92 CorresjJondence—jWr. T. E. Waller— Mr. W. Gunn. 



compensate the effect, that at any point of the luminous brushes 

 darkness couhl be restored ; and this was found to be the case when 

 the stretched part of the glass strip was superposed. On the other 

 hand, either pair of the quadrants could be brightened by placing the 

 length of the glass corresponding with their direction and bringing 

 the compressed side into action. The effects are still more marked 

 when a selenite is used, but disapj)ear entirely on inserting a quartz 

 plate until the analyser is somewhat rotated. 



In one case I have met with appearances quite the reverse of those 

 described by Rutley in the case of perlitic cracks. One of these 

 surrounds a crystal to the extent of about f, and toifhin this the field 

 is dark, while the line of the crack is fringed with light due to 

 pressure in the glass surrounding the pearl. Connected with this 

 encircling crack, although, owing to the thickness of the section, the 

 connection is not very definitely traceable, is a long straight crack ex- 

 tending both ways from the crystal. At the extreme points of this the 

 brushes reappear, and when the crack is parallel to the principal 

 plane of either of the Nicols prisms, they are quite brilliant. We 

 have here evidently the expression of the rending force which 

 at this point was not able actually to separate the particles of the 

 glass, but only to produce the strain which results in depolarization. 



The connection which Mr. Kutley suggests between strain and 

 crystallization is, I think, scarcely available here, as the glass is 

 compressed, and the phenomena seem to me rather to point to a 

 higher coefficient of expansion by heat for the obsidian glass than 

 for felspar and magnetite producing in the former, since it was fitted, so 

 to speak, to the crystal at a high temperature, a strain similar to that 

 of an iron tire shrunk on the woodwork of a wheel. If any data of 

 the expansions of the substances involved are known it would be easy 

 to test this, but I do not know where to find anj^. 



Birmingham. Thos. H. Waller. 



POSIDONOMYA BECHERI.i 

 Sir, — I have just seen in your interesting article in the last Num- 

 ber of the Geological Magazine, " On the Discovery of Trilobites 

 in the Culm Shales of Devonshire," a reference (on p. 540) to the 

 occurrence of Posidonomya Beclieri in the rocks at Budle, Northum- 

 berland. Please allow me to point out that this shell was found 

 there by Sir Roderick Murchison and others, many years ago (see 

 p. 291, Siluria, 4th edition), and the late Mr. G. Tate, of Alnwick, 

 has noticed its presence here along with many well-known Carbon- 

 iferous limestone fossils, eg. Griffithides, Bellerophon, TJnio, Euompha- 

 lus, Chonetes, Hardrensis, etc. (Trans. Ber. Nat. Club, vol. v. p. 73, 

 1863). I never heard till now that any one regarded these beds as 

 " Tuedian." They are underlain and entirely surrounded by beds 

 of the ordinary Carboniferous Limestone type, and there is no 

 Tuedian within many miles of them. If the Posidonomya Beclieri 

 is characteristic of genuine Tuedian, why is it not shown to occur 



1 See ante, pp. 73-76. 



