BUCANIELI.A TRILOBATA. 29 



Specific Character)^.— The original description of this species, the type of which 

 came from the Upper Ludlow Passage Beds of Felindre, is as follows : " Convoluted, 

 smooth, 3-lobed, central lobe largest; inner whorls small, visible; aperture about 

 twice as Avide as long; length and breadth four lines." McCoy (oj). eit.) described 

 it more fully as follows : " Globose, umbilicus small, deep ; whorls trilobed by two 

 deep spiral furrows, the lateral lobes half the width of the mesial one ; very convex > 

 mesial lobe most prominent, slightly flattened. Width of small specimens three 

 lines, length the same; proportional width of umljilicus -i-y-y as compared with the 

 diameter of the shell." McCoy's specimens are poor internal casts from the 

 Tilestones (Upper Ludlow) of Storm Hill, Llandeilo, which is the only locality he 

 mentions for this species. 



RemarJis. — There is some doubt as to whether the Wenlock shells attributed 

 to this species ought to be separated as a variety or even distinct species ; they 

 agree with those from Gotland figured by Lindstrom {op. cit.) in possessing a very 

 wide flattened back with an unusually broad non-elevated median lobe, such as is 

 found in Bncanla trilobafa (Conrad) from the Niagara Formation in America. One 

 of the Wenlock specimens (a/870) in the Sedgwick Museum, from Dudley, was 

 stated by Salter^ to be the largest specimen known and measures 45 mm. across 

 the mouth and about 32 mm. in height (PI. VI, fig. 1). 



The specimens from the Upper Ludlow of Storm Hill and of the Kendal district 

 (PI. VI, figs. 2, 3) have the median lobe relatively more elevated and more convex, 

 and the whole shell seems more laterally compressed and less globose, thus resembling 

 some Lower Devonian species from South America. 



All the specimens occur as internal casts, and as the correct generic position of 

 the species depends on the presence or absence of a slit-band, which the state of 

 preservation does not prove, there has been much difference of opinion on this 

 point. As mentioned above, the nauie Bu.caniella was proposed by Meek for broad- 

 backed, trilobed Silurian species of Belleroplwn without a slit-band. For certain 

 early Devonian similarly trilobed shells possessing a slit-band, Clarke^ suggested 

 the name Plectouotus, and he has recently^ reviewed the whole question of the 

 validity of these genera. Clarke (oj). cit, 1900, p. 36) was of the opinion that 

 Sowerby's BelUvoplion trilohatiis was a Devonian shell ; but this is erroneous, 

 Sowerby having founded the species on a Silurian specimen. But as no slit-band 

 has been definitely proved to exist in the British Silurian B. trUohatns it appears best 

 to refer this species to Bucaniella, in the strict sense in which Meek used the term. 

 Many of the Devonian forms described by Sandberger and others from Europe, 

 America, and South Africa bear an external resemblance to this Silurian species, 

 and Whidborne* has described a British Upper Devonian shell as Trojndodiscus 



1 Salter, ' Cat. Cambr. Silur. Foss. Woodw. Mus.,' p. 157. 



2 Clarke, ' Palseoz. Faunas of Para' (1900), p. 40. 



3 Id., " Foss. Devon. Parana " (' Mon. Serv. Geol. Miner. Brasil,' vol. i, 1918), pp. 168—170. 

 * Whidborne, ' Mon. Biit. Devon. Fauna ' (Palteout. See), vol. iii, (1896), p. 68. 



