POLYCH^ETA. 



481 



of the Silurian is just as likely to be Annelidan as Nemertean, and 

 the nature of the Legnodesmns of the Solenhofen Slates is wholly 

 problematical. In fact, the entire subject of the remains of fossil 

 Errant Annelides is one of the most obscure and difficult with 

 which the palaeontologist is called upon to deal ; and all that can 

 be done here is to glance at some of the leading points of interest 

 connected with it, under the following heads : — 



I. Burrows of Habitation. — Various recent Annelides live buried 

 in the sand or mud, between tide-marks or in shallow water, and 

 communicate with the surface by means of a perpendicular shaft 

 or burrow. Such shafts may, for convenience' sake, be termed 

 " burrows of habitation," though the animal forms a fresh one at 

 will, as it moves from one spot to another ; and, as a matter of 

 course, they run in a direction more or less opposed to the surfaces 

 of the laminae of the rock, being often quite vertical. Sometimes 



Fig. 346. — Annelide-burrows {Scolithus Canadensis) from the Potsdam Sandstone (Upper 

 Cambrian). (After Billings.) 



such burrows are hollow, but they are more commonly filled up by 

 the matrix of the rock. Among the genera which have been founded 

 upon remains of this kind are Scolithus, Histioderma, and Areni- 

 colites, all of which occur in rocks of Cambrian or Ordovician age. 

 Scolithus is founded upon long burrows, which are nearly straight, 

 and descend vertically through the rock (fig. 346). They often 

 become somewhat widened out superiorly, and are generally found 

 in great numbers together. They occur abundantly in the Pots- 

 dam Sandstone (Upper Cambrian), and Clinton formation (Silurian) 

 of North America, and also in the hard sandstones of the Stiper 

 Stones in Shropshire (Ordovician). They have been supposed to 

 have been formed by sea-weeds, but there is little doubt that they 

 vol. 1. 2 H 



