Full text of DOI:10.1300/J091v14n01_03

Journal of Sustainable Forestry
ISSN: 1054-9811 (Print) 1540-756X (Online) Journal homepage: http://www.tandfonline.com/loi/wjsf20
Post-Clearcutting Chronosequence in the B.C.
Coastal Western Hemlock Zone
J. P. (Hamish) Kimmins , Wayne L. Martin & Robert L. Bradley
To cite this article: J. P. (Hamish) Kimmins , Wayne L. Martin & Robert L. Bradley (2001)
Post-Clearcutting Chronosequence in the B.C. Coastal Western Hemlock Zone, Journal of
Sustainable Forestry, 14:1, 45-68, DOI: 10.1300/J091v14n01_03
To link to this article: http://dx.doi.org/10.1300/J091v14n01_03
Published online: 12 Oct 2008.
Submit your article to this journal
Article views: 18
View related articles
Citing articles: 8 View citing articles
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=wjsf20
Download by: [George Mason University]
Date: 05 June 2016, At: 20:49

Post-Clearcutting Chronosequence
in the B.C. Coastal Western Hemlock Zone:
III. Sinks for Mineralised
or Dissolved Organic N
J. P. (Hamish) Kimmins
Wayne L. Martin
Robert L. Bradley
ABSTRACT. We estimated the magnitude of various sinks for forest
floor N released by decomposition in a 400-year-old stand and along a
26-year post-clearcutting chronosequence in a Pacific silver fir (Abies
amabilis [Dougl.] Forbes) and western hemlock (Tsuga heterophylla
[Raf.] Sarg.) ecosystem on Vancouver Island. The total biomass and
N-content of various components of the overstory and understory vege-
tation were measured, as well as the upper mineral-soil N-storage, and
the fluxes of dissolved forms of N in forest floor and mineral-soil
leachates. The overstory vegetation was not an important sink for N
until at least 10 years after clearcutting; uptake by trees was not a
significant sink for the post-clearcutting assart flush. In contrast, the
early-seral understory vegetation played a small but probably signifi-
cant role during the assart period. Competition for N between the un-
derstory and overstory vegetation during the first 10 years was prob-
ably not significant because the amount of N released from the forest
floor far exceeded the amount of N taken up by either of these two
J. P. (Hamish) Kimmins is Professor of Forest Ecology at the Department of
Forest Sciences, University of British Columbia, 2424 Main Mall, Vancouver, B.C.,
Canada, V6T 1Z4.
Wayne L. Martin is Regional Staff Manager, B.C. Ministry of Forests, Prince
George Forest Region, Bag 3500, Burns Lake, B.C., Canada, V0J 1E0.
Robert L. Bradley is Assistant Professor of Soil Science at the Département de
Biologie, Université de Sherbrooke, Sherbrooke, Québec, Canada, J1K 2R1.
Address correspondence to: Robert L. Bradley at the above address (E-mail:
robert.bradley@courrier.usherb.ca).
Journal of Sustainable Forestry, Vol. 14(1) 2002
E 2002 by The Haworth Press, Inc. All rights reserved.
45

JOURNAL OF SUSTAINABLE FORESTRY
46
components. Total leaching losses of dissolved-N between 3 and 12
years after clearcutting did not appear to vary significantly from total
leaching losses in the old forest. Based on the lysimeter data, the upper
mineral soil appeared to have the biological and/or chemical capacity to
prevent most of the N released from the forest floor after clearcutting
from moving down to below the rooting zone. However, we do not
conclude from this that the majority of the N released is retained within
the ecosystem. In a subsequent paper we develop a forest floor N-bud-
get model to evaluate whether or not this is the case. [Article copies
available for a fee from The Haworth Document Delivery Service:
1-800-342-9678. E-mail address: <getinfo@haworthpressinc.com> Website:
<http://www.HaworthPress.com> E 2002 by The Haworth Press, Inc. All rights
reserved.]
KEYWORDS. N budget, forest floor, chronosequence, clearcutting
INTRODUCTION
One of the biggest challenges facing the forestry profession today is
to develop accurate predictions of future timber and non-timber forest
values, and ecosystem conditions and functions, under alternative man-
agement systems. Silvicultural systems of all types must be evaluated
in terms of their potential to achieve society’s long-term goals of
sustainable ecosystem management across the high level of ecological
diversity found in most forest landscapes. In the absence of experience
of these systems at appropriate temporal and spatial scales, knowl-
edge-based decision-support tools in the form of ecosystem manage-
ment simulation models can be very useful.
The debate over clearcutting is not exclusively a scientific issue, but
it should be soundly based on the scientific evidence that supports or
contradicts claims that this silvicultural option can meet society’s long-
term goals. In coastal British Columbia, where some of the world’s
most productive stands of commercially valuable softwoods grow,
there is a paucity of empirical evidence on this question. Empirical
yield tables over multiple rotations are not available due to the re-
gion’s recent logging history, and we lack experience of many of the
alternative silvicultural systems that are being advocated. Adaptive
management (AM) of forests in British Columbia has been proposed
as a response to this lack of an empirical foundation for the design of
changes in management (Taylor et al., 1997). However, in spite of the

