We measured soil bacteria, fungi, protozoa, nematodes, and biological activ
ity in serpentine and sandstone annual grasslands after 4 years of exposure
to elevated atmospheric CO2. Measurements were made during the early part
of the season, when plants were in vegetative growth, and later in the seas
on, when plants were approaching their maximum biomass. In general, under a
mbient CO2, bacterial biomass, total protozoan numbers, and numbers of bact
ivorous nematodes were similar in the two grasslands. Active and total fung
al biomasses were higher on the more productive sandstone grassland compare
d to the serpentine. However, serpentine soils contained nearly twice the n
umber of fungivorous nematodes compared to the sandstone, perhaps explainin
g the lower standing crop of fungal biomass in the serpentine and suggestin
g higher rates of energy flow through the fungal-based soil food web. Furth
ermore, root biomass in the surface soils of these grasslands is comparable
, but the serpentine contains 6 times more phytophagous nematodes compared
to the sandstone, indicating greater below-ground grazing pressure on plant
s in stressful serpentine soils. Elevated CO2 increased the biomass of acti
ve fungi and the numbers of flagellates in both grasslands during the early
part of the season and increased the number of phytophagous nematodes in t
he serpentine. Elevated CO2 had no effect on the total numbers of bactivoro
us or fungivorous nematodes, but decreased the diversity of the nematode as
semblage in the serpentine at both sampling dates. Excepting this reduction
in nematode diversity, the effects of elevated CO2 disappeared later in th
e season as plants approached their maximum biomass. Elevated CO2 had no ef
fect on total and active bacterial biomass, total fungal biomass, or the to
tal numbers of amoebae and ciliates in either grassland during either sampl
ing period. However, soil metabolic activity was higher in the sandstone gr
assland in the early season under elevated CO2, and elevated CO2 altered th
e patterns of use of individual carbon substrates in both grasslands at thi
s time. Rates of substrate use were also significantly higher in the sandst
one, indicating increased bacterial metabolic activity. These changes in so
il microbiota are likely due to an increase in the flux of carbon from root
s to soil in elevated CO2, as has been previously reported for these grassl
ands. Results presented here suggest that some of the carbon distributed be
low ground in response to elevated CO2 affects the soil microbial food web,
but that these effects may be more pronounced during the early part of the
growing season.