170 PROCEEDINGS OF THE ACADEMY OF [May, 



There can be no question, however, as to the igneous origin of the 

 rock. When studied in thin section it is found to consist of inter- 

 lacing laths of labradorite surrounded by augite similar to the contact 

 facies of Triassic diabases which have been described elsewhere. 2 

 The lines turn out to be dikes, composed of coarser crystals of 

 the same minerals (fig. 3, Plate II). These dikes vary from \ to 5 mm. 

 in thickness, and penetrate into the rock to depths of several centi- 

 meters, gradually losing their identity as the size of grain of the 

 background increases. 



The explanation of this occurrence is apparently to be found in the 

 tendency of magmas to contract on solidifying, exemplified in the 

 well-known columnar structure of many basalts. In the course 

 of the intrusion of the magma the outer portions were rapidly chilled 

 by the cold wall-rocks, and solidified in correspondingly fine-grained 

 form. At the same time there developed, perpendicular to the 

 contact surfaces, hexagonally arranged shrinkage cracks. Into these 

 the still liquid material beneath found its way, but since in the 

 meantime the surroundings had become heated, cooling was now less 

 rapid and larger crystals were formed. 



The boulders which everywhere characterize the diabase areas 

 often show on their surface a hexagonal-crack effect resembling at 

 first sight that just described (fig. 2, Plate II). The two phenomena 

 are, however, quite distinct, for thin sections of the rock beneath 

 these cracks fail to indicate the existence of any structural peculiarity 

 (fig. 4, Plate II), and the cracking is evidently caused by expansion 

 of the surface layers during the exfoliation of the boulders under the 

 action of frost. 



The striking similarity of these two effects to one another, and of 

 both to sun-cracks, is due, then, to the fact that all three have their 

 origin in tension exerted' uniformly in a plane, the normal result of 

 which is, as is well known, the development of cracks at approximate 

 angles of 120 degrees. 



The blocks of diabase are occasionally collected into "boulder 

 fields" — in Germany termed Felsenmeere, or rock seas — tracts often 

 an acre or more in extent, which are practically barren of vegetation 

 (because of the absence of soil), in striking contrast to most of the 

 surrounding region (fig. 7, Plate II). Many of the boulders ring like 

 a metal when struck, and under the name of "ringing-rocks" have 

 attracted considerable attention from the inhabitants of the neighbor- 



2 Lewis, J. V.: Ann. Rept., State Geol, New Jersey, 1907, 115, pi. XVI. 



