The methods described here of fertilizing stage VI oocytes are lengthy and quite difficult techniques. They would become more attractive if the success rate (i.e., the number of fertilizations compared to the numbers of matured oocytes) could be improved. An important step toward this for the host transfer technique would be to monitor carefully the status of mature Xenopus females ovaries in relation to cyclical HCG stimulation, so that we could predict more accurately whether stage VI oocytes are fertilizable. The in vitro technique would obviously be improved if oocytes could be fertilized without removing their membranes, perhaps by using oviduct extracts. So far, this approach has had only limited success. It seems that the rewards of using these techniques could be great, in terms of understanding the maternal contribution to development. Although our experiments have not yet shown that oocyte injection of DNA has any advantage over egg injection, it is clear that it is possible to make "mRNA-minus mutants" by this approach. In the message depletion experiments mentioned here, we targeted the cleavage of an mRNA which is of low abundance in the full grown oocyte, but preliminary experiments have shown that we can deplete more abundant messages and produce specific phenotypes. Of course such experiments need to be controlled to show that the effect is specific, and the best proof that this is the case is to rescue the effect with injection of the appropriate mRNA. Finally, it seems likely that the method can be used to study the function of both localized molecules, such as the putative primordial germ cell (PGC) or dorsal determinants, and more ubiquitous molecules such as cytoskeletal elements.