12 
HARKER : PRESIDENTIAL ADDRESS 
precession. Probably the most noticeable effect will be a recurrence 
of limestones and chemical deposits alternating with detrital sediments. 
If the matter were no more complex than this, it would be sufficient, 
where such alternations can be detected, to count them, as we count 
the rings of growth of a tree, and reckon 21,000 years for each sedi- 
mentary cycle, that being the period of the precession corrected for 
the movement of the perihelion. If the alternations can be dis- 
tinguished only in some parts of the succession, some hypothesis must 
be devised to take account of the intervals. Gilbert has discussed in 
this way a succession of beds, 3,900 feet thick, forming part of the 
Cretaceous system in Colorado. Alternations of calcareous beds with 
shales come in four times, being separated by unbroken thicknesses 
of shale. Gilbert calculates for the part of Cretaceous time repre- 
sented a duration of about 20 million years, with an uncertainty 
indicated by the number 2 as a " factor of safety." 
We have to remember, however, that sedimentation is controlled 
by other conditions besides climate, and climate depends upon other 
causes besides the precession of the equinoxes ; and, further, that most 
of these contributing causes cannot be described as periodic in any 
intelligible sense. There is, it is true, a second astronomical movement 
to which both Croll and Blytt have made appeal, viz., the variation in 
the eccentricity of the earth's orbit. This goes through a period of 
about 90,000 years ; but there are considerable irregularities which 
repeat themselves in the course of 1,450,000 years, giving a larger 
cycle which embraces sixteen of the smaller cycles. The change of 
eccentricity must modify the effect of the precessional movement ; 
but Blytt argues that it will also react on the ellipsoidal shape of the 
globe itself, and so give rise to a displacement of shore-lines. He 
claims to have traced this effect, as well as the climatic cycle, in such 
cases as the succession of the Tertiaries in the Paris basin and the 
Isle of Wight. His conclusion is that Tertiary time comprises two 
of the larger cycles, i.e., about three million years. 
It has usually been assumed that the year is too short a period to 
leave any recognizable mark on the geological record. This is 
probably true in general, but in certain favourable circumstances it 
may perhaps be possible to count annual layers of sediment. De 
Geer has recently attempted this in the case of certain finely laminated 
clays of late Glacial and post-Glacial age in Sweden. The material 
was brought down by sub-glacial streams at a time when the ice had 
retreated to the higher ground. Consequently the seasonal variations 
