413 
Silurian of their immediate derivation from Nautiloid ancestors. 
This is precisely what actually occurred and in the Nautiloidea such 
evidence is easily obtained as has already been stated above in the 
pages of the Introduction and other publications. 
It-also follows, if the theories advanced by the author are true, 
that the Nautilinide among Goniatitinz, as ancestors of the Am- 
monoidea, and especially the genus Mimoceras as the centre of 
derivation, should also show more prolonged retention of nautiloid 
characters in their ontogeny than is usual in their supposed descend- 
ants. The researches of Sandberger, Barrande, Branco and the 
author show this to be a fact. The figures of Pl. ii copied from 
Barrande and Branco exhibit this in Af@moceras compressum, am- 
bigena and the whole of the Nautilinidze of the Silurian, and the 
essential distinctive characteristic of this family is the nautiloid 
form of the septa and lateral sutures. The shells of this genus also 
do not possess a contact furrow, as noted above, and have no an- 
nular lobes on the dorsum. 
The first suture of AZimoceras compressum, Figs. 3, 4, Pl. 11, and 
in some other allied species of the Devonian is bent into a slight 
lobe on the venter, which is a purely nautiloid character, and not 
to be confounded with the ammonoidal lobe in the same situation 
in the third suture that follows this. This is shown by the occur- 
rence of similar lobes in the Endoceratidz and some cyrtoceran 
forms of Nautiloidea and in figures of sutures of /Vautilus deslong- 
champseanus and clementinus of the Cretaceous, also copied from 
Branco, which have similar first and second sutures. The aselate 
rst septum is in AZ. compressum, followed on the second septum by 
a broad, almost imperceptible saddle, also considered aselate by 
Branco, but which is obviously a transition to the latisellate, or 
broad-saddle type of suture in the more specialized forms. The 
limits of the ananepionic substage in this form, which, as said 
above, is directly transitional to Bactrites, is therefore that part of 
the whorl which is represented by these two septa and the living 
chamber in which the animal rested while constructing the second 
one. 
The characters of these two septa, however, are not repeated in 
the closer-coiled forms of the Nautilinidee and Primordialidz. In 
these the repetition of the outline of the second suture may be 
entirely omitted, the shell passing immediately in the second sep- 
tum to the repetition of the peculiar undivided ventral of the 
