained by this gradual transformation is obviously the avoidance 

 of excessively large or excessively small cartilages, while yet 

 securing a diminution, on the whole, of the antero-posterior 

 length of the elements. 



The most recent view on the subject of caudal diplospondyly 

 iu Selachians is that expressed by Dr. Gadow, who attributes 

 (4. p. 194) the phenomenon to the " chorda centra " being inde- 

 pendent of the " arcualia " and to the difference between the 

 metamerism of the centra and that of the arches. He explains 

 that in the middle of the trunk region of Heptanchus the centra 

 are double their normal length, extend through two myomeres, 

 and have four pairs of " dorsalia " *. After stating that it is the 



* It is not clear from Dr. Gadow's text whether he is here describing obser- 

 vations of his own, or is merely enlarging upon the unfortunate sentence of 

 Kolliker's (11. p. 199) " bei Heptanchus im hintern iind vordern Theil der 

 Wirbelsaule die Zahl der Wirbel um das doppelte grosser ist als in der Mitte." 

 But in either case, it should not escape notice that Hasse failed to confirm 

 Kolliker's observation, and stated (7. p. 46) that this anatomist was probably- 

 misled by the deceptive appearance of " ein dunkler, doppelt contourirter 



