Dr. F. A. Bather— The Mount Torksse Annelid. 535 



silica. In section the fossils are generally somewhat hour-glass or 

 dumb-bell shaped, but this is an ordinary effect of compression, and 

 there is no trace of lateral ridges such as characterise Serpulites. 

 Unequal compression and a slight tendency to cleavage in the rock 

 have produced inequalities of thickness, but the general impression 

 produced is that the tubes were remarkably straight and not curved 

 as in Serpula. The surface appears to have been smooth, but, if 

 anything, irregularly and finely striate longitudinally rather than 

 polished (Fig. Ic). There are no transverse annulations, wrinkles, 

 or stride, but specimen 19 shows a slight transverse swelling at one 

 level (Fig. 2). The evidence at hand does not show whether the tube 

 was closed or open at the smaller end. Probably it was not operculate 

 at the broader end, or traces of the opercula would have been found. 



Owing, perhaps, to its silicification, no microscopic structure can 

 be detected in a thin transverse section of the tube (Fig. 3). 



The specimens of rock, of which an average sample measures 

 100 X 62 X 18 mm., contain as a rule traces of more than one 

 individual, but they are not very closely associated. Since there are 

 in the rock no other solid bodies to which they can have been 

 attached, it must be inferred that the tubes were free and isolated. 

 While the evidence of the vertical specimens suggests that they were 

 sedentary and embedded in the miid, the still greater number of 

 specimens in the plane of bedding leads one to enquire how .and 

 when they left their vertical station. 



The systematic position of this fossil is far from clear. There 

 seems, indeed, no reason to doubt the unanimous ascription of it 

 to the Polychjeta tubicola, at least with as much reason as 

 Serpulites, Gornulites, and similar forms ; but the genus is not 

 so obvious. Captain Hutton (1885) has suggested that certain 

 annelids from the Upper Maitai beds of the Dun Mountain ^ are 

 "perhaps Gornulites.''' The present specimens, however, assuming 

 that they are of the same species, do not support that view, since 

 they are devoid of annulations and taper much less rapidly. 



From Serpulites, as defined by Mr. R. Etheridge, fil. (1880, p. 304), 

 this species differs in the absence of the lateral ridges and in its 

 straightness. The genus Serpula, as defined by Linnaeus (Syst. Nat., 

 1758, p. 786), had "Testa uuivalvis, tubulosa, adhserens, ssepe 

 isthmis integris passim intercepta." It is true that the first species 

 mentioned by him, Serpula seminulum, " recedit a congeneribus quod 

 libera sit nee adhasreat aliis corporibus," but this species cannot 

 be taken as the genotype, and it seems justifiable to follow 

 E. Etheridge, fil. (1880, p. 362) in regarding adherence as 

 a necessary criterion of the genus. Another feature almost 

 universally ascribed to Serpula is the irregular contortion or 

 enrolment of the tube. On both these counts the Mt. Torlesse 

 annelid is not a Serpula. 



^ Accounts of the Maitai beds of the Maitai Eiver and Dun Mountain district 

 near Nelson have been given by M'Kay (1878 and 1879), Avho, however, mentions 

 only " Annelid tracks " and a shell like Inoeeramiis. But these beds were not then 

 correlated with the Annelid slates and sandstones of "Westland and Mount Torlesse, 

 which were placed in an underlying Eimutaka series (see Hector, 1878). 



