QL and QR are generated through identical patterns of division from ABpl and ABpr, and they divide identically, to produce three neurons and two programmed cell deaths. Yet QL descendants stay near V5.a or migrate back into the tail, while QR and its descendants migrate anteriorly past the gonad into the vicinity of V2.a or the head. Five genes are known to influence the direction of migration of these neuroblasts. Recessive mutations in the genes
mab-5, 3), and
hch-1 cause both QL and QR to migrate anteriorly, while a semidominant mutation in the gene
lin-21 prevents QR and its descendants from migrating past the gonad. There are three reasons to suspect that
lin-21(
e1751) might be a gain of function allele of
mab-5. First,
e1751 has the opposite effect on Q migration. Second,
mab-5 and
lin-21 have not been separated in mapping experiments. 16 recombination events between dpy- 17
(e164) and
unc-32(
e189) failed to separate
mab-5(
e1239) and
lin-21(
e1751) (E. Hedgecock, personal communication). Third, a restriction fragment length polymorphism associated with
e1751 has been seen on two Southern blots probed with putative
mab-5 DNA sequences. (R. Hoskins, MRC pers. comm., repeated by M. Costa, UCSF) If
mab-5(
e1239) is a null allele of
lin-21, then it should behave like a deficiency when in trans to
e1751. Otherwise, since known mab- 5 alleles are recessive,
e1751/e1239 should resemble
e1751/+. Here are the positions of Q descendents in L2 heterozygotes. There were regularly two Q descendants in the body region illustrated: [See Figure 1] First, since
e1751/nDf16=
e1751/e1751,
e1751/+,
e1751 is likely to be a neomorphic mutation whose effects on Q migration can be alleviated by wild type gene activity. Second, the phenotype of
e1751/e1239 resembles that of
e1751/+, while the phenotype of
e1751/nDf16 is quite different, and resembles the
e1751/e1751 phenotype. This suggests, as argued above, that
e1751 is not an allele of
mab-5. However, it is always possible that
e1239, while null for the
mab-5 wild-type activities, retains activity in suppressing
e1751. Examination of more
mab-5 alleles over
e1751 will help rule out this possibility.