Kimmins, Martin, and Bradley
47
unquestionable value of AM as a tool of sustainable forest manage-
ment, it can be argued that this option is less useful for long-lived
forests than for shorter-term crops such as fisheries and agriculture.
AM is most useful where the consequences of management actions are
readily observed after only a few years, such as in agriculture, fish-
eries, or short rotation forestry. For most forests, it is many decades
before the long-term consequences of management choices become
apparent. As a consequence, foresters have started to use computer
models to simulate the effects of ecosystem management on stand
productivity and other forest-based values over multiple rotations as a
complement to, or an integral part of, adaptive management (e.g.,
Aber et al., 1982; Kimmins and Scoullar, 1984; Morris et al., 1997).
A simulation model’s ability to predict future forest growth and
harvestable values is determined by (1) our understanding of the pro-
cesses controlling forest growth dynamics, (2) the representation of
these processes in the model, and (3) the quality of the data used to
calibrate the model. There is a paucity of data describing the fate of
forest floor N following clearcutting in the coastal western hemlock
forests of British Columbia. This has been an impediment to model
formulation, calibration and validation, and identifies the need to doc-
ument N dynamics in these forests. Martin et al. (this issue) showed
that clearcutting on unslashburned sites on southern Vancouver Island
resulted in a significant temporary decrease in the mass and N content
of the forest floor. Bradley et al. (this issue) showed an increase in
mineral-N availability, the so-called ‘‘assart flush,’’ and an increase in
potential leaching losses of soil-N following the complete removal of
the forest canopy on these sites.
The degree to which the assart flush benefits the crop vs. the non-
crop vegetation, and the amount of N which is lost through leaching or
other pathways, is key information with which to improve the repre-
sentation of N-cycling and stand productivity during the stand-estab-
lishment phase in ecosystem management models. The overall objec-
tive of the study reported here was to estimate the magnitude of
various possible sinks for forest floor N along the 26-year post-clear-
cutting chronosequence in the Pacific silver fir (Abies amabilis [Dougl.]
Forbes) and western hemlock (Tsuga heterophylla [Raf.] Sarg.) eco-
systems on Vancouver Island investigated by Martin et al. and Bradley
et al. (this issue). We measured total biomass and N-content of various
components of the overstory and understory vegetation, mineral-soil

JOURNAL OF SUSTAINABLE FORESTRY
48
N-storage, and the fluxes of dissolved forms of N in forest floor and
mineral-soil leachates in a 400-year-old forest and four clearcuts of
varying age in the same ecosystem type. Motivation for the study was
the desire to develop credible simulation models for use by foresters as
decision-support tools with which to evaluate ecosystem resiliency
and sustainability (Kimmins 1993).
METHODS
Study sites and chronosequence design--A detailed description of
the study area and of the experimental design was given in a previous
paper (Martin et al., this issue). Briefly, the study was conducted on
southwestern Vancouver Island, Canada, about 15 km southwest of the
village of Lake Cowichan, in the Estevan and West Vancouver Island
Submontane Wetter Maritime Coastal Western Hemlock (CWHb1,2)
biogeoclimatic zone (Pojar et al., 1991). The mean annual temperature
is 9.2_C and the mean annual precipitation is 3016 mm, of which 24%
falls between April and September.
Five intensively studied mid-elevation (550-to-600 m) sites were
selected on the basis of their similarities in pre-disturbance, distur-
bance, and post-disturbance conditions. The five sites represented a
post-clearcutting chronosequence (3-, 6-, 10- and 26-year-old clear-
cuts in 1981) and a 400-year-old unevenaged climax forest of Pacific
silver fir and western hemlock. None of the clearcuts had been slash-
burned, and consequently carried advanced regeneration of Pacific
silver fir and western hemlock. The dominant early seral species was
fireweed (Epilobium angustifolium L.). According to pre-logging re-
cords and forest cover maps, the four clearcuts had all carried a rela-
tively uniform climax forest very similar to that of the ‘‘old-growth’’
site.1
The soils of the sites are all humo-ferric podzols (CSSC, 1978) with
a forest-floor and mineral-soil pH (10ꢀ M CaCl2) of 3.2-to-3.7 and
2
4.2-to-4.7, respectively. Alaska blueberry (Vaccinium alaskaense Ho-
well), salal (Gaultheria shallon Pursh.) and lanky moss (Rhytidiadel-
phus loreus [Hedw.] Warnst.) dominated the understory in the ‘‘old-
1. The old growth index (OGI) of these old forests was not determined, so we
cannot say with confidence the degree to which they satisfied the definition of
‘‘old-growth’’ for these forests.

