1868.] NicnoLsoN — coniston-flag graptolites. 535 



with a slender, curved, spiral portion, provided proximally with a 

 small radicle. The cellules here are not so long as in the adult por- 

 tion of the stipe, and they arc more closely arranged, attaining the 

 number of from 30 to 40 in the space of an inch. This coiled-up 

 proximal portion of G. Sedgwichii was described by Hisinger under 

 the name of G. convolutus, and it also corresponds to G. proteus of 

 Barrande and G. millipeda of M'Coy. All these pseudo-species, 

 however, are partially referable to the base, not of the ordinary 

 form of G. Sedgiuickii, but of the variety triangulatus, Harkn., the 

 proximal extremities of the two forms often very closely resembling 

 one another. 



Judging from the figures of G. Clintonensis of Hall, descrihed 

 from the Clinton formation (of Middle- Silurian age), this Graptolite 

 is merely a form of G. SedgiuicJcii ; and its occurrence so high in the 

 Silurian series is worthy of note. 



The normal form of G, SedgwicJcii, Portl., as above described, 

 occurs commonly, in the Upper Llandeilo rocks of Dumfriesshire, 

 and is found, though rarely, in the Coniston Flags of Westmoreland 

 (Skelgill Beck near Ambleside, and Broughton Moor in Furness). 



Var. a. spinigekus, Nich. PL XIX. fig. 32. 



This variety is dubiously distinct from the ordinary form, and 

 differs from it only in the fact that the cellules are provided with 

 strong horizontal spines proceeding from their lower surfaces, and 

 apparently attached to the inferior margins of the ceU-mouths. The 

 spines are ordinarily from -^ to -^ of an inch in length ; but a line 

 or more is not unfrequently attained. It is possible that these 

 spines are normally present in G. SedgivicJcii; but, as the great 

 majority of examples (even when best preserved) do not exhibit 

 any traces of them, I have thought it best to form a distinct variety 

 for those specimens in which they occur. This view is strongly 

 corroborated by the distribution of the species in the Coniston Flags. 

 Thus in the flags of Mosedale in Long Sleddale it occurs in great 

 plenty and of great size, whilst the spineless form is seldom, or 

 never, found. In the flags of Skelgill Beck, however, which are 

 exactly on the same horizon, the spine-bearing variety is, as far as 

 I have seen, entirely wanting, and the species is represented by the 

 normal form, and by the variety triangulatus of Harkuess. In the 

 flags of Broughton Moor, again, the ordinary form is by no means 

 uncommon, but neither of the varieties has hitherto presented itself. 



Var. /3. TRIANGULATUS, Harkn. PL XIX. figs. 33, 34; PL XX. 

 figs. 1, 2. 



Rastrites triangulatus, Harkness, Quart. Journ. Geol. Soc. vol. 

 vii. pi. 1. figs. ^a~d. 



This is a very well-marked variety of G. SedgwicJcii ; but it 

 seems on the whole to be only a variety, and not to be a distinct 

 species. In some localities it prevails to the total exclusion of the 

 ordinary form, and it is almost always the most abundant phase of 

 the species. The stipe is simple, monoprionidian, and more or less 



