METOPASTER MANTELLI. 41 



the length of the side about 25*5 mm. In a very finely preserved cast from the 

 Upper Cretaceous beds of Haldon the length of the side is 29'5 mm. 



Locality and Stratigraphical Position. — All the examples of this form with 

 which I am acquainted are from the Upper Chalk. One of Forbes's type is from 

 Gravesend, but the locality of the other is not recorded. The large example 

 figured on PI. XIII, fig. 2 a, is from the Upper Chalk near Bromley. 



An extremely well-preserved cast in flint from the Upper Cretaceous beds 

 of Haldon (Devonshire) is in the collection of the Albert Museum, Exeter ; and a 

 cast of this example may be seen in the Museum of the Geological Survey, Jermyn 

 Street. 



History. — This form appears to have been first recognised by Parkinson, who 

 erroneously referred it to the Pcntagonaster semilunatus of Linck. The latter is a 

 well-known recent species, and quite distinct from the fossil under consideration. 

 Mantell, following Parkinson's determination, referred to the form under the 

 name of Goniaster semilunata. Forbes was the first to indicate that these views 

 of his predecessors were incorrect, and diagnosed the species in his memoir ' On 

 the Asteriadae found fossil in British Strata ' under the name of Goniaster {Gonio- 

 discus) Mantelli ; and figures of two examples were subsequently given in Dixon's 

 ' Geology and Fossils of the Tertiary and Cretaceous Formations of Sussex,' 

 London, 1850. Both these specimens are now preserved in the British Museum. 

 Careful drawings of each fossil are given on PL XIII, figs. 3 a and 4 a. 



Hemarks. — It is not without hesitation that I maintain this species of Forbes's 

 as independent from Metopaster Parkinsoni. For the present, however, I 

 consider it to be distinguished by the smaller size, the comparative sijuareness 

 of the supero-margiual plates, the small size of the ultimate paired plates, as 

 well as by the character of the ornamentation of the supero-marginal plates and 

 of the abactinal plates. Whether a more extensive series of specimens will break 

 down or uphold these distinctions I do not feel prepared to say. It is undoubted 

 that the two forms are very nearly allied. 



I feel considerable doubt as to whether one of Forbes's types — that shown in 

 PI. XIII, fig. 4rt — really belongs to the same species as the examples illustrated in 

 figs. 2 a and 3 a on the same plate. 



