Several circumstantial arguments point to the formation of the third r -process peak at A \sim 190 , near platinum , in stars of mass of \sim 8-10 M _ { \odot } : 1 ) The delayed production of europium with respect to iron imposes a time scale that restricts the progenitor stars to M \mathrel { \vbox { \hbox { $ < $ } \nointerlineskip \hbox { $ \sim$ } } } 10 M _ { \odot } ; 2 ) the r -process demands a dominant robust mechanism at least for Z \geq 56 , barium and above , since the relative abundance pattern of those r -process elements in the low-metallicity stars , [ Fe/H ] < -2 , is consistent with the solar pattern ; 3 ) stars of \sim 8-10 M _ { \odot } produce nearly identical degenerate O/Ne/Mg cores that collapse due to electron capture ; and 4 ) the resulting low-mass cores may produce both an r -process in a prompt explosion and a subsequent r -process in a neutrino driven wind .
A special case of the r -process singles out low entropies for initial Y _ { e } \sim \bar { Z } /A , where \bar { Z } is the mean atomic number of the seed nuclei and A is the atomic weight of the target .
For \bar { Z } \simeq 35 and Y _ { e } \simeq 0.18 the A \simeq 190 peak results in a natural way .
The prompt explosion of an O/Ne/Mg core yields low entropy , S \sim 15 , and low electron fraction , Y _ { e } \simeq 0.2 , and hence may produce a reasonable r -process peak at A \simeq 190 , as well as all of the r -process elements with Z \geq 56 .
The possible differences in the \nu -driven wind and associated r -process due to the low-mass neutron stars expected in this mass range are also discussed .