Let me explain this a little further by means of examples.
The crystals, present in a rock, i.e. in a pebble, are each for themselves uniform entities having a definite intrinsic pattern. But the spatial distribution of those crystals within (i.e. across) such a rock does not represent a definite repeatable pattern. This distribution is caused by several, only loosely related factors.
In the case of the formation of plutonic rocks from molten material there were more than one crystallization processes going on, each one accounting for each individual crystal. The spatial distribution of those crystals (as they finally end up in a rock) is caused by several more or less independently operating factors, relating for example to the cavity in which the melt was cooling, the walls of that cavity, and the location of particles serving as deposition sites. All these factors belong to a much broader dynamical system and are themselves only fragments thereof, or they belong to several different such systems.
In the case of the formation of sedimentary rocks we have to do with crystals already formed elsewhere and now being collected together in virtue of their being subjected to factors belonging to larger systems like weather systems and the motion of (large quantities of) water. A local process of sedimentation resulting in local sedimentary rocks is - as a distribution process - consequently only a fragment of one or more broader dynamical systems.
back to main text .