Only two occurrences of igneous rock are known: (a) those in Elliott County, and (b) those in the fluorspar region of Caldwell, Crittenden, and Livingston counties, both basic dikes. As indicated elsewhere (Crypto-volcanic Structures, p. 151) there is some reason to believe that deep lying intrusives exist beneath central Kentucky.

Elliott County dikes.—The dikes occur in four small areas, three in the drainage of Ison Creek and one in the drainage of Creech's Creek in the headwaters of the Little Fork of Little Sandy River. They were discovered by Crandall and a study made of them in 1884 by Diller (1885, 1886, 1887, 1892) and Crandall (1886, 1886a).

The rock is a porphyritic peridotite with abundant phenocrysts of olivine, in places altered to serpentine. Garnet (pyrope) and ilmenite are common, along with some plates of biotite, enstatite, and apatite. It has a brecciated appearance and contains angular inclusions of bed rock through which it was injected, particularly of unaltered black shale regarded by Phalen (1912) as resembling the black shales of the Pennsylvanian.

The dikes have been found penetrating the Allegheny, which dates the intrusion as post-Allegheny. The writer has recently examined samples from a core of the Weir sand of Magoffin County. In places the sand contains some biotite flakes and seams of fluorspar. It may be that volcanism was in part earlier.

The area of outcrop is small and the presence of peridotite is recognized by the residual products of decay, principally garnet and ilmenite. The occurrence is of interest in that: (a) The rock resembles the south African peridotite of the Kimberly diamond district. The dikes penetrate both carbonaceous shales and coal beds. Though prospected by shaft, no diamonds have been found, (b) It is an uncommon type of rock in the Appalachians, (c) It may well be related in origin to the western Kentucky mica peridotite.

Western Kentucky Mica-Peridotite.—Dikes and sills have been known for years. Currier (1923) reported 20 exposures in the Kentucky district and ten in the Illinois field. Though genetically related to the fluorspar (Currier, 1923, p. 21), the dikes have no structural connection with the vein matter. "The period of most pronounced faulting and its later mineralization appears to have occurred subsequently to the intrusion of the igneous rock. . . ." Two types of rock are recognized (Diller, 1892; Smith, 1905; Bain, 1905; Currier, 1923): (a) mica peridotite, and (b) pyroxene lamprophyre. They contain conspicuous foils of a dark brown mica and an appreciable amount of small pyrite grains. "Under the microscope the most striking features are the abundance of apatite needles in some of the rocks, of small grains of magnetite and ilmenite with much associated leucoxene, the presence of moderate amounts of a brownish to yellow colored mineral of strong relief but sometimes nearly opaque, which is probably perofskite, and the absence of any feldspar grains" (Currier, 1923, p. 23). The rock weathers rapidly forming a dark brown clayey and slightly sandy soil. A maximum thickness is given as 10 to 12 feet. The dikes strike northwesterly and occupy fissures of little or no displacement.