86 BRITISH FOSSIL CORALS. 



high. Each corallite usually attains the length of about one inch and a half before 

 dividing, and the diameter of the sinijjle calices is in general about eight lines, but the 

 compound calices are often double that size. 



We have examined a great number of specimens of Thccosmilia anmdarls found at 

 Steeple Ashton, Wiltshire. Most of these fossils belonged to the collections of the Museum 

 of Practical Geology, the Geological Society, the Bristol Museum, the Paris Museum, 

 Mr. Bowerbank, Mr. Phillips, Mr. Stokes, and Mr. Pratt. 



The same species is met with at Slingsby in Yorkshire, and we also refer to it a fossil 

 found at Malton, and belonging to Mr. Bowerbank's collection, which resembles nuich the 

 specimen figured by Mr. Phillips, but is not in a state of preservation sufficiently good to 

 enable us to be certain as to its specific characters. W. Smith, who gave some good 

 figures of this coral in his remarkable work on ' Organic Remains,' mentions it as having been 

 found in the following localities : Longleat Park, Stratton, Ensham Bridge, Wotton Basset, 

 Banner's Ash, Well near Swindon, Wilts and Berks Canal, Shippon, Bagley Wood Pit, and 

 Stanton, near Highworth. Mr. Phillips also points out its existence at Seamer. We must 

 add that some specimens found at RadcliiT by the collectors of the Geological survey, and 

 communicated to us by Sir H. De la Beche, do not appear to differ from the above-described 

 species, but some other fossils from the same locality appear to be more similar to the 

 MonUivaliia Lesueiiri from the Kimmeridge clay near Havre. 



The well-preserved specimens which we have met with in some of the English collec- 

 tions enable us to rectify, concerning Thecosmilia, an error of the same kind as that we 

 formerly fell into with respect to MonlJivaUia. The septa are not terminated by an undi- 

 vided edge, as in the tribe of JEusmilina, where we placed this genus when the first part 

 of this Monograph was published ; they are denticulated in a regular manner, and the 

 Thecosmilia may be briefly defined " compound Montlwaltia!' 



This generical division contains fossils belonging to cretaceous as well as Jurassic 

 formations, and we are also inclined to admit in it some fragments of corals found in the 

 celebrated fossiliferous deposit at St. Cassian. We have given the list of all these species 

 in the synopsis joined to our Monograph of the Palaeozoic corals, but we regret not having 

 had an opportunity of examining some of them, and others that have of late been submitted 

 to our investigation were in a very bad state of preservation ; much uncertainty, therefore, 

 still exists respecting the specific character of many Thecosmilia. However, the fossil 

 above described is easily recognisable by the unequal size of the calices, which take their 

 origin on the same stem and are of the same age, and by the form of the septa. Thus, 

 T. Konincld} a species of which we have only seen a young specimen, differs from T. 

 annularis by its septa being thinner, and more equal in size. In T. trichotoma' and 

 T. lohatcc" the calices are almost circular, smaller, and placed on the same level. T. semi- 



1 Milne Edwards and J. Ilaime, Ann. Sc. Nat., 3'"" ser. vol. x, p. iTL'. 



2 Lithodendron tricJiotomunt, Goldfuss, Petref. Germ., pi. .xiii, fig. (i. 

 •^ Lobopliyllia lobata, Micliclin, Icon., pi. ixvii, fig. 3. 



