QUANTITATIVE OBSERVATIONS 55 



often two or more observations were made close together as a check on 

 the one recorded. The possible error is all on the side of conservatism 

 for two reasons. In the first place, along the elevated shoreline occasional 

 barnacles are still living just above the zone of barnacle growth, having 

 been able to survive these six years with an occasional bath of salt water 

 from waves at high tide. Consequently when we took the highest living 

 barnacle, always searched out very carefully, it was often the only living 

 one among many dead forms and several inches above the upper limit of 

 abundant live barnacles. In the second place, the highest dead barnacle 

 actually attached to the rock (and we took account of no others) was 

 often probably not as high as barnacles had grown at the former level of 

 the sea. If the conservative error on both ends could be corrected, we 

 have no doubt that it would often increase the uplift by 6 inches or a foot, 

 but would never diminish the amount stated. 



In a very few places where living barnacles were absent from the present 

 shoreline, as along the coast south of Turner glacier, our measurements 

 are between the high-tide mark and the top of the zone of abundant dead 

 barnacles, which was assumed to represent high tide on the old strand. 

 On the elevated beaches we made one or two measurements between par- 

 allel lines of driftwood; and two or three measurements were made be- 

 tween the lower limit of land plants on the ancient beaches and the lowest 

 old bushes above it, a possible error of only a foot or two. It should be 

 pointed out that the greater number of our measurements (approximately 

 80 per cent of them) were made by the conservative method of measure- 

 ment between highest living and highest dead barnacles, and that all other 

 cases, where another method was necessary, checked well with closely 

 adjacent localities where barnacle observations were possible. The coast 

 south of Turner glacier is an exception, for here barnacles were used only 

 at the upper limit. 



Our measurements of depression are much less exact. For these we 

 measured the vertical distance between the base of the lowest tree in place 

 and of the highest on which beach gravels were being piled at the time. 



CHANGES OF LEVEL ON THE FORELAND 



Taken as a whole, the foreland and its associated islands may be con- 

 sidered as a region of no change of level, though with small areas of 

 slight depression, usually too slight for quantitative measurement. On 

 the west side of Yakutat bay, from opposite point Latouche to the Kwik 

 delta, the shoreline was studied carefully, but no change in level could be 

 detected on the foreland. On the southeast side of the bay, near the 

 mountains, both on Knight island and on the mainland, the changes of 



