Microbial methylation of selenium (Se) to volatile (CH3)(2)Se may cont
ribute to a considerable loss of the Se inventory from seleniferous so
ils over time. A field experiment was carried out to assess microbial
volatilization of Se as a bioremediation approach to dissipate Se. Fie
ld plots of size of 3.7 x 3.7 m(2) were established at the Kesterson R
eservoir, California. To enhance microbial activity, the plots were tr
eated with different C and protein amendments and were periodically ti
lled and irrigated. Over a period of 100 months, 68-88% of the total a
mount of Se dissipated from the topsoil (0-15 cm). The pattern of Se d
epletion in soil was not correlated with rainfall events nor with temp
erature. As most of the rainfall occurred during the cold winter month
s, it is likely that leaching dominated the Se depletion in winter whe
reas volatilization prevailed in summer. The highest amount of Se depl
etion occurred with the amendment of the protein casein. However, ther
e was no statistically significant difference in Se removal from the d
ifferent treatments compared to the case where there was no C or prote
in amendment. Tillage and irrigation presumably prevailed over the eff
ects of the amendments. A two-compartment model was superior to a one-
compartment model for describing the long-term kinetics of Se depletio
n in soil. The rate of Se dissipation was initially faster than at lat
er times. This indicates that the rate-limiting mechanisms have change
d during the time of the study. (C) 1997 Elsevier Science B.V.