Kimmins, Martin, and Bradley
49
growth’’ stand, suggesting a submesotrophic (sensu Krajina, 1969)
nutrient regime. The ‘‘old-growth’’ stand appeared to have been free
of fire for at least 400 years during which time an average of 26 cm of
forest floor, containing about 2500 kg N haꢀ1, had accumulated. The
humus form in the uncut forest is mor and lignomor (Klinka et al.
1981).
Three study plots, ranging from 625-to-2500 m2 in size, were estab-
lished in each of the five unreplicated sites. The three plots within each
age-class were nevertheless treated statistically as true replicates since
they were located several hundred metres apart from each other,
whereas the spatial dependence of forest floor depth, C and N content
in Coastal Western Hemlock forests does not exceed 10 m (Qian and
Klinka, 1995). The specific objectives of the study were to estimate
(1) the amount of N accumulating in the overstory and understory
biomass, (2) the leaching losses of N, and (3), the storage of N in the
upper mineral soil horizon along the 26-year post-clearcutting chrono-
sequence. In a parallel study on the same sites, Martin et al. (this issue)
reported a net loss of 940 kg N haꢀ1 from the forest floor during the
first six years after clearcutting. An underlying assumption in the
present study was that the N taken up by the regenerating vegetation
originated largely from the N-pool released from the forest floor after
clearcutting. We think that this assumption may be valid for the three
younger clearcut sites (Fahey and Hughes, 1994), but it becomes
tenuous for the 26-year-old site because of the longer period of
geochemical inputs.
Overstory accumulation--The diameter of all trees found within
four randomly-located 10 m dia. circular quadrats in each of the 15
plots were tallied in mid-summer, 1981. All trees greater than 2.5 cm
dia. at the base of the stem (dbs) were measured in the 3-year-old,
6-year-old and 10-year-old clearcuts, and all trees greater than 2.5 cm
dia. at breast height (dbh) were measured in the 26-year-old clearcut
and the ‘‘old-growth’’ stand. In these latter two stands, smaller indi-
viduals were inventoried as part of the understory vegetation.
Seven amabilis fir and seven western hemlock trees, ranging in size
from 1.0 cm dbs to 22 cm dbs were harvested from the four clearcut
age-classes. The total mass of foliage, branches and stems of the
sampled trees was measured in the field. Stem disks and subsamples
of the foliage and branches were taken back to the laboratory and
weighed fresh and after drying to a constant weight at 70_C. Double-

JOURNAL OF SUSTAINABLE FORESTRY
50
log regression equations of the aboveground biomass of each tree
component as a function of stem diameter were computed (Table 1).
Biomass components for trees sampled in the ‘‘old-growth’’ stands
were predicted from dbh using the regression equations of Gholtz et
al. (1979).
A portion of the ‘‘old-growth’’ stand was clearcut in the summer of
1983, which allowed us to collect foliar samples throughout the
crowns of freshly harvested, mature amabilis fir, western redcedar and
western hemlock trees. Foliage from the ‘‘old-growth’’ stand as well
as from the clearcut sites were ground in a Wiley mill, subsamples
(0.100 g) were wet-digested in a mixture of sulfuric acid and hydrogen
peroxide, and the digests were analyzed colorimetrically for Kjeldahl-N
using a Technicon Autoanalyzer. The N concentrations of the woody
components were assumed to be the same as those given by Krumlik
(1979). The total aboveground accumulation of N per unit area in the
overstory vegetation was calculated for each tree component in each
plot using the measured and estimated values of N concentrations and
biomass.
Above-ground understory accumulation--Understory biomass was
sampled in each plot from three randomly located 1 m2 quadrats in
mid-summer, 1981 (9 quadrats per study site). The aboveground bio-
mass was clipped and sorted into six categories: (1) mosses, (2) live
fireweed (3) standing dead fireweed, (4) Vaccinium spp. (mainly V.
alaskaense), (5) salal, and (6) other plants of minor abundance includ-
ing tree seedlings. These were oven-dried (70_C for 48 h), weighed
(á 1 g) and ground in a Wiley mill. Subsamples were analyzed for
Kjeldahl-N and the total aboveground accumulation of N in the six
categories of understory species was calculated for each quadrat.
Accumulation in roots--Belowground biomass of both overstory and
understory vegetation was estimated based on root-shoot ratios and
root biomass values of similar-aged stands on Vancouver Island (Fell-
er et al., 1983) and in the Washington State Cascade mountain range
(Grier et al., 1981). Root-N concentrations were obtained from root
samples excavated during the inventory of the forest floor (Martin et
al., this issue). These were sorted by species, dried, weighed, digested
and analyzed for total-N concentration as described above. An esti-
mate of the total belowground accumulation of N was calculated for
each category of plant.
Leaching of DON and DIN--The flux of dissolved organic (DON)

