316 DE. A. VATTGHAtf ON THE EATJNAL SUCCESSION [May I906, 



broad shallow depression in each tabula, and is occupied by one or more 

 short septa. Figs. 9 & 13 of Thomson, above cited, illustrate this stage. 



(4) In the latest stage the outer wall partly splits, and the interspace is 

 broadly vesicular, thus simulating a Caninia. 



This subgenus occurs in the Cyathaxonia-Beds of Rush and 

 Bradbourne; it is also found in the uppermost beds of' the Avonian 

 throughout the South- Western Province, where it is especially 

 common in D 2 and e of Oystermouth (Grower). 1 



Cyathaxonia rushiana, sp. nov. (PI. XXIX, figs. 3, 3 a, & 3 b.) 



Compare Cyathaxonia cornu, Mich., M'Coy (' Brit. Palaeoz. Foss.' 1855, 

 p. 109). 



Form. — An elongated narrow cylinder, starting as a short curved 

 cone. 



Calyx. — Boundary a circle, with its plane perpendicular to the 

 axis ; rim sharp ; from the middle of the calyx-floor projects a 

 strong ' columella.' This ' columella ' is a complex structure, con- 

 sisting essentially of a thick tube surrounding a stout columellarian 

 lath, the interspace being completely filled up. 



Horizontal section. — Central area : oval and completely 

 filled up, but the enclosing ring well-marked. 



Septa : regular, radial, and well-spaced ; 20 to 25 in number ; all 

 the septa are bilamellar, and, in the greater number, the lamellae 

 are separated near the periphery. The septal break is marked out 

 by a unique thin septum, collinear with the columellarian lath. 



This species is common at Bush in the Cyathaxonia-Be&s, to 

 which zone it is confined. The same species is abundant at 

 Bradbourne (Derbyshire) at the same level. 



So closely does M'Coy's description of Cyathaoconia cornu apply 

 to the forms here included, that a new specific name might seem 

 unwarranted ; the reasons for its creation are as follows : — 



(1) The axis of C. cornu is merely described by M'Coy as ' solid,' whereas 



the axis of all specimens of C. rushiana exhibits a "very characteristic 

 tubular structure. 



(2) Cyathaxonia cornu is insufficiently defined by Michelin, and the 

 identity of C. cornu, as a species, rests upon its interpretation by 

 Edwards & Haime. 2 The description given by these authors does 

 not suggest the characters of C. rushiana. 



1 I have carefully examined the specimens from Grower, and have had 

 several sections cut from the same individuals, in order to observe the stages of 

 growth ; the above description of Amplexi-Zaphrentis is founded on the results 

 of this examination. More study is, however, required to determine the 

 relationship of the Upper Avonian forms with certain corals which occur in y 

 and G x of the South -Western Province and also with the ' Amplexi- Caninia' 

 which occurs in the Zaphrentis-Beds of Eush. The full description and illus- 



Nfc Aration of this subgenus is consequently deferred. 



2 ' Monogr. des Polypiers Foss. des Terr. Paleoz.' 1851, p. 320 (Archives 

 Mus. Hist. Nat. vol. v). 



