BARKER : PRESIDENTIAL ADDRESS 
11 
and the ratio of helium to uranium is found to be higher in the older 
deposits. Estimates of age calculated from these data give high 
figures, e.g., the age of the haematite overlying the Carboniferous 
Limestone in Cumberland is given as 140 millions of years, and even 
that of the Eocene iron -ores of Antrim 30 millions. The results show 
some irregularities, and it is of course admitted that the method has its 
own difficulties. If, however, the chief source of error is, as appears 
probable, the loss of helium by leakage, the figures found will be under- 
estimates. Helium comes from the thorium series of derivatives 
as well as from the uranium series, and this is to be taken into account 
where thorium is found. Zircons from various igneous rocks have 
also been examined by Strutt, and found to give consistent results as 
regards the helium-ratio. Mr. Holmes has approached the question 
in a different way by considering the ratio of lead to uranium in various 
minerals rich in the latter element. The igneous rocks of the 
Christiania district, of Devonian age, are in this way calculated to be 
about 370 million years old. For the Archaean rocks of different 
countries the estimates range from 1,000 to 1,600 millions of years. 
Holmes' results are in general nearly twice as high as those of Strutt ; 
but, if we bear in mind the error due to the escape of helium, which 
is proved to take place, a discrepancy to this extent is no more than 
should be expected at this early stage of the inquiry. 
The other method which has been suojorested for obtaining an 
absolute measure of geological time is of a more speculative kind, 
although the principle of it is sufficiently simple. It consists in 
detecting some clearly marked rhythm or cycle in the geological record, 
and correlating it with one of the known periodic movements of the 
earth. It was on these lines that CroU attempted to explain recurrent 
glacial epochs ; but more to our present purpose is the theory which 
Blytt has based upon a study of the alternations observed in a suc- 
cession of sedimentary strata. The most important astronomical 
cycle of long period is doubtless that which depends upon the pre- 
cessional movement, by which the relation of summer and winter to 
perihelion and aphelion is gradually changed. This involves a change 
in the relative lengths of summer and winter, and must undoubtedly 
exercise a marked effect upon climatic conditions, though there is 
much difference of opinion concerning the exact nature of this effect. 
Changes of climate may in their turn cause differences in the nature 
of the sediments deposited successively at a given place, differences 
which will repeat themselves in a cycle corresponding with that of the 