Kimmins, Martin, and Bradley
51
TABLE 1. Regression equations used to estimate aboveground biomass1 com-
ponents (Y) for Tree species as a function of stem diameter2 (X) for tree species
in the sample; all equations are of the form ln(Y) = m*ln(X) + b.
2
Slope (m )
Intercept (b)
R
A. Clearcuts
(3 to 26 year-old)
Amabilis fir
(1.4-22.0 cm dia.)
Foliage
2.20
2.48
2.97
3.50
2.48
2.23
0.98 (n = 7)
0.99 (n = 7)
0.99 (n = 7)
Branches
3
Stemwood
Western hemlock
(1.0-21.0 cm dia.)
Foliage
2.27
2.49
2.67
3.15
2.43
2.83
0.99 (n = 7)
0.99 (n = 7)
0.98 (n = 7)
Branches
3
Stemwood
B. Old growth stand
(from Gholtz et al., 1979)
Amabilis fir
Foliage
2.19
2.63
2.57
2.84
ꢀ4.55
ꢀ5.24
ꢀ3.51
ꢀ6.12
Branches
Stemwood
Bark
Western red cedar
Foliage
1.78
2.09
2.19
2.11
ꢀ2.62
ꢀ3.27
ꢀ2.09
ꢀ4.19
Branches
Stemwood
Bark
Western hemlock
Foliage
2.13
2.79
2.26
2.26
1.31
ꢀ4.13
ꢀ5.15
ꢀ2.17
ꢀ4.37
ꢀ2.41
Branches
Stemwood
Bark
Dead branches
1
2
3
Biomass units are grams for 3- to 26-year-old clearcuts, and kilograms for the old-growth stand.
Refers to basal stem diameter in the 3- to 10-year-old clearcuts, and to dbh in the older stands.
Bark component was not separated from stemwood, in the clearcut sites.

JOURNAL OF SUSTAINABLE FORESTRY
52
and inorganic N (DIN) through the soil profile was estimated using
nutrient concentrations measured in lysimeters, from the fall of 1981
to the fall of 1982 (Bradley et al., this issue). These were placed
beneath the forest floor (three tensionless lysimeters per plot) and at a
depth of 60 cm in the mineral soil (three tension lysimeters per plot).
The resulting data refer to sites of 3-5, 6-8, 10-12 and 26-28 years post
clearcutting. A water balance equation was used to estimate net soil
drainage:
D = P ꢀ R ꢀ áS ꢀ Ea
where D = drainage (mm dayꢀ
)
1
P = precipitation (mm dayꢀ
)
1
R = surface runoff (mm dayꢀ
)
1
Ea = actual evapotranspiration (mm dayꢀ
)
1
áS = changes in soil moisture storage between sampling dates
(mm dayꢀ1).
Assumptions used in applying this model were as follows:
1. surface runoff equals zero (i.e., R = 0)
2. no change in soil moisture storage (i.e., áS = 0) between Sep-
tember and April
3. soil moisture storage can be estimated from a soil moisture reten-
tion curve (Hillel, 1980) adjusted for coarse fragment content
4. actual evapotranspiration (Ea) can be calculated by comparing
precipitation inputs, soil storage and potential evapotranspiration
(Ep)
5. potential evapotranspiration (Ep) can be calculated from the Pre-
istley-Taylor equation (Ep = (a*(Ta + b)*St)/2430 J gꢀ1) where
constants a and b equal 0.025_C and 3_C, respectively (Camp-
bell, 1977), Ta is air temperature and St is solar radiation
6. air temperature and solar radiation data taken from Nanaimo,
British Columbia (Environment Canada, 1982), situated approx-
imately 50 km to the north-east of the study area were assumed
to be adequate for calculating Ep. Differences in climate between
Nanaimo and the Fleet River study site render this assumption
somewhat tenuous, but these were the best available data at the
time of the study, and errors in the assumption should not invali-
date comparisons between the different chronosequence sites.

Kimmins, Martin, and Bradley
53
Upper mineral soil N storage--Mineral soil was excavated, in the
summer of 1982, to a depth of 30 cm at 25 randomly selected locations
in each plot, except in the 26-year-old clearcut. A polyethylene bag
was inserted to line the contour of each pit, and the volume of water
needed to fill each pit was determined using a graduated cylinder.
Coarse fragments were removed, their volume measured in the field
by water displacement, and converted to a mass basis assuming a
specific density of 2.65 g cmꢀ (Hillel, 1980). The remaining soil
3
samples were taken to the laboratory, sieved and separated into fine (<
2 mm) and coarse (> 2 mm) fractions, and both fractions were
weighed. Subsamples of each size fraction were dried at 105_C for
48 h and re-weighed to determine moisture content. The fine fraction
of the 25 samples from each plot were randomly bulked into 8 larger
samples, and these were wet digested and analyzed for total Kjeldahl-
N (TKN).
RESULTS
Accumulation in trees--Prior to clearcutting, the total biomass and N
content of the trees in theꢀ‘‘old-growth’’ stand were approximately 660
t haꢀ and 862 kg N ha
, respectively (Table 2). Roots accounted
1
1
for over 25% of the total-N; 38% of total root N was in the fine root
compartment. Foliage accounted for only 4.5% (29.9 t haꢀ1) of the
total tree biomass but for 27% (232.9 kg N haꢀ1) of total tree N.
Stemwood and bark, the main components removed from the site by
harvesting, contained approximately 121 and 115 kg N haꢀ1, respec-
tively; together they accounted for 27% of total tree N.
The total biomass and N content of the advance regeneration and
new tree seedlings onꢀthe 3-year-old clearcut were very low (1.4 t
haꢀ and 8.0 kg N ha 1, respectively) compared to the trees in the
1
‘‘old-growth’’ stand (Table 2). The foliage component accounted for
65%, and roots accounted for 9%, of the total-N in the conifer seed-
lings at this site. Fine roots accounted for 80% of root-N.
There were no significant changes in the total biomass and N con-
tent of tree seedlings between the 3-year-old and the 6-year-old clear-
cut sites (Table 2), but both of these variables increased 7-to-8 fold,

