(Book 23.5, 6)
Mixing layers:
Asuming that the layer is dominated by a balance of inviscid
instability (stronger for thinner layers) creating turbulence, and the
diffusive effects of turbulence (which thicken the layer), viscosity
would not play a big part. Then there is no obvious nondimensional
quantity that the nondimensional ratio 
could depend on,
giving 
linear in x. For the same reason, the
velocity profile will be similar.
Exercise:
Discuss how well that seems to agree with flow visualizations.
Jets:
must be
constant, so, assuming similarity, we get the decay rates for the
velocity.
2D Wakes:
For 2D wake flow, will vary with the nondimensional
parameter 
. The profiles will be similar since
the shape should tend to a finite limit far downstream.
Because of the momentum integral, 
remains constant
going downstream. Estimating the change in momentum 
as being caused by a Reynolds shear
force 
,(eg, a mixing length estimate with 
,) we get that
increases as 
and decreases as
going downstream.