The trochoblasts in the pilidium larva break an ancient spiralian constraint to enable continuous larval growth and maximally indirect development

Table 1 Summary of cell-marking results

For each quartet by quadrant, the numerator corresponds to normal patterns observed—conforming to images shown in figures; denominator is the total number of scored larvae assignable to that quadrant, including those with deviant labeling patterns (e.g., missing a portion of, or having an extra part to the domain). Each type of experiment was replicated 2–4 times with separate pools of embryos
^aInjecting one cell at the four-cell stage apparently biases D-quadrant specification: the injected cell is biased against assuming the D-quadrant fate (i.e., we got no 1d² cases in any of these trials). All other experiments were done injecting dye into zygotes instead
^bThe conditions in these experiments somehow biased specification of the D quadrant, resulting in almost entire absence of 2d cases. Our best guess is that this bias comes about because the labeled cell necessarily faces the coverslip and may experience restricted respiration. In subsequent experiments, we were very careful to release the embryos promptly, and the results were no longer severely biased against 2d

ISSN: 2041-9139