80 Mr. T. H. Withers on 
ancestors of the Scalpellide. Stramentum, on the other 
hand, constitutes a highly specialized and aberrant form. 
Instead of developing a definite peduncle, it specialized in 
the disposition and arrangement of its plates to form a 
completely armoured shell. When attached to its object, 
usually an ammouite, the cirri would not have nearly so 
much sweep and freedom of movement as in a peduncu- 
late form. The subsequent splitting of the carina, and the 
non-intersection or overlap of the plates along the outer 
margins of the shell, was no doubt evolved to obtain that 
freedom of movement, and this specialization, while giving a 
temporary advantage, probably led to its extinction under 
changing conditions; and the fact that the two sides of the 
shell were so easily parted would render it especially vulner- 
able to its enemies. 
Prof. Gruvel (1905), who has been followed by later 
authors, has a different conception of the phylogenetic 
position of Stramentum (= Loricula). He considers that the 
first remains of cirripedes are represented by the fossils 
Turrilepas, H. Woodward, and Plumulites, Barrande, and that 
those fossils constitute the complete imbricated covering of 
a primitive cirripede. The animal may be said to have been 
enclosed in a scaly cylinder, which afforded protection to 
the appendages and soft parts. He considered that later the 
upper row of plates were more specially developed to form 
the capitulum and that the remaining rows remained 
undifferentiated and served to form the peduncle. Loricula 
is supposed to represent this second stage, and Prof. Gruvel 
has given some very convincing figures, which have been 
reproduced in the text-books, as to the structural relation- 
ship of those two forms. He does not refer at all to Archeo- 
lepas or any other Jurassic cirripede. 
Apart from the fact that Turrilepas may not be a cirripede, 
it is quite clear that Prof. Gruvel has misunderstood the 
structure of the example of 7. wrightianus figured by 
Dr. H. Woodward, and on which he based his figure. Instead 
of Turrilepas having a laterally flattened shell with five rows. 
of plates on either side as given in Gruvel’s figure (see 
text-fiz. la, 6), the shell is subtriangular in transverse 
section (see text-fig. 1a’) and there are four rows of plates 
only. In fact, the real structure of the two forms is so 
fundamentally different that it is difficult to imagine that 
there can be any relationship between them. 
While it is difficult therefore to see how Turrilepas could 
give rise to a form like Loricula (text-fig. 2), there is little 
doubt that certain of the stalked cirripedes were evolved 
