In the simpler forms, Chætopoda and Gephyrea, the supraœsophageal ganglia are usually stated to be formed as an unpaired thickening at the apex of the præoral lobe, which in most cases becomes subsequently bilobed.
In the Arthropoda the unpaired præoral lobe of the Chætopoda is replaced by the so-called procephalic lobes, which are themselves bilobed; and the supraœsophageal ganglia are formed of two independent halves; further complications in development are also generally found.
There is not as yet sufficient evidence to decide whether the supraœsophageal ganglia were primitively developed continuously with, or independently of, the ventral cords.
The ventral cord appears in the embryo as two independent unsegmented strands, although in a few cases (some Crustacea and Gephyrea) these cords, by an abbreviation in development, arise as an unpaired median thickening of the epiblast.
The form of nervous system of the Chætopoda, Arthropoda, and Gephyrea is clearly therefore to be derived, as was first pointed out by Gegenbaur, from a more or less similar type to that now found in the Nemertines; and as suggested in the chapter on larval forms (vide p. 378) may perhaps be derived from the elongation of a circular ring, of which the anterior end has become developed into the supraœsophageal ganglia, the lateral parts into the two lateral strands, while the posterior part persists in some forms in the junction of the ventral cords above the anus (Enopla and Peripatus).
Mollusca. While study of the anatomy of the nervous system of the Mollusca, especially of certain primitive genera (Chiton, Haliotis, Fissurella, &c.) leaves little doubt that it is formed on the same type as that of the groups just spoken of, the development, so far as our imperfect knowledge enables us to make definite statements on the subject, is somewhat abnormal[156].
In the Gasteropoda and Pteropoda the supraœsophageal ganglia appear most probably to be developed either as paired thickenings of the epiblast of the velar area, or as invaginated pits of the velar area, which become detached from the surface, and then become solid (Hyaleacea and Limax). In either case the supraœsophageal ganglia appear to be developed quite independently of the pedal ganglia. The latter, as might be anticipated, are earlier in their development and more constant than the various visceral ganglia; and, if the views above expressed are correct, are homologous with the ventral cord of the Chætopods and Arthropods. Their actual development is very imperfectly known.
The most precise statements on the subject, viz. those of Bobretzky and Fol, would lead us to suppose that they arise in the mesoblast, but it seems more probable that they are formed as thickenings of the sides of the foot.
In the Cephalopods all the ganglia are stated to be differentiated in the mesoblast (Lankester, Bobretzky).
Hatschek[157] has recently given a detailed description of the development of the supraœsophageal and pedal ganglia of Teredo. He finds that the former ganglia arise as an unpaired thickening of the epiblast in the centre of the velar area, and the latter as an unpaired thickening of the epiblast of the ventral side of the body between the mouth and the anus. The two ganglia would thus seem to be disconnected with each other in their development.