T. S. Rail — Graptolite Rocks, Victoria, Australia. 443 



perhaps indicative of the higher beds of the series. Didymograptus 

 cadaceus, again, is rare and small, and it is interesting to notice, 

 as we pass up through a long series of rocks above those of Bendigo, 

 that it increases in relative numbers, and at the same time gradually 

 attains a much larger size, till it reaches its maximum near the 

 horizon of the uppermost Castlemaine beds, where it crowds the 

 rocks to the almost entire exclusion of other forms. It then enters 

 on the period of its decline, is but sparingly represented by 

 stunted forms at Darriwill, and perhaps ranges up into the Upper 

 Ordovician. 



3. Castlemaine Series. — The rocks which form this series overlie 

 the Bendigo beds conformably. The change in fauna is slight, but 

 Tetragrapfiis fruticosus, so characteristic of the older beds, is absent, 

 and Didyniograptus bifidus becomes abundant at the base of the 

 Castlemaine rocks. I have subdivided the rocks into several minor 

 zones,^ but need not here consider the details. The summit of the 

 series has not been clearly marked off from the beds which succeed 

 it, as they are not known to come into contact. 



la the lowermost beds Didymograptus bifidns, J. Hall, as just 

 stated, is abundant, but soon disappears. Fhyllograptus typus long 

 persists, and is survived for a sbort time by P. angnstifolius, 

 J. Hall. Species of Diplograptus appear in the higher beds, but 

 not, apparently, in the lower. A species of Clonograptns occurs 

 in the lower beds, but soon disappears. Loganograptus Logani puts 

 in an appearance in the highest zones and ranges up into the 

 Darriwill Series. 



Though the position of Didymograptus bifidus and its allies would 

 appear to be somewhat lower, relatively, than what it is in the 

 Northern Hemisphere, yet if it be true, as has been suggested 

 by Nicholson and Marr, that the 'tuning-fork' group is phylo- 

 genetically connected with Tetragraptus fruticosus, then its position 

 in beds so closely associated with those in which the latter species 

 occurs is no more than we should reasonably expect to find. The 

 really striking point is that it should die out so soon with us, and 

 characterize rocks that are closely connected by their fauna with 

 the Lower Arenig and not with the Upper. Although we seem to 

 be able to trace the descent in a general way of the ' tuning-fork' 

 group from the Tetragraptus fruticosus group, still we are unable, 

 I venture to think, to descend from generals to particulars. 



The late appearance of Loganograptus is of interest. As it is 

 one of the species figured by McCoy in his " Prodromus of the 

 Palaeontology of Victoria," and was stated to occur commonly at 

 Castlemaine, I was naturally on the look out for it, and for a long 

 time in vain. I found it at last about a mile to the south of the 

 town, where I made a prolonged search for it, on the strike of 

 McCoy's locality, which is in the centre of the town and inaccessible. 

 It is never associated with Phyllograptus typus, which does not 

 range as high as the beds in which Loganograptus first appears. 

 Dichograptus octobrachiatus has also disappeared, after having long 

 1 Proc. Eoy. Soc. Victoria, new ser., vol. vii (1894), pp. 55-88. 



