12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I42 



oriented line of fusion or a line of shearing stress between the parts. 

 In both genera postterminal centrum 2 seems to have fused with one 

 or more of the upper hypurals. In Aphredodorus, hypurals 2 and 3 

 are closely associated and hypural i has its articular base in part over 

 the preterminal vertebra. 



Order Cyprinodontiformes. — As with so many other round-tailed 

 fishes, considerable difficulty has been encountered in interpreting the 

 caudal structure of the Cyprinodontiformes. Also as in other round- 

 tailed groups, the number of branched caudal rays varies greatly. For 

 example, there were 9 branched caudal rays in three specimens of 

 Chologaster examined, 13 in a specimen of Fimduhis, and 17 in a 

 Belonesox. 



The caudal skeletons of Amhlyopsis (fig. 3 A) and Chologaster 

 (fig. 3B) are presumably more "primitive" than those of other 

 members of the order. In these two genera the tip of the vertebral 

 column (presumably represented by PC 2) fuses with the upper 

 hypural plate. The lower half of the hypural fan seems best inter- 

 preted by comparison with the caudal skeleton of Aphredodorus (fig. 

 2D). If the Aphredodorus caudal structure really represents a gen- 

 eralized form of that seen in Amhlyopsis and Chologaster, then hy- 

 purals 2 and 3 have fused, and hypural i has lost all basal attachment. 

 (Seemingly hypural i frequently does this in round-tailed fishes, 

 judging by the illustrations in Whitehouse, 1910, and others.) 



The caudal skeletons of various nonamblyopsoid cyprinodonts have 

 been illustrated by Hollister (1940). Part of that of Fundulus is 

 shown here (fig. 3C). The simplest explanation for Fundulus is that 

 the terminal vertebra, postterminal centra, and all the hypurals have 

 fused into a single plate. If this is correct, the intervertebral articu- 

 lation between the postterminal centra i and 2 has been lost in 

 Fundulus, leaving it with one fewer vertebra at the base of the 

 caudal fin than Amhlyopsis and Chologaster. However, no sign of the 

 lost intervertebral articulation could be seen in the stained specimen 

 of Fundulus examined, or in the juvenile specimens of Mollienisia 

 illustrated by Hollister (1940, figs. 7-10). 



OrderLampridiformes. — The only caudal skeleton of a member of 

 this order available was one of Velifer (fig. 3D). Once again there is 

 the intervertebral type of articulation between postterminal centrum 

 (i?) and the structure behind it (a postterminal centrum fused with 

 a hypural?). This feature sets Velifer to one side of the lineage lead- 

 ing to the percoids, as far as caudal skeleton is concerned. The slight 

 resemblance to Amhlyopsis and Aphredodorus seems of highly dubi- 

 ous significance. 



