appearance will greatly aid the mechanics of metamorphosis. The increasing precocity of gene action can explain adultation most effectively.

In conclusion, I do not know if recapitulation is a more common result than paedomorphosis. Stebbins' concept of increasing precocity and the predominance of increase in phyletic size argue, respectively, for a greater frequency of acceleration over retardation and of prolongation over truncation. Both would justify a belief in the dominance of recapitulation. But I am also convinced that the impression of an overwhelming dominance rests, ironically, on cases that have nothing to do with recapitulation, but only mimic it in the workings of von Baer's laws.

Dissociability and Heterochrony

Correlation and Dissociability

When we eliminate the numerous cases that only reflect von Baer's laws, we are still left with thousands of documented heterochronies and the conclusion that this process has dominated the evolution of many important lineages (Chapter 8). Is there any general observation about the nature of ontogeny that will help us to understand why acceleration and retardation are so common in phylogeny? I believe that the notion of dissociability provides such a key.

There is an old prejudice in biology, expressed as much in ancient notions of perfect harmony as in Cuvier's correlation of parts. It dictates that organisms be treated as sublimely integrated systems, always changing in perfect coordination. In phylogenetics, this prejudice surfaced as the "harmonious development of the type." Fossils were often denied an ancestral status if they showed a mixture of primitive and advanced characters, rather than perfect intermediacy between an earlier ancestor and its modern descendant. Ernst Mayr (personal communication) tells of an argument he had in the 1940s with Franz Weidenreich on the status of Australopithecus. Weidenreich would not allow this small-brained but fully erect hominid into our lineage because it was an inharmonious type. The concept of "mosaic evolution," developed by Louis Dollo and others, refuted the notion of harmonious development by affirming that individual organs could have independent phyletic histories, despite the evident correlation of parts within any organism.5 Correlations are no more immutable than species themselves.

The notion of primary harmony motivated Haeckel's view of recapitulation, while the recognition of dissociability prompted the first