The next example shows an improved prediction of relative diversity value for the same bee faunas, using a measure of tree length to count the relative expected number of gene or character differences within each fauna (the numbers on the map are percentage scores compared to the entire set and are rounded down; the richer areas are again shown in orange, the poorer areas in dark blue) (below):
This time, an Asian fauna (Kashmir, highlighted with the magnifying glass) scores most highly. The genealogical classification of species on the right shows the fauna of this cell, traced in red and black. The bee species in Kashmir are more distantly related from one another than are those in Ecuador and are consequently expected to be more different from one another.
