more characteristic beds of quartz rock with which it is associated.
The next remark I shall make is on the term syenite, which in the paper alluded to is applied to that rock which I have called granite. The difference here is merely a question of expediency, and I shall content myself, in addition to what I have already said on that subject, with a brief statement of my reasons, as I shall have occasion to discuss this point at some length in giving an account of the Western islands.
Werner has described syenite as an overlying formation, and incumbent on granite. With a rock of these characters I have no acquaintance, and his description is therefore unintelligible to me in a practical view.
But there are in Scotland (and elsewhere) two compound rocks formed of the ingredients described in his definition, namely, quartz, felspar, and hornblende, but occupying two positions, in a geological view most distinct, and neither of them agreeing with that which he has assigned for his syenite. That to which I have thought it expedient to limit the term, is incumbent upon the secondary strata and at the same time interferes with them in the same way that certain varieties of trap and porphyry, with which it is intimately connected, are known to do; examples of this are to be found in many of the Western islands. It is therefore not only posterior to these strata and distinct from the syenite of Werner in its geological relations, but it differs from it in composition, inasmuch as it does not contain mica, unless, as some also of the traps do, accidentally.
The other rock which agrees with the mineralogical definition of syenite, is found connected with granite and consequently subjacent to the most ancient stratified rocks. I have attempted to shew
