Btylasteridce from the Victorian Tertiaries. 179 



relied on in the main as specitio are not so in reality. He is 

 inclined to believe that the colour variation is merely an indic- 

 ation of sex and age, and says " In all cases it is ditHcult to 

 determine specific differences in Hydrocoral lines ; but the diffi- 

 culty is considerably increased w^hen there are only small pieces 

 of the coralla at the disposal of the naturalist, for the general 

 form, colour and mode of branching of the entire colony have 

 always been taken as some of the most impoi'tant specific cha- 

 racters of these corals, and these features cannot be determined by 

 the examination of small pieces. Whether these characters are 

 satisfactory is another matter." These difficulties press the more 

 heavily upon the palaeontologist, as he has usually to deal with 

 very small fragments. 



Numerous fragments of the genus have oeourred to me from 

 both our Miocene and Eocene beds, but I leave their specific 

 determination undecided. A series of measurements of the 

 diameters of the pores showed that the range of variation was so 

 great that satisfactory results could not be attained in this way. 

 Comparison was made with a recent specimen of D. coccifiea. 

 Gray. The pores as a rule were larger in the fossil than in the 

 recent form. In many of the specimens, which were somewhat 

 rolled and corroded, the thinner tissue near the mouth of the 

 pores had broken down so that the gastropores appeared to open 

 into a longitudinal sulcus while the dactylopores occupied grooves 

 leading into this and also trenching the outer wall of the corallum. 

 For some time I thought that this might serve as a specific 

 difference, but as most of the specimens showing it seemed to b? 

 worn, and were moreover small and presumably young branches, 

 whereas the characteis did not appear in thicker specimens, 

 stress cannot be laid upon it. Still I think it very probable that 

 better specimens will show that the peculiarity of the dactylo- 

 pores opening into grooves does normally occur in our Eocente 

 species. Till specific differences are found in the corallum of the 

 recent species which are decipherable even in fragments, we 

 must, I think, rest content with a generic i-ecord from our 

 tertiary beds. 



Localities. — Miocene: Grange Burn, near Hamilton; Eocene: 

 Muddy Creek ; Batesford. 



