ISOSTASY SHS 
some cases erosion to a peneplain has been followed by subsidence 
and in other cases by uplift’? and notes that the context seems to 
imply that this cannot be reconciled with the theory of isostasy, 
one is puzzled to understand why no reference is made to that part 
of Hayford’s address in which two long paragraphs are devoted 
to setting forth the errors in this particular line of reasoning. 
Thus far in this rejoinder only such evidence has been cited as 
was available to Mr. Lewis when he wrote his article. In May, 
1912, the Coast and Geodetic Survey issued a publication entitled 
The Effect of Topography and Isostatic Compensation upon the 
Intensity of Gravity, written by John F. Hayford and William Bowie. 
This contains further evidence pertinent to the questions under 
discussion. 
In the principal computations set forth in this new publication 
the effects of isostatic compensation upon the intensity of gravity 
are computed upon the assumption that the compensation is 
complete and is uniformly distributed to the depth 113.7 km. 
The depth was adopted from the first figure of the earth investi- 
gation by Hayford and the assumption of complete compensation 
was retained. 
Let the evidence furnished by this new publication on gravity 
now be considered in connection with Mr. Lewis’ article. 
In the new gravity publication the computations are made by 
concentric circular zones as in the computations of topographic 
deflections, but the zones are not the same. In the new publica- 
tion in the table on p. too the effect upon gravity, at each of 41 
stations in the United States, of the assumed compensation for all 
zones out to zone O, covering all areas within 166.7 km. of the 
station, is tabulated separately from the effect of the topography 
itself. This enables one to apply two tests as indicated below. 
First, let it be assumed that for these 41 stations the compen- 
sation is only nine-tenths complete. Then the computed effect 
of the compensation as shown in the table should in each case be 
diminished by one-tenth of itself. This would produce a contrary 
change of the same magnitude in the anomaly, or residual, shown 
in the fourth column from the last in this table. An inspection 
of the values shows that for the 41 stations 19 residuals would be 
