348 Dr. H. A. Nicholson on the Distribution in Time 



tion of Sir Roderick Murchison, that they are exclusively 

 confined to this epoch, still holds good as far as all the typical 

 forms of the order are concerned. The somewhat aberrant 

 genus Dictyonema (which, along with some others, might fairly 

 be placed in a distinct suborder) has been found by Hail in 

 the Middle Old Red in America. This discovery would lead 

 us to anticipate a similar revelation in Britain, whenever beds 

 shall have been examined in this formation which present 

 evidence of having offered the requisite conditions for the 

 growth and preservation of these organisms. Further, the 

 same genus Dictyonema occurs in the Tremadoc Slates, which 

 are by some looked upon as the top of the Upper Cambrian series. 

 ISTevertheless it remains certain that the Graptolites as a family 

 are characteristically Silurian ; and further researches are not 

 likely to alter this statement in any essential point. Not only 

 is this the case, but the Graptolites, as regards their abundance 

 as individuals, and the number of generic and specific types, 

 are far more characteristic of the Lower- than of the Upper- 

 Silurian period. And, finally, the inferior portions of the 

 Lower Silurian rocks can claim a decided predominance in the 

 number of genera when compared with the superior members 

 of the same. 



Contrary to what might have been expected, the various 

 genera, and often the species also, of Graptolites are very 

 constant in their range and distribution. They afford, there- 

 fore, very valuable and reliable data, whereby formations in 

 different parts of the world may be correlated with one another 

 or the exact position held by any group of beds in the strati- 

 fied series may be more or less exactly ascertained. 



In Britain Graptolites are known to range from the Tremadoc 

 Slates up to the Upper Ludlow rocks, inclusive, the Lower 

 Llandeilo, Upper Llandeilo, and Caradoc groups being those 

 in which there is the maximum development of genera and 

 species, usually accompanied in the two latter cases by an 

 extraordinary abundance of individuals. On the whole, the 

 lower part of the Lower Llandeilo (Skiddaw Slates), as stated 

 by Salter, must be looked upon as the '"'' metropolis " of the 

 family, since it contains a larger number and a more varied 

 series of generic types than is found in any other formation. 

 The Upper Llandeilo and Caradoc groups, on the other hand, 

 possess together a much greater number of species and of in- 

 dividuals than is the case with the Lower Llandeilos, though 

 this is, perhaps, largely due to the more favourable nature of 

 their sediments, the same disproportion not being recognized 

 in America. 



Looking merely to Britain, the Lower Silurian rocks are 



