458 0. Meyer—Species in the Southern Old-tertiary. 
in my present collection, and future collecting may add many 
new points. 
In tracing out relationships between different species the fol- 
lowing considerations were guiding. 
Species, which disagree above a certain moderate extent In 
their embryonic parts, cannot be nearly allied, however similar 
they may appear. 
The differences between related species must be generally of 
a more quantitative than qualitative nature. 
While some of the qualities may differ, perhaps even consid: 
erably, most of them must be expected to remain unchanged. 
If two related species differ in certain points, it is to be ex- 
pected that they will vary somewhat in the same points in at 
Jeast one of the localities. 
small peculiarities in teeth, stris, pits, etc., usually not 
mentioned in descriptions, are sometimes very characteristic | 
of a species and remain constant, when other qualities vary. 
These little things sometimes give good indications of relation- 
hij 
