Correspondence — Dr. F. A. Bather. 527 



find them parallel to it. Sometimes they are as thick as a pencil. 

 Sometimes they are trumpet-shaped. At one time we thought they 

 were restricted to one geological horizon, but they evidently persist 

 from our Cambrian or Canibro-Ordovician conglomerate, in which 

 they occur sparsely, through the sandstone of doubtful Cambrian 

 or Ordovician age to Silurian sandstone, in which they are also rare." 

 The specimens are now in the Geological Department, registered 

 A 1658. 



In a reply to Mr. Twelvetrees, dated September 30, 1911, I com- 

 pared the specimens with similar structures in the Camhrian 

 sandstones of this country and of Sweden, in particular with 

 a specimen obtained by me at Bergquara in Smaland (Brit. 

 Mus., Geol. Dept., A 1356), where such appearances are fairly 

 common and have been referred to Scolithus linearis, Hall. Descrip- 

 tions of these are given by N. 0. Hoist (1893, Sveriges Geol. 

 Under soltning, ser. C, No. 130, p. 6) and Nathorst (1892, Sveriges 

 Geologi, p. 117 and text-figures on p. 118). Though usually 

 interpreted as the filled burrows of worms, Nathorst questions 

 Avhether the structure has not rather arisen in a mechanical way. 

 The cylinders are often packed so closely that I too was " inclined to 

 regard the structure as due to some mechanical agency". Some 

 American specimens in the British Museum, labelled Scolithus, 

 do indicate the existence of burrows subsequently filled; but in the 

 pipe-rock there is no evidence of this. I do not, however, under- 

 stand how the tubes can ever be parallel to the bedding, as stated by 

 Mr. Twelvetrees, if formed by ascending bubbles. 



In another note (Proc. Boy. Soc. Victoria, sr.s., vol. x, pp. 209-10) 

 which Mr. Dunn published in May, 1898, he was tempted to ascribe 

 numerous perforations in a decomposed steatitic rock near Coolgardie 

 to worms, larvae, or flies. He mentions, however, that the roots of 

 eucalyptus trees follow these perforations to as great a depth as 

 150 feet from the surface. May it not be the case that the 

 perforations were actually made by the eucalyptus roots? Such an 

 action is by no means uncommon. 



The weathering action of salt-solution through repeated wettings 

 and dryings is one that I have attempted to apply in removing the 

 matrix from the surface of fossils, or inducing differential weathering 

 of a fossiliferous limestone with impure matrix. The mechanical 

 principle of crystallization involved, in the weathering is, as 

 Mr. Dunn says, an intensification of the principle of repeated 

 freezing and thawing in winter, and is more convenient for the 

 palaeontologist. The same effect may be attained by the use of the more 

 readily crystallizing salt, sulphate of magnesia (Epsom salts). It is, 

 however, a question whether there may not be in the case of sodium 

 chloride some chemical action as well. Here reference may be made 

 to the paper read by Professor B. C. Wallace at the Manchester 

 Meeting of the British Association, 1915, " On the Corrosive Action 

 of certain Brines in Manitoba." The third paragraph of the 

 Abstract (published in the Association Beport, p. 427, and in the 

 Geol. Mag., Jan. 1916, p. 31) indicates a considerable chemical 

 action, due to the fact that a persistent film of concentrated sodium 



