192 Proceedings of the Royal Physical Society. 
There are certain cases, I admit, where the occurrence of a 
single species goes far to prove the age of the rocks from 
which it was derived. Experience has taught us that a 
greater paleeontological value for determining horizons belongs 
to certain genera and species than to others, and there are 
many cases where certain plants and animals have never 
been found but on one particular horizon ; still, great care is 
necessary in dealing with such negative data, for there is 
always the possibility of such organisms being found in other 
horizons. 
If, on the other hand, we deal with a collection of fossils, 
containing a greater or less number of species, the risk of 
giving undue importance to any given species is done away 
with, and here another very important factor enters in—the 
relative proportion of the species. 
The value of this relative proportion of the species is well 
illustrated when dealing with the correlation of the various 
divisions of the Coal-Measures, as developed in Britain. 
Tf one carefully examines the floras of the Upper, Middle, 
and Lower Coal-Measures, it is found that of the few species 
which extend through all the divisions, such species are usually 
common in one, or even two of the divisions, and extremely 
rare or absent in the other division or divisions, as the case 
may be; hence any possibility of giving an undue importance 
to the rare occurrence of a species, which may be common 
in some other horizon, is entirely eliminated by taking into 
consideration the prevalence of the other species with which 
it is associated, and which may be extremely rare or quite 
unknown in the other divisions. 
Any characters derived from colour or lithologieal character- 
istics of strata are of little use in the correlation of horizons, 
except in restricted areas. In many cases where colour has 
been accepted as evidence of the age of a rock, it has led to 
error. On one occasion Professor Hull says: “ In connection 
with the colour of these sandstones, it may be stated that it 
has been found in the neighbourhood of Rainford, that sand- 
stones, which at thé surface assume a purple colour, are 
found to be brown or grey where pierced by coal-shafts, and 
