GOMPHOCERATIDiE. 
213 
(loc. cit. p. 198, pi. ccxliv. ff. 6-12) as being rounded, and covered 
with a cap {calotte)^ which is sometimes smooth, as in Gomphoceras 
\_Phragm.'] Broderijpi (pis. Ivii., lix.) and Gomphoceras [Phragm.'] 
perversum (pi. c.), and sometimes ornamented with fine concentric 
striae, as in Gomphoceras [Phragm.^ imbricatum (pi. ccxliv.). In the 
last two species there is a small cicatrix at the apex, where the 
siphnncle appears to terminate. 
The T-shaped aperture may have its larger orifice widely ex- 
panded {Gomphoceras fe7'um, Barr., loc. cit. pi. Ixxx. f. 21), or it may 
be contracted in the middle and expanded at both ends (Gompho- 
ceras tumescens, Barr., pi. Ixxxi.). 
The channel connecting this orifice with the smaller (ventral) one 
varies considerably in length in difi’erent species, and may be re- 
duced almost to zero. The smaller orifice, which is supposed to 
have given vent to the respiratory funnel, is sometimes merely an 
ovate expansion of the channel, or it may be distinctly bilobed, thus 
repeating on a smaller scale the shape of the larger orifice. 
The body-chamber is large in comparison with the septate part 
of the shell ; it is frequently crenulated at the base. The siph uncle 
varies from a submarginal to a nearly central position, and is often 
infiated between the septa. In the endogastric forms it is often 
filled with radiating calcareous deposits, the centre being sometimes 
found to be occupied by a canal (? endosiphon) which is filled with 
the matrix in which the fossil was imbedded h The septa are ap- 
proximate and direct. 
BemarTcs. The great development of the Gomphoceratidoe in the 
Bohemian Basin enabled Barrande to observe the very considerable 
range of variation which the group passes through, and which could 
hardly have been suspected from its comparatively meagre repre- 
sentation in the British rocks. 
Finding it convenient to subdivide the many species coming 
under his notice, Barrande selected as the basis for his subdivisions 
the characters of the aperture, and still retaining the old groups 
Gomphoceras and Phragmoceras, he constructed a table under each 
of these, in which the species are grouped according to the number 
of lobes in the aperture, the names chosen for them indicating 
these numbers, such as, e. ^., Dimorion, Dimeres, Tetramorion, 
Tetrameres, &c. &c. These groups he regarded as of subgeneric 
rank, but Hyatt ^ has erected some of them into genera, slightly 
altering the terminations of Barrande’s names to suit their extended 
^ Barrande, loc. cit. pi. liv. f. 9, pi. Ixxxiii, ff. 12, 140, 
^ Loc. cit. 
