22 



this to Graham. (See reference in Bibliography.) He found that 

 through a thin membrane, as parchment, crystallizable bodies 

 will pass, colloid bodies will not. Hence this supplies a means of 

 separating the two. A liquid holding silicia in solution, such as 

 one containing salt, and different carbonates, would pass through 

 the membrane of the original sponge perhaps, or through the 

 first thin coating of colloid silica deposited on the exterior, and 

 the colloid silica be left behind . 



Here is a singularly interesting specimen of a flint which has 

 evidently received a considerable addition to it since its first 

 formation. It is clear from this that the silica of the flint which 

 served as a nucleus must have been consolidated previous to the 

 later deposition. But the question is, and much depends upon 

 it — How long after the^r«^ deposition did the second take place? 



Looking at the flints then generally we may note that there 

 is almost always something not flint which served as a nucleus 

 round which the silica aggregated. It is a shell, or the test of an 

 echinus, or more probably a sponge. But the manner in which 

 the silica was held in solution, what caused it to form con- 

 cretions round these substances, what determined its solidi- 

 fication — are some of the most important problems yet to be 

 solved. 



Banded Flints. — In all probability Nature has followed more 

 than one method in the production of these. The " banding " 

 has sometimes arisen from structural differences in the sponge 

 about which they are formed, or from chemical reactions 

 between the silica and the iron, &c., dissolved in the water 

 which has enveloped the nodule since its formation. (See 

 references to Euskin and Woodward in Bibliography.) 



PROBLEMS CONNECTED WITH THE MANNER IN 

 WHICH FLINT OCCURS IN THE UPPER CHALK. 



In most places where there are cliffs or quarries in the Upper 

 Chalk in this neighbourhood, the flint may be observed to occur 

 in the following modes : — 



1st, Layers of nodular concretions, passing often into tabular 

 masses, both following the planes of bedding. These layers are 

 from two to five feet apart. 



2nd, Sporadic nodules, nodules appearing here and there 

 irregularly in the chalk. 



3rd, True vein courses inclined at various angles to the 

 planes of bedding. 



