These two studies clearly show that these three RPTPs play important roles during several steps of motoneuron axon guidance and suggest many questions. Why does each mutant affect only certain axons? Why are Dlar and dptp69D dptp99A not more fully penetrant? Partially redundant function and expression of different RPTPs at different levels on distinct subsets of axons could explain both. It will be interesting to check all the mutant combinations (eventually including dptp10D) to see whether different sets of axons are affected and to see how these sets correspond to the details of expression of each RPTP. It will also be important to analyze the defects at the level of single growth cones by using intracellular dye injection into the RP cells, the identified neurons whose axons make up most of SNb. What are the downstream molecules involved in axon guidance? One expects a panoply of adaptors, cytoplasmic kinases, and small G proteins just as found in the RPTK pathways that control cell fate. Indeed, the cytoplasmic PTK abl has been implicated in axon guidance in the CNS, the transmembrane glycoprotein gp150 has been shown to interact with and be a substrate for DPTP10D (Tian and Zinn 1994xTian, S.S and Zinn, K. J. Biol. Chem. 1994; 269: 28478–28486PubMedSee all ReferencesTian and Zinn 1994), and the SH2/SH3 adaptor protein encoded by dreadlocks has recently been shown to be necessary for axon guidance by fly photoreceptor axons (Garrity et al. 1996xGarrity, P.A, Rao, Y, Salecker, I, McGlade, J, Pawson, T, and Zipursky, S.L. Cell. 1996; 85: 639–650Abstract | Full Text | Full Text PDF | PubMed | Scopus (205)See all ReferencesGarrity et al. 1996). Finally, since the RPTPs seem to act in the growth cones, are they in fact ligand-gated phosphatases? Ligands might be found using biochemical methods, or perhaps a genetic interaction screen. With these two papers and other recent results in growth cone guidance, asking about entire signaling pathways is no longer just pie in the sky.