BRACHIOPODA. 
289 
loid shell from the upper Wenlock shales which Davidson has described* as 
Waldheimia Mawi; a species which strikingly resembles Cyclospira bisulcata, both 
in form and size. This shell has a low median septum in the brachial valve, 
and its brachidium is longer and much broader than in those of the Devonian. 
In all of Davidson’s representations of the interior of these species, the hinge- 
plate, which we may assume to be somewhat constructive, is given with a 
distinct cardinal process in the Devonian species, like that of the living 
Magellania, though in W. Mawi there appears to be a trace of a perforation in 
the plate. These structures, however, are not fully described. The actual 
differenc.e in the composition of this plate in the recent Magellanias and the 
Devonian Cryptonellas, as above described, may be regarded as a highly impor¬ 
tant basis of distinction between these forms. Were it necessary, however, to 
rely upon this difference alone, we should fall far short of separating their 
remote predecessors of the palaeozoic era as widely from Magellania as the 
evidence seems to require. 
The form of the long, recurved adult loop in such living genera as Magella¬ 
nia, Macandkewia and Terebratella, has been shown by various investigators 
to be but the terminal condition of a series of metamorphoses. Evidence con¬ 
cerning the immature condition of the loop in any of the fossil terebratuloids 
is extremely difficult to obtain. In the very early growth-stages of Cryptonella 
planirostra^ where the shell has a length of not more than 4 or 6 mm., the bra¬ 
chidium is simply a miniature of its adult condition. However, from what we 
now know of the changes in living and extinct Brachiopoda of similar char¬ 
acter, it seems a natural and necessary inference that the brachidia of all such 
terebratuloids have undergone modifications or metamorphoses which, though 
slight in comparison with similar changes in the living species, yet do involve 
a progressive change from the simple loop of Renssel^ria and Centronella to 
the resultant acquired in Magellania and Terebratella of modern seas. 
It should not be overlooked, however, that in the recent genera of terebratu¬ 
loids these modifications of the loop are complicated by the presence of a median 
septum, which is an integral part of the brachidium, and the absence of such a 
* Silurian Supplement, 1882, pp. 76, 77, pi. iv, figs. 1-3. 
