eLearning

changes in the 0–10-cm depth represented shifts
among soil horizons rather than novel inputs to the
soil; perhaps N from unmeasured aboveground lit-
ter contributed to the 0–10-cm horizon under the
conifers, but mixing by earthworms under the
hardwoods mixed the litterfall N to great depths
under hardwoods. This study provides a very in-
triguing possibility of differential occult N input
among tree species. However, the statistical signif-
icance of the species effect on the N content of the
0–10-cm horizon is tempered by the lack of sam-
pling of the rest of the profile.We conclude that this
study warrants moderate confidence as a case of
occult N input.
In southeastern Sweden, Eriksson and Rosen
(1994) estimated the total N content of five species
in monoculture plots (n 5 3 plots/species). After 35
years, the total N content under silver fir (Abies alba
Mill.) averaged 1700 kg N/ha higher than under
Japanese larch. If the difference in forest N content
were real, the rate of divergence would be approx-
imately 50 kg N ha-1
y-1
. However, the only signif-
icant differences among the species was in the N
content of vegetation and O-horizons; the majority
of the N in these systems was found in the mineral
soil, and no differences were significant. We con-
clude that this study provides only weak evidence
of possible effects of tree species on forest N con-
tent, and any further interpretation in support of
occult N inputs would not be warranted.
CHRONOSEQUENCE STUDIES
Dickson and Crocker (1953) analyzed soils of a
several-thousand-year chronosequence of mud-
flows at Mt. Shasta, California, USA. The forest
floor was variable (and discontinuous in some cas-
es), so the investigators sampled only the “zones of
maximum accumulation under Pinus ponderosa”
and claimed that their samples “bear no relation-
ship to areal extent over the mudflows as a whole.”
Furthermore, they chose to sample only one loca-
tion within each mudflow, and had no replicates of
mudflows from the same eruption. Despite these
sampling limitations, the authors calculated a rate
of N accretion of 3760 kg N/ha in a 60-year-old
mudflow (mudflow B), for an N accretion rate of 63
kgNha-1
y-1
. However, later work indicated that
the age of this mudflow was closer to 300 years
than 60 years (Glauser 1967), which would give an
accretion rate of approximately 12 kg N ha-1
y-1
.
Glauser also estimated N accretion in a younger
flow of known age and calculated an N accretion
rate of 14 kg N ha-1
y-1
. Sollins and others (1983)
again sampled the chronosequence of mudflows