SALTERASTER? CORONELLA. 153 



in S. asjperrimus. The arms are swollen. Salter describes them as somewhat 

 carinate. According to my observations the keel is very slight, if present at all. 

 I should describe the arms as being cylindrical. The radial ia are small, irregularly- 

 shaped plates not nearly so large or swollen as in S. asjperrimus. The adradialia 

 are similar and arranged in rows which form a sharp V across the middle line of 

 the arm. It is this arrangement of the radialia which makes the apical surface of 

 the arm at first sight appear carinate. There is a distinct suggestion that each 

 adradial had a stout paxilla-shaft. 



The disc is almost bare of plates, only a few remaining near the edge. These 

 appear to be very similar in character to the adradialia. R : r : : 14 mm. : 3*5 mm. 

 These measurements are made from the left-hand arm which appears to be 

 preserved in almost full length. 



Horizon and Locality. — Upper Ordovician (Caradocian) of Llanfyllin, Mont- 

 gomeryshire. 



3. Salteraster? coronella (Salter). Plate XI, fig. 6; Text-fig. 99. 



1857. Palseaster coronella, Salter, Ann. Mag. Nat. Hist., ser. 2, vol. xx, p. 326. 



1862. „ ,, Wright, Mon. Brit. Foss. Echinoderni., Oolitic, vol. ii, pt. 1 (Palseontogr. 



Soc. for 1861), p. 25. 

 1886. Stenaster ? coronella, Sturtz, Neues Jahrb. fur Min., vol. ii, p. 153. 



1914. „ ,, Schuchert, Fossiliuin Catalogus, Animalia, pt. 3, pp. 28, 39. 



1915. „ „ Schucliert, Bull. 88, U.S. Nat. Mus., pp. 165, 167. 



Material. — A single specimen in the Museum of Practical Geology, Jermyn 

 Street (7402) is the only material of the species. It is a mould showing the 

 greater portion of the apical surface. 



Specific Characters. — Infero-marginalia oblong, closely touching, each with a 

 distinct transverse ridge. Radialia with a large evenly swollen boss with no 

 distinct paxilla-shaft. 



Apical Surface (Plate XI, fig. 6; Text-fig. 99). — An examination of the photo- 

 graph shows that the disc has been broken away in the region of the primary 

 circlet. The plan of the ossicles (Text-fig. 99) makes it clear that the circlet is 

 incomplete. The four larger ossicles are undoubtedly primary radialia, and the 

 smaller ossicle probably represents the primary interraclial usually associated with 

 the madreporite. The fifth primary radial and the madreporite are missing. If 

 this interpretation be correct, the arrangement of the circlet is exactly as in 

 Urasterella thraivensis (Text-fig. 88, p. 138). The radialia are stout and closely 

 touching. Each has a boss-like eminence with no distinct paxilla-shaft. The 

 matrix in which the fossil is embedded is very coarse, and there may have been 

 originally a slender paxilla-shaft which has not been preserved. 



There are two rows of adradialia very similar in appearance. All have paxilla- 



