872 The American Naturalist. [October, 
impressed zone in the eyrtoceran or gyroceran stages of devel- 
opment. In several examples also, the disappearance of this 
characteristic has been observed in the last stages of old 
whorls. There is, therefore, every reason for regarding the 
impressed zone as a ctetic characteristic acquired in the later 
stages of growth and not hereditary so far as is known in any 
shells of the earlier paleozoic periods. 
The same statement may also be made with regard to 
the majority of Carboniferous shells. There is, however, a 
notable exception in Coloceras globatum (sp. De Kon.) Hyatt, 
and very likely some other species of closely coiled nautilian 
shells. In C globatum of Visé, Belgium, I found in seven 
specimens that the impressed zone appeared while the whorl 
was still in the cyrtoceran stage. Pl. XVIII, figs. 9-10, give out- 
lines of the adult of this species, and figs. 11-12, of the young 
and the zone, showing that the impression appeared long before 
the whorls touched each other and began to assume nautilian 
characters. Section, fig. 13b, shows the impressed zone occurr- 
ing in the cyrtoceran stage while the venter or outer side of 
the whorl was rounded. Such facts admit of but one explana- 
tion, namely, that in this species the impressed zone had 
become hereditary and was in consequence repeated at an 
early age, previous to the occurrence of close coiling which 
produced it in the ancestral forms of the same group. 
There are certain correlative characters which lead me to 
think that this is only a partial statement and perhaps a more 
complete and better one would be as follows: that the 
impressed zone, together with a peculiar broadening out of 
the dorsum and helmet-shaped section of the whorl, and per- 
haps also certain forms of sutures occurred in the early stages 
of some Carboniferous species before the nautilian stage, and 
consequently they must have been introduced by heredity into 
the development of this species before the tendency to close 
coiling had completed the first whorl. Thus these characters, 
although purely ctetic in origin, were repeated before the 
usual conditions recurred in the ontogeny of this species 
which had obviously and repeatedly produced them in the 
nautilian forms of the earlier paleozoie and the more general- 
