359 
pierced the corneal facet, a sense bud not unlike some found in fishes 
etc. would be formed (fig. 3, o. £. and fig. 4 E.). 
It is almost certain that the four central retinula cells of Belostoma 
arein various of stages withdrawal to form ganglion cells, and that the axial 
fibres in the retinula of Tabanus are the outer ends of cells that have 
completed this process! Theoretically an indefinite number of retinophorae 
may withdraw in this way from the same point, each leaving its 
attenuated outer end between two newly formed sense cells (compare 
Vespa, Acilius, Belostoma, Tabanus, Pecten); a string 
of double ganglion cells might thus be formed, the whole series 
terminating at one end in a perfect ganglion cell and at the other in 
a newly formed retinophora (fig. 4 D.). This telescoping of indi- 
vidual, epithelial cells, or of a small number of them, seems to me 
to be a true phylogenetic process and invagination a purely ontogenetic 
one. Invagination probably occurs only in compound 
sense organs and there as an incidental result of the 
rapid inwandering of ganglion cells which thus causes 
an enlargement of the inner surface of the sensory 
layer and consequently a warping of the whole organ. 
In the convex eye of Vespa, the temporary invagination produced in 
this way is very deep (see Eyes of Vespa), yet it is not so deep but 
that, when the warping ceases, the eye returns to its natural con- 
dition. But we can readily see how an increase in the strength and 
duration of the warping might make a return to the original condition 
impossible; if so, this convex eye, might, in a single generation, 
be converted into an open pit or a closed vesicle; and it is obvious 
that the advantage thus gained in shape and protection would be a 
purely incidental result of its increasing complexity. Thus it is 
not at all necessary toassume, as is usually done, that 
the adult ancestors of animals with vesicular eyes had 
eyes in progressive stages of invagination. I believe we 
may safely assume that primitive sense organs, ganglia, nerve centres etc. 
have been formed, phylogenetically, by the telescoping of in- 
dividual epithelial cells; this process when repeated ontogenetic- 
ally gives rise to invaginations, for the reasons given above. The 
present condition of the more complex sense organs is a resultant of 
_ the phylogenetic and the ontogenetic process. 
This explanation may be extensively applied to explain invagin- 
ation and evagination, and also to account for the sudden appearance 
of new types of organs. 
Grand Forks, N. Dak., May 15” 1890. 