54

Kimmins, Martin, and Bradley
55
between the sixth and tenth years after clearcutting. However, total-N
in the trees on the 10-year-old clearcut was still only 6% of total-N in
the tree component of the ‘‘old-growth’’ stand. The foliage component
accounted for 64% and roots accounted for 11% of total-N in the
overstory on the 10-year-old clearcut.
The highest rate of increase in tree growth was observed between
10 and 26 years after clearcutting. The total biomass and N content of
the overstory in the 26-year-old clearcut were approximately 36% and
73% of those values in the ‘‘old-growth’’ stand. The foliage compo-
nent accounted for 56% and roots accounted for 19% of total-N in the
overstory on the 26-year-old clearcut. Fine roots accounted for 64% of
root-N.
Understory accumulation--The biomass and total-N content of un-
derstory vegetation were only slightly lower in the 3-year-old clearcut
ꢀ
(0.6 t haꢀ and 7.7ꢀkg ha 1) than in the ‘‘old-growth’’ stand (1.0 t
1
haꢀ and 9.1 kg ha 1) (Table 3). Both of these variables increased
1
approximately 4-to-5 fold between 3- and 6-year-old clearcuts, and
increased by another 60% and 25%, respectively, between 6 and 10
years after clearcutting. The total biomass and N content of the under-
story vegetation on the 26-year-old clearcut was slightly higher (5.2 t
haꢀ and 51.0 kg N haꢀ1) than on the 10-year-old clearcut (4.3 t
1
haꢀ1 and 47.8 kg haꢀ1). The N-content of belowground components
contained 16% to 23% of total N-content of the understory vegetation,
except on the 26-year-old site where belowground components con-
tained 37% of total-N.
The dominant understory sinks for N in the ‘‘old-growth’’ stand
were salal (2.70 kg N haꢀ1), Vaccinium sp. (2.46 kg N haꢀ1) and
TABLE 3. Total biomass and N content of the understory vegetation in chrono-
sequence stands; numbers in parentheses = 1 S.E. (n = 3).
ꢀ
1
N content (kg ha
)
ꢀ
1
Site
Biomass (t ha
1.0 (0.2)
)
Aboveground
7.5 (1.8)
Belowground
Total
9.1
Old-growth
3-year-old
6-year-old
10-year-old
26-year-old
1.6
1.8
0.6 (0.1)
5.9 (1.4)
7.7
2.7 (0.1)
32.1 (1.3)
36.8 (3.1)
32.2 (7.2)
6.3
38.4
47.8
51.0
4.3 (0.4)
11.0
18.8
5.2 (1.2)

JOURNAL OF SUSTAINABLE FORESTRY
56
mosses (2.30 kg N haꢀ1) (Figure 1--note the log scale). Three years
after clearcutting, these three categories of understory plants de-
creased markedly, whereas fireweed (2.01 kg N haꢀ1) emerged as the
dominant sink for N. Fireweed continued to be the major sink for N
among understory plants in the 6-year-old (15.91 kg N haꢀ1) and
10-year-old clearcuts (12.48 kg N haꢀ1). Dead fireweed was a rela-
tively large pool of understory-N on the 10-year-old clearcut (4.87 kg
N haꢀ1). Salal was a relatively major additional understory sink for N
FIGURE 1. Total N content (kg haꢀ1) of overstory trees and of important
understory species in the old-growth stand and in the chronosequence of
clearcut sites.
Overstory
Mosses
Live fireweed
Dead fireweed
Salal
Vaccinii
Others
3
2
1
0
0
5
10
15
20
25
Age of clearcut

Kimmins, Martin, and Bradley
57
in the 6-year-old (7.05 kg N haꢀ1) and 10-year-old clearcuts (6.57 kg
N haꢀ1), but its biomass was markedly lower in the 26-year-old
clearcut (0.11 kg N haꢀ1). The various Vaccinium species comprised a
large understory sink for N starting in the 10-yꢀear-old clearcut (10.68
kg N haꢀ1) and increasing to (30.94 kg N ha 1) in the 26-year-old
clearcut. No other understory plant species were measured in the
26-year-old clearcut site.
Leaching of DON and DIN--The estimated solution fluxes of DIN
(i.e., NH4+ + NO3ꢀ) from the forest floor were less than 3 kg N haꢀ
1
yearꢀ1 in each clearcut age-class and in the ‘‘old-growth’’ stand, with
the lowest values occuring in the 11- and 27-year-old clearcuts (Fig-
+
ure 2a). The concentration of NH4 in the forest floor leachates was
consistently 2-to-4 times higher than that of NO3ꢀ. The amount of
DON leaching annually from the forest floor was about 5-to-10 times
higher than that of DIN (Figure 2b). The highest estimated rate of
ꢀ
total-N leaching loss from the forest floor (21.2 kg N haꢀ year
)
1
1
occurred in the 4-year-old clearcut.
Only small amounts (< 1 kg N haꢀ1 yearꢀ1) of DIN were collected
in the mineral soil lysimeters, except in the 4-year-old clearcut site
where approximately 3 kg of both NHꢀ4 -N and NO3ꢀ-N were col-
+
+
lected (Figure 2c). The NH4 -to-NO3 ratio was generally much
closer to unity in the mineral soil percolate than in the forest floor
leachate. As was the case for the forest floor, the amount of DON
percolating through the mineral soil was about 5-to-10 times higher
than that of DIN (Figure 2d). At each site, the amounts of DON and
total-N percolating through the mineral soil were comparable to the
amounts leaching from the forest floor and into the mineral soil, sug-
gesting little net mineral soil storage of N.
Upper mineral soil N storage--The measured levels of total N in the
top 30 cm of the mineral soil on the 4-year-old clearcut was more than
double that on the old-growth site (i.e., 2000 kg N haꢀ1 vs. 921 kg N
haꢀ1) (Figure 3). N-storage in the upper mineral sꢀoil was 1604 kg N
haꢀ1 on the 7-year-old clearcut, and 1284 kg N ha 1 on the 11-year-
old site.
DISCUSSION
Trees as a post-clearcutting sink for N--The lack of change in bio-
mass and N content of young trees between the third and sixth years

