Off-lattice Monte Carlo simulations employing the pivot algorithm are used
to generate ideal and excluded volume linear polymers in three dimensions.
The structure function at large wavevectors is calculated from the resultin
g configurations. This is compared to the exact equation for ideal chains a
nd to experimental data and both scaling and renormalization group predicti
ons for excluded volume chains. It is found that using the des Cloizeaux fo
rm for the distance distribution function in an analytic calculation of the
structure function leads to dose agreement with the experimental and Monte
Carlo data.