F. H. Lahee — Dodecahedral Jointing. 169 



Art. X. — Dodecahedral Jointing due to Strain of Cool- 

 ing • by Fred. H. Lahee. 



In the woods just south of Beacon St., and about a quarter 

 of a mile west of Hammond St., Chestnut Hill, Mass., the 

 Roxbury conglomerate is cut by a fine-grained, basaltic dike* 

 which, in some parts, disintegrates into small (average dia- 

 meter, -§- in.), polyhedral, often roughly dodecahedral, frag- 

 ments. The dike has approximately plane-parallel sides, is 

 four feet thick, strikes K 20° E., and dips 78° E., its attitude 

 being parallel to a prominent joint set of the country rock. 

 That it entered a relatively cool rock is indicated by the fact 

 that its texture, moderately fine in the middle, becomes very 

 line at the contact. It contains occasional large phenocrysts 

 (xenocrysts) of apatite and biotite, both of which are well 

 shaped, and of pink acid feldspar, which has outlines made irre- 

 gular by the invasion of short tongues of the groundmass. 



Near the country rock, on each side, in a zone from four to 

 ten inches wide, hexagonal columnar jointing is poorly devel- 

 oped perpendicular to the contact surface. Inwards, the col- 

 umns give place to the polyhedral blocks already mentioned as 

 conspicuous in the disintegration of the rock. 



Obviously the jointsf which give rise to such many-sided 

 fragments are of small extent. Furthermore, they (the joints) 

 often vary in direction, or may die out entirely; but in spite of 

 this irregularity, it is not difficult to find blocks bounded by 

 twelve rhombohedral faces that intersect one another at angles 

 nearly equal to the similar angles of a perfect dodecahedron. 

 The surfaces of the blocks are relatively coarsely granular, 

 without a well-developed feather fracture, a feature not uncom- 

 mon on the more finely textured hexagonal columns. Where 

 the phenocrysts are in the path of fracture, the break either 

 passes round the obstacle, or takes advantage of the mineral 

 cleavage. These statements clearly point to the inference that 

 the dodecahedral jointing, like the hexagonal, is an effect of 

 tension due to cooling. 



Hexagonal columnar jointing has long been so interpreted. 

 In the ideal case, according to the principles of least action, 

 the columns begin their development as a series of three-way 

 fractures (each fissure at an angle of 120° to the other two) 

 radiating from equally spaced points in a surface which is per- 

 pendicular to their (the columns') direction of growth. This 

 type of fracture is therefore a two-dimensional, or surface, 

 phenomenon, and the extension of the columns may be regarded 



* The writer wishes to thank Mr. E. W. Sayles for bringing this dike to 

 his notice, and Professor Chas. Palache and Professor J. B. Woodworth, 

 who discovered the dodecahedral jointing, for valuable suggestions. 



t Similar jointing has recently been seen by the writer in trap dikes on 

 Eagged Island, Casco Bay, Maine. 



