170 DEVONIAN FAUNA. 



Description. — Ventral valve usually rather small, elongate, gibbose, but 

 flattened on the marginal half of the centre of the valve. Umbo more or less 

 elevated, recurved upon the hinge. Hinge-line shorter than the width of the 

 shell. Surface having (1) from fifteen to twenty-five rather irregular, rounded 

 rays, rising some distance in front of the umbo, and bearing some large, long 

 spines, which tend to arrange themselves in transverse rows ; (2) exceedingly 

 minute fibrous markings ; and (3) a few irregular transverse undulations on 

 the wings and umbonal parts. 



Size. — An unusually large specimen measures 33 mm. long, 20 mm. wide, and 

 10 mm. deep ; its umbo extends about 8 mm. behind the hinge-line. 



Localities. — In the Woodwardian Museum are eight specimens from Croyde 

 and three from Top Orchard ; in the Museum of Practical Geology one from 

 Braunton ; in the Porter Collection five from Pilton and one from Poleshill; and 

 in the Barnstaple Athenaeum a slab containing specimens mentioned and figured 

 by Davidson as " P. long i spinas, Sowerby ? " 



Remarks. — The specimens in the Barnstaple Athenaeum, which Davidson 

 referred somewhat doubtfully to P. longispinus, are very imperfect and indistinct; 

 but they are evidently identical with the other specimens mentioned above, 

 which are most of them in a better state of preservation, and afford clear proof 

 that they do not belong to the same species as Sowerby's Carboniferous 

 P. longispinus. They have much stronger and fewer ribs, and none of them show 

 any signs whatever of an angular median sinus. 



From the typical form of P. prsdongus they usually differ by (1) their smaller 

 size ; (2) the absence of a sinus bearing a vertical row of very long spines ; (3) 

 the presence of spines on other parts of the body, which are often arranged in 

 concentric rows ; and (4) their very much more definite and regular ribs. 



On the other hand, these points of difference are not constant. One or two 

 of the specimens show a distinct tendency to a slight median groove or concavity, 

 and have median spines arranged longitudinally. It appears to me that it will 

 therefore be best to regard them as a variety of P. praslongus, as it may prove to 

 be only an immature or stunted form of that shell. 



3. Productus scabriculus, Martin, sp. Plate XX, figs. 16 — 18, and Plate XXI, 



fiff. 12. 



'D' 



1809. Anomites scabeictjlus, Martin. Petrif. Derb., p. 8, pi. xxxvi, 



fig. 5. 

 1841. Lept.ena scabeicula, Phillips. Pal. Foss., p. 58, pi. xxiv, figs. 97 a, b. 



