Tachylite of Cleveland Dyke. 67 



This supplies us with a good example of the power of differentiation 

 by crystallization, and it seems probable that if pressure had supervened, 

 so as to strain off the crystals of the ' later groundmass ' and squeeze 

 out the latter, an acidic rock would have resulted. Although this 

 possibility was not realized in this case it has been an important factor 

 controlling differentiation in other instances. 



General Eeview and Conclosions. 



We find a 'stony rock', consisting of a base in which there is evidence 

 of incipient crystallization, passing quite abruptly into a layer of 

 tachylite (5 to \ inch) in thickness. The ' stony rock ' is granular in 

 appearance owing to the presence of minute specks of iron-oxide. 

 These increase markedly in size and number in the rock adjacent to 

 the tachylite, giving rise to the ' dark area '. Dark streaks of 

 a similar nature mark the lines of flow in the paler rock (PI. IV, Fig. 1). 



The spherulitic devitrification of the tachylite in contact with the 

 adjacent layers may be evidence of a not perfectly homogeneous 

 condition at the time of solidification, but the outer part of the glass 

 is certainly homogeneous, for there the devitrification is clearly 

 guided by the cooling cracks along which it started. 



There are crystals — porphyritic and smaller individuals — in both 

 parts of the slides, but they are much more altered, faulted, and 

 deformed in the ' stony rock '. Some curved felspar laths in the 

 tachylite show that the temperature there had approached that of their 

 melting-point. Small masses of granular augite associated with 

 felspar laths, either in stellate groups or tangentially disposed to form 

 a border, occur here as well as in other parts of the intrusion. They 

 have been found in every section from all localities, but they are 

 certainly largest and most numerous in the central parts of the dyke. 

 That they are smaller and rarer here is distinctly opposed to the idea 

 that they are a fine-grained lateral consolidation, which has been 

 broken up by a later melting of the rock, while there is nothing to 

 disprove the suggestion of a local eutectic mixture produced by the 

 progressive abstraction of material in a small area walled off by the 

 crystals (felspars usually), which by withdrawing materials for their 

 ■own growth give rise, instantaneously perhaps, to the conditions 

 necessary for the spontaneous crystallization of the remainder. 



The distribution of the cracks is interesting. In the outer part of 

 the tachylite they are ubiquitous ; then they become scarce and never 

 extend into the adjacent ' stony rock '. In this latter one set of 

 cracks runs at riglit-angles to the junction, but no crack passes into 

 the tachylite. Another set, more or less parallel to the junction of 

 the two types of rock, bounds the very irregular limits of the ' dark' 

 rock, while a third set converges towards the second at a large angle 

 and does not as a rule pass it. 



The dark area seems to bear the same relation to the adjacent rock 

 that the tachylite bears to it; in other words, it would seem that the 

 tachylite cooled first, forming a rather firm viscous film, past which the 

 still molten magma continued to flow. Evidence of this is found in 

 the appearance of one or two large felspars, which suggest having 



