October 1, 1892.] 



KNOWLEDGE 



195 



too that they have the advantage of a bird's-eye view, 

 which anyone entering a real structure made of walls or 

 hedges would not. 



I have at the principal doubling points merely left open 

 two more of the roads than would have resulted from simply 

 treating the dividing wall as another path. Though not 

 necessary, this seems to give a little more variety to the plan. 



It will be seen that there are four places where the right 

 way has to be chosen between five, and as no one would 

 be so stupid as to choose the road by which he had just 

 entered, the chances are about three to one against his 

 taking the right one of the other four. This difficulty is 

 repeated four times in the course of the labyrinth, so that 

 it would seem on a first calculation that his probability of 

 getting in by the best route is about J of ^ of ^ of -^, or, 

 say, 1 to 81. But anyroad will, after an uncertain number 

 of turns and doublings, take him either to the centre or 

 back again to the entrance ; so that the task is not nearly 

 so difficult as it would at first sight seem, nor, indeed, is 

 it at all probable that the designer would have made it too 

 difficult, especially if any body of men had to thrive on the 

 profits, like the artificers who made silver images of Diana at 

 Ephesus. Perhaps even the labyrinth coins took their origin 

 from tokens, given to those who had visited the maze and 

 duly paid for the privilege. They might also serve as a sort 

 of pass token and souvenir of the visit, as well as a key to the 

 mystery. The legend of the colom-ed thread of Ariadne 

 becomes intelligible on this plan of the maze as now shown, 

 because, on returning to any place where he had been 

 before, Theseus would see his own thread and retrace his 

 steps, gathering up the line until he could make a better 

 start — while of course the thread would serve him as a 

 guide to enable him to get out again. If I gave the coin 

 plan, it would simply be each alternate circle of this one, 

 and it would furnish a key by which anyone could imme- 

 diately find the way to the centre. 



The solution, when put into words, is — avoid every 

 alternate circle or path. 



Yours faithfully, Richard Inwards. 



A FLAKE OF FLINT AND ITS HISTORY. 



By R. Lydekker, B.A.Cantab. 



IN our article entitled "A Lump of Chalk and its 

 Lessons," published in Knowledge for June last, it 

 was stated that the upper white chalk, so far as 

 hand specimens are concerned, is a nearly pure lime- 

 stone. This, indeed, is a perfectly true statement as 

 regards such hand specimens ; but when we consider the 

 upper chalk as a whole, we must not omit to regard the 

 numerous bands and nodules of Hint with which it is 

 interstratitied as a very important constituent of the whole 

 rock. We say a constituent of the whole rock advisedly, 

 because although the flint is now separated from the white 

 limestone which we call chalk in the form of nodules and 

 bands, yet there is e\-idence that it was originally dissemi- 

 nated throughout the entire mass, and that the upper chalk 

 then formed a slightly siliceous limestone. Probably 

 everybody is more or less familiar -nith flint as it occurs in 

 a chalk pit, or in the form of gravel derived from the 

 disintegi'ation of chalk strata ; but it may be taken for 

 granted that comparatively few have ever seriously 

 considered how the solid masses of flint have originated in 

 the soft chalk limestone. As this is a subject of consider- 

 able interest, and one which has given rise to much 

 discussion, we propose to devote the greater part of the 

 present article to its consideration, while we shall add 

 some observations on the history of flints after they have 

 been removed from their native chalk. 



We shall assume, in the first place, that all our readers 

 are aware that flint is one of the manifold forms assumed 

 by that abundant constituent of the earth's crust technically 

 known as silica — the oxide of the element silicon. When 

 crystallized, silica occurs in the form of rock-crystal, or 

 quartz ; but flint is one of the many non-crystalline, or 

 amorphous, developments of the mineral. It is generally 

 defined as a massive dark-coloured or black, semi-trans- 

 lucent, dull-looking variety of silica ; which when pure 

 burns to an opaque white, and has what is termed a 

 conchoidal or shell-like fracture. The dark colour, it may 

 be added, is due to the presence of a smaU quantity of 

 organic matter, or carbon. 



If we fracture a nodule of flint freshly taken from a 

 chalk pit, we shall find that the thin edges of the sharp 

 flakes which are seen to be produced have a pale-brown 

 horn colour when viewed by transmitted light, and that 

 as the flake becomes thicker the colovn- gradually darkens 

 till it assumes the blue-black tint characteristic of the 

 mass. If, however, our flake contains a portion of the 

 external coat of the nodule, we shall see that for about a 

 quarter of an inch in depth the outer layer is far less 

 compact than the rest of the flint, and is of an opaque 

 white colour. In other specimens, again, we shall observe 

 that the colour, instead of the usual deep blue-black, is 

 of a pale whitish blue, frequently marked by more or less 

 distinct bands. Needless to say that on trying to scratch 

 our flint flake with a knife we shall signally fail, and if 

 any result happens it will be that a thin film from the 

 metal of the blade will be left on the stone at the point of 

 contact. 



FlQ. 1. — A Chipped Flint Implement, from Iclilingham, half 

 natural size. (From Sir J. Evans' " Stone Implements.") 



Our flake will likewise exhibit in great perfection the 

 characteristic conchoidal fracture of flint ; that is to say, 

 its surfaces will be smooth and undulating, swelling here 

 into a prominent convexity, and falling there into a deeper 

 or shallower hollow. Frequently, moreover, there may be 

 observed a number of small parallel wavy ridges on the 

 fractured surface. If we submit the edges of the freshly- 



