308 
INi^^ERIOR OOLITE AMMONITES. 
DORSETENSIA SUBTECTA, LIOSTRACA, TECTA. 
These three " species " may be known for general purposes as the " Liostraca' 
group." They possess in common one feature which separates them from the 
other members of the genus, namely, a large and distinctly-hollow carina. This 
is, in reality, the only feature which separates " stibtecta" PI. LIV, fig. 3, from 
" C07n2jla7iata and the specimens are arranged in PI. LIV in order to show a 
gradual decrease of ribbing correlating with an umbilicus proportionately slightly 
smaller in each side-view. Figs. 1 and 3 are placed together especially on account 
of their similar appearance ; but in fig. 1 no trace can be found of the hollow- 
keel- structure so prominently shown in fig. 3.^ 
That the " Liostraca-grouTp " was derived from a species like Dorsetensia 
complanata, which had not a hollow carina, is shown not only by the external 
resemblance of Dors, complanata and subtecta, but also because the young of Dors, 
liostraca do not exhibit a hollow carina until somewhat advanced in growth (PI. 
LIII, figs. 11, 15). That the " L'io6'^racft-group " is not directly descended from 
" complanata " in a strict sense of the term is shown by the decadence of the 
carina in " complanata,'^ and by the enlargement of its umbilicus in the last half- 
whorl. An imaginary enlargement of the inner whorls shown by Dors, complanata 
in PI. LIII, fig. 10 would indicate the common ancestor of " complanata " and the 
Liostraca-grow^ and a form very much like the complanata figured in 
PI. LIY, but perhaps somewhat more involute, would probably represent the link 
connecting this ancestor with the " Liostraca-gYow^." 
The species of the " Liostraca-grow^ " are separable from one another only by 
the decreasing size of the umbilicus and the increasing thickness of the whorl. 
It is not possible to draw any sharp lines of demarcation between them, because 
I possess intermediate forms connecting those selected for figuring. There is, in 
fact, a completely transitional series from PI. LIV, fig. 3, to PI. LVI, fig. 5, 
almost in the order in which the specimens are arranged in the plates ; while the 
close relationship of the specimen which is depicted in PI. LIV, fig. 3, with 
" Dors, complanata " is obvious ; and the probable connection of the latter with 
Dors. Edouardiana has already been noted (page 307). 
The " Liostraca -group " furnishes another good instance of retrogression 
producing smooth, involute, compressed forms, converging in general appearance 
1 The lettering used in PI. LIII, et seq., with reference to the carina is the following : — a, solid 
carina ; h, complete hollow carina ; c, infilling of hollow carina (test lost) ; e, end of the partition- 
band core of carina on body-chamber where there is no partition-baud ; d, apparent commencement 
of hollow carina. 