y r a h N k g
y r a h N k g
1
ꢀ
1
1
1
ꢀ
ꢀ
ꢀ
y r a h N k g
y r a h N k g
1
ꢀ
1
1
1
ꢀ
ꢀ
ꢀ
58

Kimmins, Martin, and Bradley
59
FIGURE 3. Total N storage in the top 30 cm of mineral soil in the old-growth
stand and in each of the three youngest clearcuts; error bars indicate 1 SE (n= 3).
2000
1500
1000
500
0
Old-growth
4-yr-old
7-yr-old
11-yr-old
Age of clearcut
after clearcutting reflects a period of physiological and morphological
adjustment of the previously heavily shaded advance regeneration,
and some mortality of these seedlings. There was also recruitment of
new seedlings by natural regeneration during this period. The presence
of these smaller trees resulted in the mean seedling height remaining
unchanged between the 3- and 6-year-old sites (Martin et al., 2002) in
spite of accelerating height growth of the advance regeneration (Brad-
ley et al. this issue). As a result there was much greater variation in
seedling height on the 6-year-old clearcut than on the 3-year-old clear-
cut. Conifer regeneration dynamics during the 3- to 6-year-old period
were thus characterized by a combination of vigorous growth for some
of the established advance regeneration seedlings, slow growth or
mortality for others, and recruitment of new seedlings. Death of a
significant proportion of regenerating conifer seedlings can occur on
cutovers in a number of ways, including direct heat damage, larval
infections, wildlife grazing and increased competition from early seral
vegetation (Graham and Kingery, 1990; Grehan et al., 1992; McInnis
and Roberts, 1995; Kronzucker et al., 1997). We conclude that, al-
though surviving conifer seedlings showed vigorous growth during
the ‘‘assart flush’’ (Bradley et al., this issue), total-N uptake by the
overstory vegetation during the 3 to 6 year post-clearcutting period
was insignificant.
Trees became a much stronger sink for N during the post-assart

JOURNAL OF SUSTAINABLE FORESTRY
60
period, between 10 and 26 years after clearcutting. A twelve-fold
increase in total-N content during this period, from 6% to 73% of
total-N sequestered in the overstory vegetation in the ‘‘old-growth’’
stand, coincided with the attainment of the full age class complement
by Pacific silver fir seedlings (i.e., needle longevity = 11 years) and
the onset of canopy closure. The latter would have resulted in reduced
competition from early seral species, such as fireweed.
The foliar-N:total-N ratio decreased with stand-age as N storage in
woody tissues increased during stand development. As the regenerat-
ing stands grew older, the root-N:total-N ratio increased, and by the
‘‘old-growth’’ stage, the root compartment of this type of forest is an
important storage organ for plant-N.
Understory vegetation as an N sink--The understory vegetation in
the ‘‘old-growth’’ stand contained an amount of N similar to that in the
3-year-old clearcut, although the species composition in the two
stands was different. Therefore, the understory vegetation was not a
net sink for forest floor N during the first three years after clearcutting.
On the 6-year-old site, the N-content of the understory vegetation was
five times greater than the N-content of the young trees, and five times
greater than the N-content of the understory vegetation on the 3-year-
old site. Live fireweed comprised the largest understory component
during this 3- to 6-year post-disturbance ‘‘assart period,’’ accounting
for over 40% of the total understory-N. The high rate of N-uptake by
the understory vegetation compared to that by the young trees between
3- and 6-years after clearcutting raises an important question about
interspecific nutrient competition: does soil-N supply exceed total
plant-N uptake during this period, or does the understory vegetation
compete with the overstory vegetation for soil-N? The question has
been raised in the discussion over the ecological and silvicultural
consequences of clearcutting. Kronzucker et al. (1997) suggested that
early-seral herbacious species (e.g., fireweed) are detrimental to re-
generating conifers on clearcut sites because of their competitive ad-
vantage in acquiring soil-NO3ꢀ and because clearcutting promotes ni-
trification. However, evaluation of competitive aspects of early seral
vegetation must be balanced by a consideration of their role in ‘‘cap-
turing’’ the assart flush of nutrients more efficiently than slower grow-
ing conifers. This characteristic may minimize the loss of soil-N from
the ecosystem and thereby help to sustain long-term site productivity
(e.g., Marks 1974).

