250 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1949 
nor even in successive species of ages as much apart as 1 million years, 
to believe that the minute differences present had any value as a 
determining factor in survival. 
Returning to the duration of a genus, it will be interesting to 
consult other groups, to see whether the figure of 5% million years 
found in the horses is more generally applicable. Similar values are 
found in other groups of Mammalia, such as the land carnivores. 
For these Simpson constructed a ‘‘survivorship curve’ showing that 
the mean duration of a genus of carnivores is 6% million years. But 
for the bivalve Mollusca, he obtained a mean as high as 78 million 
years. 
Just to add a few more figures: Swinnerton found 20 million years 
as the average lifetime of a genus of Triassic ammonites. Lingula, on 
the other hand, a brachiopod, has been known for its persistence since 
the Cambrian. This genus has lasted over 400 million years. From 
this evidence the terrestrial Mammalia appear to have genera of a 
short duration, but even this is a matter of a few million years. 
One might be inclined to interpret these different rates of generic 
evolution in terms of species steps, assuming that each lineage would 
pass through several evolutionary changes, each constituting a new 
species, before the accumulated differences justify one in calling the 
descendants a new genus. It is certain that this process underlies 
the evolution of the horse, but it is conceivable that in other groups 
genera may have originated without the interposition of a series of 
species steps. The time rate of species formation, therefore, must 
now be considered. 
Selecting again terrestrial forms of life, mainly mammals, as our 
first examples, because more detailed evidence is available, we find 
that since the end of the last glaciation, some 10,000 to 20,000 years 
ago, only minor subspecies have appeared. The differences are 
confined to body size, color, slight differences in body proportions, 
in the development of appendages, etc. The British race of the red 
deer, for instance, has evolved since Britain became separated from 
the Continent some 7,500 years ago. This is shown by fossil evidence, 
since the early postglacial specimens found in the Thames belong to 
the Continental race. Moreover the characters of Cervus elaphus 
scoticus, as the British race is called, are probably not fixed genetically, 
since, when the breed was transferred to a favorable environment, 
in New Zealand, it reverted in many respects to the Continental type. 
Other examples could be given of the insignificant character of 
Postglacial differentiation in species. ; 
Greater differences are observed in some, but by no means the 
majority, of Upper Pleistocene species, of about 50,000 to 100,000 
years ago. For the marmot of that time, for instance, Wehrli found 
that the shape of the temporal ridges has since become stabilized. 
