﻿1850.] HARKNESS ON THE GRAPTOLITES OF DUMFRIESSHIRE. 59 



of the cells, according to the position which they occupy, an occur- 

 rence similar thereto being met with in the genus Virgularia, in 

 which the cells near the base undergo considerable modification. 



As regards Graptolites proper, and Rastrites, their existing ana- 

 logues are not so well ascertained. On the whole, they appear not 

 far removed from the modern genus Sertularia, and partake rather 

 of the characters of hydroid than asteroid zoophytes *. 



Rastrites, Barrande. 



1. R. peregrin us, Ban*. 



Pl. I. fig. 1 (and 2?). 



(Barrande, Grapt. Bohm.pl. 3. f. 10-13.) Syn. G. spiralis, Geinitz, 

 Neues Jahrb. 1842, pl. 10. f. 28. 



The extreme thinness of the portion on which the cells are placed 

 is very characteristic of this species. It commonly appears devoid 

 of any trace of a central canal, but occasionally an individual may 

 be found in which a depression is seen, such as would result from the 

 collapsing of the sides after the internal portion had been removed. 

 It has generally a curved form, but (unless fig. 2 belong to the same 

 species) only small portions have been obtained, and therefore the 

 nature of the young part is uncertain. The cells, which are some- 

 times on the convex and sometimes on the concave side of the curve, 

 are placed at right angles to the axis, and are far removed from each 

 other. They are much thicker than the axis which supports them. 

 In some rare cases they appear to increase in breadth from their base 

 upwards ; a character also represented in Geinitz' s figure above 

 quoted. A well-marked line runs along the length of each cell, the 

 result of compression, the cells being hollow. In a line there occur 

 about four cells, measuring at their base. The extremities of the cel- 

 lules when these are situated on the convex side of the axis are remote, 

 but when they occur on the concave side they are approximated to 

 each other. This species is generally converted into iron pyrites, and 

 shows no traces of its external envelope. The locality where it is 

 found most commonly is in the course of the Little Queensberry Burn ; 

 but it likewise occurs in other parts of Dumfriesshire where the 

 graptolite shale is met with. 



Fig. 2 is in its inner or younger portion very like this species, see 

 the magnified figure at 2b; in the older part (2 c) the canal is 

 broader ; this specimen therefore seems intermediate between R.pere- 

 grinus and our next species. 



2. R. triangulatus, Harkness. 

 Pl. I. fig. 3. 



This species when perfect is spirally rolled, and the younger por- 



* I have to express my obligations to M. Barrande and Mr. Salter, who have 

 examined the Dumfriesshire Graptolites ; and to whom I am indebted for many 

 important observations concerning them. 



