4 86 



NEW YORK STATE MUSEUM 



they are found to run in all possible directions [fig. 2]. In individual 

 exposures the majority of the joints are seen to be curving instead 

 of straight. This shows that some latitude in direction must be 

 allowed each joint set, but is not decisive as to the amount of allow- 

 ance necessary. Few exposures show more than two good sets of 

 parallel joints, though some show three and even four. Certain 

 compass directions are frequent in certain portions of the quad- 

 rangle, and utterly fail elsewhere, indicating a shifting of direction, 

 rather than a different joint set, it is thought. From the best ex- 

 posures it can be learned that in general there are two sets of joint 

 couples, each couple consisting of two sets of parallel joints which 



v^UUU/// 



Fig. 2 Diagram of the readings on joints within the Long Lake quadrangle. The inner 

 figures represent the compass degrees east and west of true north; the outer row the num- 

 ber of joint readings observed for each 5 direction. Four numerical groups are also 

 indicated. 



approximate a right angle with one another, and that one couple 

 tends to occupy the meridional and equatorial directions, the othet 

 running northeast and northwest. An attempt may be made to 

 classify the entire number of -readings on this basis, assuming that 

 each set has a variation in direction of 20 owing to swerve. Thus 

 the n. 8o° e. to n. 8o° w. joints are grouped together, forming a couple 

 with then. io°e. to n. io° w. joints; in like manner the n. 40 e. to n. 

 6o° e. and n. 30 w. to n. 50 w. joints are grouped. The numerical 

 results of this grouping are indicated in the diagram, and a greater 

 number of the readings are included, on this basis of subdivision, 

 than on any other possible basis using the same amplitude of swerve. 

 It is quite certain that some of the joints have a swerve of this 

 amount, but it is not impossible that they may have even more. 