Kimmins, Martin, and Bradley
61
The mean total-N content of the forest floor declined by 940 kg N
haꢀ1 in the first six years after clearcutting (Martin et al., this issue);
that is, twenty times more than the combined amounts of N seques-
tered in the understory and overstory vegetation during the same
6-year period. Given the apparently high rates of nutrient release from
the forest floor during the assart flush, the N demands of fireweed and
other fast-growing early-seral species was small compared to that
which was potentially available (but many times greater than the de-
mands of the young conifers). It would therefore appear that the
amount of N they sequestered was of little significance for the supply
of N to the regenerating trees. The standing dead fireweed on the
10-year-old clearcut represented 10% the N contained in the young
trees on this site. The subsequent incorporation of this fireweed litter
into the forest floor constitutes a ‘‘green manure’’ for the regenerating
overstory trees after the tenth post-clearcutting year. Similarly, the
various Vaccinium species on the 26-year-old clearcut repreꢀsented a
net capture of about 51 kg N haꢀ1, that is, about 42 kg N ha 1 more
than the amount of N contained in the understory on the old-growth
site. Thus, between 26-years after clearcutting and the ‘‘old-growth’’
stage, the reduction in understory vegetation can return an amount of
N to the forest floor that is approximately equal to 18% of net N-up-
take by the overstory vegetation during the same period of time.
Leaching losses--A key characteristic of ‘‘old-growth’’ forest soils
is the predominance of DON in soil solution. Some have proposed
that, because DON may be less mobile than DIN, the development of
forest soils over time from DIN-rich to a DON-rich soil solution is a
mechanism by which forest ecosystems maintain a biogeochemically
tight N cycle and minimize N-loss following disturbance (Howard and
Howard, 1994; Northup et al., 1995). In our study, the soil solution
flux of DON was an order of magnitude greater than the soil solution
flux of DIN, and the total-N flux would probably have been greater
had the DIN:DON ratio of soil solution been higher, The higher fluxes
of DON from the forest floor, and of DIN and DON from the mineral
soil, on the 4-year-old clearcut are interpreted to reflect the assart
release of N from the forest floor, the lower evapotranspiration rates
and higher soil moisture content on this site (Martin et al., this issue),
and the delay in growth response of advance regeneration and early
seral vegetation. Both DIN and DON fluxes in the initially 6-year-old
and 10-year-old clearcuts over the subsequent two years of measure-

JOURNAL OF SUSTAINABLE FORESTRY
62
ment were lower than in the ‘‘old-growth’’ stand. N uptake by the
regenerating vegetation undoubtedly contributed to this, as did the end
of the assart period.
The chemical speciation of DIN duriꢀng the assart flush is a topic of
+
interest. The higher mobility of NO3 tꢀhan NH4 in soil solution
means that lower concentrations of NO3 are required to achieve
equal leaching rates of the two ions. The predominant ionic form of
+
DIN in our forest floor and mineral soil leachates was NH4 . In the
Hubbard Brook study, Likens et al. (1970) concluded that clearfelling
a northeastern deciduous hardwood forest without the removal of the
felled trees, followed by repeated application of broad spectrum herbi-
cides (i.e., sustained devegetation), resulted in dramatic and sustained
increases in leaching losses of NO3ꢀ. The results of this study have
ꢀ
sometimes been interpreted to mean that NO3 is the predominant
ionic form of soil-N on all clearcuts. For example, Kronzucker et al.
(1997) asserted that conifer plantation failures in B.C. (the data they
reported refer to the area of harvested forest that had not yet satisfied
the prevailing definition of ‘‘satisfactorily regenerated’’ based on pro-
vincial regeneration standards, rather than to areas which had been
regenerated and then failed, as they implied) were largely due to
increased nitrification rates after clearcut harvesting. In contrast, the
results of our study, and those of many others (e.g., Fredriksen, 1971;
Aber et al., 1983; Stevens et al., 1995), suggest that many clearcut
sites have a high resistance to NO3ꢀ-leaching after clearcutting.
Our data on post-clearcutting N sinks pose a problem. Martin et al.
and Bradley et al. (this issue) show a substantial release of N from the
forest floor during the assart period. Neither the understory and the
overstory vegetation, nor the forest floor leachates, provided a signifi-
cant sink for the N that was released. In contrast, measurements of
N-storage in the top 30 cm layer of mineral soil in the 4-year-old
clearcut (the clearcut age that appeared to coincide with the onset of
the assart period) was double that in the old-growth stand. This raises
the question as to whether the mineral soil was the largest sink for
forest floor N after disturbance. Nadelhoffer et al. (1992) identified the
mineral soil horizon of Spodosols as a major sink for added-N in
forested catchments in Maine. However, Hedin et al. (1995) showed
that forest soils have a maximum organic-N storage capacity which
determines the ability of the upper mineral soil to retain sudden large
inputs of N, and DON was the predominant form of N leaching from

