634 ON ARENIG AND LLANDEILO GKAPTOLITES EBOM ST. DAVID'S. 



The so-called "dendroid " Graptolites are here for the first time 

 united by Mr. Hopkinson into a suborder under the title of Clado- 

 phora. That the classification of the genera included in this suborder 

 is to a certain extent artificial is admitted ; but the facts at our 

 disposal, while they afford conclusive evidence of the deficiencies of 

 the scheme here proposed, are as yet insufficient to enable us to 

 make any more satisfactory arrangement. It may be objected that 

 the name given to the suborder is not sufficiently distinctive, many 

 of the Ehabdophora being branching forms ; but the tree-like mode 

 of growth of these genera, all of which bear branches, in which 

 respect they differ collectively from the Ehabdophora, seems suffi- 

 cient to warrant their being connected together under the term Cla- 

 dophora. The two sections into which they are grouped (shrub-like 

 and tree-like) also point out this as an appropriate term. They all 

 seem to have been fixed forms, differing in this respect, as in their 

 mode of growth, from the Ehabdophora, to which they are possibly 

 allied through the genus Thamnograptus, which, with the somewhat 

 anomalous genus Buthograptus, forms the section Thamnoidea. 



The section Dendroidea here includes the genera originally 

 grouped together under this head by its founder, Professor Nicholson, 

 in the first part of his ' Monograph of the British Graptolitidse.' 

 These genera fall naturally into two families, the first of which, 

 Ptilograptidce, connects the section with the Thamnoidea, and appa- 

 rently forms a connecting link between the Cladophora and their 

 recent allies the Thecaphora, Ptilograptus seeming to be a true Ser- 

 tularian zoophyte. The three genera of the Callograptidse are almost 

 inseparably connected through some of their species, the inter- 

 mediate forms being comprised in the genus Callograptus. 



DlSTKIBITTION AND COKEEIATION. 



The present state of our knowledge of the distribution of the Grap- 

 tolites in the vicinity of St. David's is shown in the accompanying 

 Table, in which every species noticed is referred to its exact position 

 in the vertical series. 



Graptolites are first known to occur in the lowest beds of the 

 Arenig rocks exposed in Whitesand Bay. It is an interesting fact 

 that the Cladophora only have as yet been found here, and that 

 the two genera of the Tremadoc rocks, Dendrograptus and Dictyo- 

 graptus, are here represented, associated for the first time with the 

 zoologically intermediate form Callograptus, species of which, too 

 badly preserved for determination, occur a few feet lower than the 

 small group of Dendroidea here referred to. Of Dictyograptus a 

 single species only has here been found ; and although several 

 species of Callograptus and Dendrograptus are known to occur, there 

 are only two species of each genus of which specimens sufficiently 

 perfect for description have been collected. All these species are 

 new, and therefore useless for purposes of comparison. 



Very near to this horizon, but most probably rather higher, are 

 the beds exposed on the north-eastern coast of Ramsey Island, at 



