The Creationist dispute of whether ontogeny recapitulates phylogeny has never seemed one of any merit to me. Maybe my education is lacking, but I've always understood "ontogeny recapitulates phylogeny " to be more of a general principle rather than a strict law. Embryo development in the more recent orders quite obviously passes through stages closely resembling older orders, such as the oft mentioned gill stage in mammals. Since evolution takes a make-do approach, I assumed that features of earlier orders manifested themselves in a variety of ways, some developing fully and then being reabsorbed, like the tail for tailless mammals like ourselves, and others developing less fully or even just barely before making a turn toward something more mammalian, like the gill slits.
I've never read Haeckel, but if he actually claimed this general principle was an invariant law, and if he claimed the gill slits were actual gills, then I guess he was wrong.
About the accuracy of Haeckel's drawings, I think it was Gould who made a very interesting point. He showed drawings by professional artists who accompanied Napolean's early 19th century campaign in Egype side-by-side with modern photographs of the same objects. Just as striking as the similarities (those artists were
*good*) were the differences. For example, one common difference was the number of and expression of scroll lines. Gould (if it was Gould) argued that when faced with the problem of making many drawings quickly that artists were often forced to capture as much as possible the essence of the object rather than every particular detail. I tend not to accept this argument, because from what others have said I gather Haeckel's own words indicate he thought his drawings very accurate.
But whether Haeckel was right or wrong at the level of detail anyone cares to evaluate his work, the fact of the matter is that in general ontogeny absolutely does recapitulate phylogeny. Quite clearly, embryo development follows a course similar to evolutionary ancestors until a certain point is reached, and the particular degree of the similarity will vary from one characteristic to the next as the vagaries of evolution having to make do with what's available are felt.
--Percy