³⁸Ar and ³He data from Antarctica and Bolivia show a distinct increase in the cosmogenic ³He to ³⁸Ar ratio (³He_c:³⁸Ar_c) with increasing sample elevation. Apatites sampled from 1.6 to 4 km suggest a sea level ratio of 2, with an increase to ~6 at 4 km elevation. This variability has three possible sources: 1) Significant production of cosmogenic ³⁶Cl_c and/or ³⁶Ar_c; 2) Erosion; or 3) An increase in ³He_c production with elevation relative to ³⁸Ar_c production. Calculation of ³⁸Ar_c includes the implicit assumption that ³⁶Ar produced from cosmogenic or other processes is minimal relative to ³⁶Ar originating from air. ³⁶Cl decays to ³⁶Ar and is produced through both cosmic ray neutron and muon processes; cosmogenic ³⁶Ar_c is also produced directly. Analysis of these data, however, indicates that excess ³⁶Ar cannot fully explain the observed trend. Second, the flux of cosmogenic muons increases with depth in rock. Erosion of an exposed surface would result in an apparent increase in ³⁸Ar_c production relative to ³He_c. Our data point to an opposite relationship, with the highest ³He_c:³⁸Ar_c occurring at the site with the most effective erosion. Finally, ³He_c:³⁸Ar_c can increase with elevation if muogenic production of ³⁸Ar is significant. Theoretical evaluation of direct and indirect ³⁸Ar_c production from muons on Ca suggests ~40 atoms g^-1 yr^-1. Model analysis of data thus far produces a best-fit where production from fast neutrons is ~12 atoms g^-1 yr^-1 and from negative muons (via capture or fast processes) is ~38 atoms g^-1 yr^-1, suggesting that noble gases could become central components of a quantitative paleoaltimetric technique.