production of the Prismatic Structure of Basalt. 223 



Such are the conditions of mutual compression which most 

 writers on the subject since Mr. Watt have had somewhat con- 

 fusedly in their minds (see Scrope, 'Volcanos/ 2nd edit, foot- 

 note p. 102 et seq.). But the conditions assumed by Mr. G. 

 Watt are even still more far-fetched and chimerical, and may be 

 stated in his own words : — " In a stratum composed of an inde- 

 finite number in superficial extent, but one only in height, of 

 impenetrable spheroids with nearly equidistant centres, if the 

 spheroids should come in contact on the same plane, it seems 

 obvious that their mutual action would form them into hexa- 

 gons ; and if these were resisted below and there was no oppo- 

 sing cause above them, it seems equally clear that they would 

 extend their dimensions upwards, and thus form hexagonal 

 prisms whose length might be indefinitely greater than their 

 diameters " (Phil. Trans. 1804, Part i. p. 307). This suppo- 

 sition leads to even more extravagantly improbable results than 

 the former*. None of these authors condescends to inform us 

 whether their spheroids under compression are to be supposed 

 cold and rigid, or heated so as to be plastic. We may infer from 

 the context, however, in the case of Mr. Watt, if not also in that 

 of Mr. Scrope, that each spheroid is in a state of plasticity by 

 heat, thus yielding to pressure as an imperfect liquid. Vertical 

 prisms are recorded by authors 1 50 feet in length and only 9 

 inches in diameter (Scope ' Volcanos/ p. 102), and others 6 

 to 8 feet in diameter and 40 feet and upwards in length. If we 

 assume a prism of the smaller of these diameters and only 100 

 feet in length, it would demand a sphere for its production in 

 the supposed manner equal to its own volume, or 4*1 feet in 

 diameter, and, in the case of the larger diameter and length of 

 prism, of 8*4 feet diameter. No instance has been observed, or 

 probably exists, of any extensive collection of spheroids of either 

 of these large sizes, or even approximately all of equal diame- 

 ters. Admitting, argumenti gratia, the existence of some un- 

 known external compressive force sufficient to overcome all the 

 resistances, including that of the hydrostatic pressure of the 



* Mr. Watt's paper presents one of those singular examples, of which 

 so many others exist in the records of scientific research, of how near an 

 author may approach to completely discerning a truth and yet be blinded 

 by a preconceived and falsely based hypothesis. At page 311 (Phil. Trans, 

 as above) it will be seen that Watt had some clear ideas as to the splittiug 

 up into hexagonal prisms by contraction in cooling ; but immediately after 

 (pp. 313-314) he dismisses this all in the words, "though these conside- 

 rations may be sufficient to explain the tendency to division into prisms 

 which is so generally extended .... they are utterly inadequate to the for- 

 mation of the more perfect basaltic prisms ; they offer no means of account- 

 ing for the extreme regularity of the sides and the precision of the angles, 

 for the articulations, for the close contact in which the perfect columns are 

 placed to one another," &c. 



