120 OBSERVATIONS ON BASALT, &C. 



the connexion that almoft uniformly exifts, between the ra- 

 diated ftructure and the formation by concentric coats. There 

 are few radiated fubftances which are not divilible into con- 

 centric fragments ; and as few concentric arrangements which 

 are not radiated. Of the firft, it may be fufficient to mention 

 haematites ; of the fecond, calcareous ftala&ites. The len- 

 ■ — dency to this union of ftrn&ure, may perhaps be produced by 



the radiation of the emitted heat, or moifture, if the folution 

 be aqueous ; and the divifions of the coats will naturally 

 take place at thofe paufes in the accumulation of par- 

 ticles, which the momentary emiffion of heat necefTarily in- 

 duced. 

 Formation of the If this be allowed to explain the formation of the nrft feries 

 larger fpheroids f globules which confolidate into the jafpideous fubftance, it 

 will alfo explain the formation of the larger and more diftinclly 

 radiated fpheroids, which have been already ftated to be very 

 eafily divifible into concentric fragments. They probably were 

 alfo formed round a central point, by the accumulation of thin 

 coats ; and the tendency to radiation, which feems almoft infe- 

 parable from this ft ru dure, was perhaps aided by the arrange- 

 ment induced by the emiffion of heat from every part of the 

 iurface of the fpheroids. This mode of formation has the ad- 

 vantage of explaining their impenetrability. Had they been 

 generated by radii diverging from a centre, their compaclnefs 

 muft have diminiflied as their diameter increafed ; but, in the 

 ftruclure which I have fuppofed, each coat is coinpofed of 

 particles folidly arranged in immediate contact with each other, 

 leaving no fpaces for penetration. The fame progrefs is rigidly 

 obferved in the extenfion of the compact nucleus, which always 

 occupies the centre of the radiated fpheroids, and finally ex- 

 tends to their peripheries. It obferves the concentric divisions 

 of the radiated part with the greateft precision ; and the line 

 of their feparation is always perfectly defined. But the ftate 

 of aggregation into which the fubftance has now entered, is 

 fo perfect as to overcome the operation of the caufes which 

 formerly induced the fibrous ftruclure, and the mafs remains 

 compact. The only change that the fubftance afterwards un- 

 dergoes, confifts in the gradual accumulation of the cryftalline 

 molecules, and their arrangement, by their individual polarity, 

 into regular folids. This depends on very different laws from 



thofe 



