78 University of California Publications in Zoology 



specimen may not be assignable to one or the other group, but a species (sample) can be so 

 assigned. 



Unfortunately, the pathway of character-state transformation cannot be analyzed by 

 outgroup comparison without making additional assumptions about the character. None of 

 the four outgroups used in this study, nor any other iguanian, possesses caudal vertebrae 

 with two pairs of transverse processes (Etheridge, 1967). Nevertheless, a close 

 correspondence between the beginning of the series of caudal vertebrae with two pairs of 

 transverse processes and the beginning of the series of autotomic vertebrae in iguanines 

 suggests that the latter might be used as the character instead. Unfortunately, not all 

 iguanines (nor all outgroup taxa) possess autotomic caudal vertebrae. Therefore, in order 

 to use this character I first must assume that the beginning of the series of caudal vertebrae 

 with two pairs of transverse processes in taxa that lack autotomy septa corresponds with 

 the beginning of the autotomic series in those taxa that possess autotomy septa. Second, I 

 must assume that the beginning of the autotomic series in taxa that lack vertebrae with two 

 pairs of transverse processes corresponds with the beginning of the series of vertebrae with 

 two pairs of transverse processes. 



Under these assumptions, outgroup comparison can be used with those outgroups 

 possessing autotomic vertebrae, but it provides ambiguous evidence concerning the 

 plesiomorphic condition of this character. The autotomic series of Basiliscus begins in a 

 range that has the tenth caudal vertebra in its midst. That of Gambelia begins posterior to 

 the tenth vertebra, while those of Enyalioides, Morunasaurus, and oplurines begin anterior 

 to the tenth vertebra. The polarity decision for this character will thus vary depending upon 

 the relationships among iguanines and the four outgroups. Because these relationships are 

 unknown, I withheld a decision on the polarity of this character in phylogenentic analysis at 

 the level of all iguanines. 



Lazell (1973:1-2) citing Etheridge (in litt.) distinguished Iguana from Cyclura by the 

 presence of "a low fmlike process above the neural arch of no more than six anterior caudal 

 vertebrae" in the former, compared to the "high, fmlike processes above the neural arches 

 of all the caudal vertebrae" in the latter. The processes in question are presumably 

 ossifications of the dorsal skeletogenous septum. When the remaining iguanine genera are 

 considered, there appears to be a continuum in the height of these processes rather than two 

 discrete morphologies, low and high. Even within an organism, the morphology of these 

 processes differs among the caudal segments. In most iguanines, the processes on the 

 anterior caudal vertebrae are merely thin, midsagittal extensions of the anterior edges of the 

 neural spines. Moving posteriorly along the column, apices form on the processes, and the 

 processes themselves are displaced anteriorly, sometimes becoming entirely separated from 

 their respective neural spines. The height of the processes increases, then gradually 

 decreases, moving anterior to posterior. Although the midsagittal processes generally 

 disappear short of the end of the tail, they are present (Fig. 3 8 A) well beyond the anterior 

 third of the caudal sequence (determined by vertebra number, not by distance from the 

 beginning of the tail) in all genera except Brachylophus and Iguana. The situation in 

 Brachylophus and Iguana differs from the one described above in that the processes are 



