When is an island an island?

[james_davis_nicoll]

Useful factoid: If recall Wilson correctly, if you assume all other factors stay the same, the number of species a given region per unit area can support is proportional to the square root of the total area. If, for example, you have two islands, one 10 units in area and the other 1000 units, the second one will usually have about 10x as many species per unit area and presumably about 1000x as many species in total. 100 small islands might have as much area as one big one but they will only a tenth as many species.

Does anyone here know what conditions are necessary to disconnect adjacent regions? Obviously water will do it, esp deep water. Roads and wide clear cuts can, since they make migration more difficult.

Obviously, this rule of thumb implies that a few large preserves are probably a more effective tool for species preservation than a lot of little ones.

It also implies large cities should have more species/area in them than small ones. I wonder if that is the case or if other factors (like the relative youth of large cities as common habitats) dominate?

It also means that in the Reefs of Space setting, bigger rocks will tend to be more various places to live (in terms of life) than little rocks. OTOH, delicate lifeforms might find small islands nice places to live, thanks to the lack of various competators and predators.

Comments

[(Anonymous)]

There's a scaling law as regards city size and number of industry and service types (I forget what the exponents are) but there's no good theory behind the empirical relationship. People have made stabs at it -- including Krugman, when he isn't writing New York Times editorials -- but it's a little intractable.

If memory serves, it's related to calculating the optimal number of firms in an industry to maximize welfare. Also a little intractable.

Carlos