ANNIVERSARY ADDRESS OF THE PRESIDENT. 85 



mineral is in the act of crystallizing and the other has crystallized, 

 the latter is inert while the former is active.) It may also happen 

 that in this crystal-building the attached microliths are novr diverted 

 in this way, now in that ; thus twin-building of various kinds will 

 result, and I should expect that this twinning might be more 

 perfectly developed by molecular rearrangement after the crystal 

 had formed, but while it was still plastic. 



If this process of aggregation of one or more minerals be carried 

 on to a considerable extent, the residual magma will obviously differ 

 much from the original. At last this also may begin to crystallize. 

 Then, if the fall of temperature, or the change of conditions be very 

 slow, we shall have a coarsely crystalline mass enclosing the earlier 

 developed crystals ; but if the change is more rapid, one more finely 

 crystalline will be produced. If the magma contain the constituents 

 of minerals, of which any one (under the circumstances) crystallizes 

 at a markedly higher temperature than the rest, we may expect 

 perfect crystals of that to occur; if not we shall find that the 

 minerals, though thoroughly crystalline in structure, are very 

 imperfect in their external form. 



In the case of slow consolidation it is obvious that if we suppose 

 at different points in the mass the existence of centres of attraction 

 of any kind, acting uniformly in every direction, the microliths as 

 they form will be aggregated around them with a radial structure and 

 thus will form spherulites. On this structure I have some further 

 remarks to make, in relation to vitreous rocks, in which it is far 

 more common ; but I may point out here that the curious orbicular 

 diorite may be thus explained. In these globular masses we can 

 generally see a more or less indistinct nucleus, then follows a series 

 of subspherical bands of anorthite and hornblende, the former predo- 

 minating. The process, then, appears to be as follows : — The two 

 minerals have crystallized almost simultaneously, the felspar having 

 had very slightly the advantage. The nucleus, possibly only in 

 consequence of the accidental presence of a slight excess of anorthite, 

 acts as a centre of attraction, and anorthite from the parts of the 

 magma in immediate continuity, separates out and collects radially 

 upon the existing nucleus. But after this segregation of the 

 felspathic constituents from the magma has gane on for a certain 

 time, there is a zone of it in which is a residual excess of the horn- 

 blendic constituents, so that circumstances now admit of the forma- 

 tion of the latter mineral ; and thus a zone of it is built up, until 

 again the inequalitj^ is more than redressed, and the formation of 

 anorthite recommences. But this will not go on indefinitely ; for 

 general crystallization of a non-radial character will have been set up 

 in the mass, so that at last molecular motion becomes impossible. 

 The construction of spheroids ceases, and they are enclosed in an 

 ordinary diorite. I notice that where the boundary of the spheroid 

 is the most sharply defined, the outer ring is a thin one and is 

 hornblende, as it should be with the mineral second in order of 

 formation. 



Leaving, then, for a moment, certain structural peculiarities, such 



VOL. XLI. i 



