486 THE NERVOUS SYSTEM. 
indicate that this layer is ganglio-genetic. In the neighbour- 
hood of the pyramidal ganglion (Pl. XXIV., Fig. 1, x) and of 
the antennal ganglion the mantle is continuous with groups of 
large nerve cells, but whether the small round cells are developed 
from the mantle is still, I think, doubtful; nevertheless, I regard 
it as probable that they are in part at least. 
Viallanes continues his description of the development of 
the optic ganglia as follows: 
“At the same time that the layers of ganglion cells are pro- 
duced a layer of fibrillar substance appears, at first on the 
inner surface of the ganglion layers; it is, so to speak, secreted 
(excuse the comparison) by the free surface of the ganglion 
layers. The fibrillar substance once produced is rapidly in- 
creased, and at the same time advances into the midst of the 
ganglion cells by a kind of invagination. The penetration of 
the fibrillar substance into the layer of ganglion cells appears 
to me to be the result of their multiplication, by which the in- 
creasing mass of cells tends to enclose the fibrillar substance.’ 
That the central white substance originates from cells is 
indubitable, but it appears in the Blow-fly long before any 
nerve cells have been produced from the mantle layer. In the 
newly-hatched larva the fibrillar substance of the brain and 
ventral cord is already developed (Fig. 64), and both are sur- 
rounded by a multiple layer of ganglio-genetic cells. In Pl. II., 
Fig. 7, I have already represented this stage, and there is no 
trace of a mantle layer. Since publishing this figure I have 
obtained sections showing the mantle layer enclosing a large 
cavity in front of the brain, and not yet applied to the hemi- 
sphere, but in immediate relation with the dermal head-discs. 
Viallanes’ own figures seem to me to indicate that there are 
numerous cellular elements in the region from which the brain 
is developed long before the mantle layer has been laid down 
over their surface, and in one the white substance is apparently 
represented amongst these cells [190, Fig. 8, near / o]. 
It appears to me that there is evidence, although not 
conclusive evidence, in favour of a totally new view of the 
manner in which the preoral ganglia are developed—a view 
