30 Records of the Indian Museum. [Vol. XXIII, 1922.] 



The general conclusion of the above investigation is this. 



There is no indication of appreciable "curtailing" of our 

 material. Further, with small samples, the i( abruptness coeffici- 

 ents " cannot be calculated with an}- reasonable degree of accuracy 

 and these " full corrections " will necessarily have to be omitted. 

 But we have already seen that Sheppard's correction can be safely 

 applied and should never be omitted. 



