CRYSTALLIZATION-DIFFERENTIATION IN MAGMAS 425 



It has been considered that the nodules consisting mainly of a 

 single mineral and occurring fairly frequently in basaltic rocks may 

 represent immiscible liquid globules. Olivine nodules in basalt are 

 perhaps the commonest examples. The term nodules is rather 

 misleading, for they are characteristically angular and justify the 

 conclusion of a great number of petrographers that they are frag- 

 ments torn from a mass of olivine rock (Olivinfelseinschliisse). 

 Rosenbusch, however, always insisted that they represent intra- 

 telluric crystallization from the rock mass in which they are found. 

 As a matter of fact it is more reasonable to suppose that both 

 conceptions are true, and that the nodules represent fragments 

 torn from an olivine rock that was formed from the basalt by 

 crystallization-differentiation. It is not necessary to suppose that 

 the olivine settled to the bottom of an intratelluric mass of basaltic 

 magma; indeed the characteristic occurrence of the nodules in 

 narrow dikes and small necks of basalt suggests an action that may 

 be of considerable importance in petrogenesis. The suggested 

 action is the formation of crystal jams, or the clogging of the 

 channels occupied by moving magma as a result of the packing 

 together of crystals at a constricted point, in much the same manner 

 as log jams and ice jams are formed in rivers. Any marked constric- 

 tion of a channel in which a liquid is flowing that contains suspended 

 solids, even in very small amount, is likely to cause such clogging. 

 The formation of a crystal jam is highly probable in a narrow dike 

 or neck through which basaltic magma is flowing with upward of 10 

 per cent olivine crystals suspended. Once formed the clogging may 

 result in the accumulation of a considerable mass of crystals, and 

 the passing of liquid through the mass, a Hquid from which ohvine 

 and only olivine is being precipitated as yet, may cause the growth 

 of the crystals of olivine until no interstices are left. Accumulation 

 of pressure on the magma, aided perhaps by terrestrial disturbance, 

 may then break the barrier and carry the olivine rock along as 

 inclusions of the nature of olivine nodules, some of which may 

 become rounded as a result of the resorption that olivine normally 

 suffers. While the clogging of a channel with very early crystals, 

 such as olivine, is to be expected only in small channels, there is 

 reason to believe that even large channels might be similarly affected 



