Re: Converting a number back to a symbol?

*To*: mathgroup at smc.vnet.net*Subject*: [mg82966] Re: Converting a number back to a symbol?*From*: AES <siegman at stanford.edu>*Date*: Mon, 5 Nov 2007 05:01:33 -0500 (EST)*Organization*: Stanford University*References*: <fgenct$fsd$1@smc.vnet.net>

In article <fgenct$fsd$1 at smc.vnet.net>, AES <siegman at stanford.edu> wrote: > If we have an expression containing integer powers of the symbol u0 and > we know that u0's numerical value is (or at least eventually will be) u0 > = Pi/2, we know what happens if we write > > expr /. {u0 -> Pi/2} > > But what about the reverse case . . . My abject apologies for this -- utter stupidity in action. My test case was (in separate cells): u0 = Pi/2; f = u0 + 2 u0^2 + 4 u0^2 f /. {Pi -> 2 u0}