PREFACE. ix 



figures should be so far accurate, that the forms described, 

 whether rightly identified or not, should be capable of ready 

 recognition by subsequent workers. 



In my work on ' The Structure and Affinities of the " Tabu- 

 late Corals" of the Palaeozoic Period' (1879), I treated gener- 

 ally of the characters and systematic relationships of Montictdi- 

 pora, and its immediate allies. Part of the matter there given, 

 greatly expanded, and improved by further and more extended 

 investigations, is here reproduced. I have found it necessary, 

 however, to restrict myself upon the present occasion entirely 

 to the genus Monticulipora proper, using this term in its wide 

 sense. Such closely allied types, therefore, as Fishtlipora, 

 M'Coy, Constellaria, Dana, and Dekayia, Edw. and Haime, 

 will only be treated of incidentally. 



I regret greatly that at the time when I wrote the above- 

 mentioned work I was wholly unaware of the fact that Dr 

 Dybowski had previously (1877) published a treatise on the 

 Monticuliporoids (Die Chaetetiden der Ostbaltischen Silur- 

 formation). I have now, however, had the opportunity of 

 examining this important memoir, and shall, as occasion arises, 

 make such observations upon it as may appear necessary. 



The material upon which the present work is based is con- 

 tained in my own cabinet, and has been mostly collected by 

 myself from the Silurian, Devonian, and Carboniferous deposits 

 of North America and Britain. I have, however, to thank the 

 kindness of my friends Dr Gustav Lindstrom, Mr Richard 

 Howse, Mr U. P. James, and Mr John M. Nickles, for the 

 gift of various specimens of forms of interest and importance. 

 I have also, where possible, consulted the collections contained 

 in our public museums, though I have not been able to avail 

 myself of this source of information as fully as I should have 

 wished to do. I may add that I have personally prepared the 

 whole of the numerous thin sections required for this work. 



