A defence of the calcichordates

Peterson (1994, Short Courses in Paleontology 7, Pal. Soc. & Univ. Tennessee; 1995, Lethaia 28) has claimed that the calcichordate theory of the origin of chordates can be dismissed on purely methodological grounds, without evaluating the anatomy of the calcichordates. He claimed this on the basis of his finding that the interrelationships for extant deuterostomes obtained by a PAUP analysis were not altered by including calcichordate fossils in the analysis, allegedly coded according to the anatomical interpretations of the calcichordate theory. This result, however, was due (1) to acceptance of dubious homologies among extant deuterostomes, (2) to reliance, when asserting that the fossils made no difference, on a majority-rule tree based on 13 equal-shortest trees; and (3) to the use of numerous multi-state characters in coding, which had the effect of denying, a priori, many of the homologies claimed by the calcichordate theory between fossil and extant forms. When these errors are corrected, a new PAUP analysis yields a tree consistent with the calcichordate theory. Peterson's claim therefore fails.