In 1895, Roux set out the problems confronting the new subject of experimental embryology and commented that, although he and his peers intended to simplify what was clearly a very complicated set of events, they knew so little about development that they would be unable to elucidate the underlying mechanisms without a great deal of work. Moreover, because they were so ignorant, they could not know which approaches would be the most helpful in their attempts to gain understanding. The initial result of any research in the area would therefore be to make the situation appear even more complicated than it already was and it would take some time for the simplicities to become apparent.

After a century of work, there are few in the field who would say that enough of those underlying simplicities have yet emerged. Much of development remains complex and, with the tools of molecular biology now being applied to the subject, it is, by Roux's conjecture, likely to become more so, in the short term at least. This is not to say that the results of 100 years of research have in any way been fruitless: we now know a great deal about what happens as development proceeds and are beginning to understand the molecular nature of the cell–cell and cell–genome interactions that underpin embryogenesis.

However, one area where a substantial gap remains in our understanding, or so it seems to me, is morphogenesis, the study of the processes by which cellular organisation emerges in embryos.