The original classification of the Burgess Shale fossils and the subsequent reclassification has been extremely controversial, because Stephen J. Gould used the original classifications as his support for a new evolutionary model in his book Wonderful Life (1989).
March of Evolution, from http://www.wf.carleton.ca/Museum/man/evnman.html
Gould believed that two incorrect icons of evolution were in use at the time. The first is known as the “March of Progress” or “Linear Progression”. Such an image suggests that human evolution proceeded by making steps or rather improvements, which ultimately led to our own species. Such an interpretation is flawed because it does not reflect the random nature of evolution and assumes directionality. The “March of Progress” is created by abbreviating the hominoid family tree and including only those species that are believed to have genetically contributed to Homo sapiens sapiens. This view ignores the existence of species such as Australopithecus bosei and Australopithecus robustus, which are a few of the hominoids whose diet consisted of celluose-rich plant matter.
Cone of Increasing Diverisity, from http://geosrv01.bgsu.edu/Yacobucci/Geology%20105/Outlines/Paleozoic.htm
The second model of evolution that Gould argued against in Wonderful Life was the cone of increasing diversity. Such a view implies that there was less diversity in the past and that diversity has increased with time.
Model of Decimation and Diversification, from http://geosrv01.bgsu.edu/Yacobucci/Geology%20105/Outlines/Paleozoic.htm
In contrast, Gould’s model of decimation and diversification hypothesizes that greater diversity existed in the past. With mass extinctions, a small number of the existing organisms pass through the filter of extinction and then rapidly diversify to fill the vacated niches. Gould believed that greater diversity in the past would be seen as an increased diversity of body plans in the past. He believed that the Burgess Shale, which originally contained so many organisms that did not fit into any existing group, proved this. While the model of decimation and diversification has been accepted, Gould’s conclusion that more diversity meant many more body plans and more phylums in the past has lost support with the reexamination of the Burgess Shale. Few fossils remain unclassified, and such fossils are often rare and thus not fully studied. Thus, the existence of a significantly greater number of body plans in the past is not supported.