Kimmins, Martin, and Bradley
63
the forest floor in our study. This capacity is dependent on the CEC,
soil texture and the mineralogy of the site, which are perhaps the most
important site factors controlling N retention and resiliency of forests
after clearcutting. Variation in these factors could explain why such
large differences in reported rates of organic-N loss from one site to
the next are commonly observed (e.g., Seely et al., 1998). For this
reason, we did not measure mineral-N storage on the 26-year-old
clearcut because of the different geological parent material at this site.
The parent material and mineral soil texture at the other four sites were
similar (Martin et al., this issue).
At first inspection, the observed chronological pattern in mineral
soil N-storage appears to provide an explanation (a sink) for the N
released from the forest floor. However, there are several problems
with this interpretation.
1. The mineral soil N content, lysimetry and vegetation regrowth
studies were not replicated, a situation which is common for chronose-
quence studies (Cole and Van Miegroet, 1989). We therefore do not
have statistical support for the interpretation that changes in N storage
in the upper mineral soil can explain the fate of the N lost from the
forest floor. Our evidence is only presumptive, based on (a) large
losses of forest floor-N, 3-to-6 years after clearcutting (Martin et al.,
this issue), and (b) plant-N uptake and leaching losses being too small
to account for this lost N.
2. We have no lysimetry data for the first three post-clearcutting
years, data that are needed to corroborate the leaching of forest floor N
to the upper mineral soil in sufficient quantities during this period. Our
data on forest floor N dynamics suggest that the majority of the loss of
forest floor N did not occur until after the time when the higher levels
of mineral soil N were observed.
3. The largest observed loss of forest floor N-capital occurred be-
tween the third and sixth years after clearcutting, a period when miner-
al soil N storage data show a decrease. The lysimetry data for the 3-8
year post-clearcutting period provide no evidence of high N fluxes.
4. The texture of the mineral soil at these sites was mainly sandy-
loam, which suggests neither a high CEC nor a high capacity to retain
DON (the major form of N leached).
5. N-leaching rates out of the forest floor were similar to N-leaching
rates at 60 cm depth into the mineral soil, although the use of different
types of lysimeter at the two depths suggests caution in making com-

JOURNAL OF SUSTAINABLE FORESTRY
64
parisons. Caution is also needed because our mineral soil lysimetry
data was taken from a depth of 60 cm whereas mineral soil N storage
was only measured down to a depth of 30 cm. We assume mineral-N
soil storage between 30 cm and 60 cm depth did not change, based on
the lack of evidence of solution N transfers from the forest floor to the
upper mineral soil, but we cannot test this assumption.
6. Pedogenic processes in the mineral soil are expected to be slower
than in the forest floor. Hence, the impact of clearcutting is expected to
influence the forest floor more rapidly than the mineral soil. It seems
unreasonable that mineral soil N could respond so dramatically and at
the same time as, or even in advance of, the changes in forest floor N.
The validity of the chronosequence in terms of changes in forest floor
N-capital has been established (Martin et al., this issue). This was not
done for the mineral soil. We have assumed that the chronosequence
characteristics of the mineral soil were more a function of historical
site differences than of the ecosystem processes occuring after clear-
cutting. We have no way of testing this based on the data we have.
The idea that the mineral soil buffered large potential losses of
forest floor N following clearcutting is appealing. It is certainly pos-
sible that significant quantities of solution-N were leached from the
forest floor during the first 3 years following disturbance. Martin et al.
(this issue) noted that forest floor N-capital measured on the 3-year-
old clearcut site may have been overestimated by as much as 16%
suggesting a larger loss of forest floor N-capital during the first three
years than the forest floor data suggest. We could not verify this
because there were no comparable 1-year-old clearcuts in the study
area. We estimated from our biomass data that the clearcuts had log-
ging slash containing an estimated 396 kg N haꢀ1 (59% green needles)
(Table 2). It is possible that a partial release of this N, in conjunction
with release of N from the forest floor, could have resulted in the
eluviation of significant amounts of dissolved-N during the first three
years. However, the data do not suggest that this could provide the
major explanation as to where the N released from the forest floor
went; it could not explain the approximately 1080 kg N haꢀ ob-
1
served difference in the upper mineral soil between the ‘‘old growth’’
and the 4-year-old clearcut. We are left with the conclusion that we
have not satisfactorily explained the post-clearcutting fate of the forest
floor N on these sites. The search for the missing sink is presented in a
subsequent paper.

Kimmins, Martin, and Bradley
65
SUMMARY AND CONCLUSION
Following clearcutting on south-central Vancouver Island, the re-
generating Pacific silver fir--western hemlock did not accumulate sig-
nificant amounts of N until at least 10 years after disturbance. The
early-seral understory vegetation provided a larger sink for N released
from the forest floor in the first 10 years after clearcutting. Since the
amount of N released from the forest floor during this period far
exceeded the amount of N sequestered in both the overstory and the
understory vegetation, the understory vegetation was considered to be
an asset for the long-term productivity of the site, rather than a com-
petitor for N. The short-lived understory vegetation ‘‘captured’’ some
of the assart flush of N and subsequently returned it to the forest floor
as ‘‘green manure.’’ Annual leaching losses of dissolved-N increased
by 10 kg N haꢀ1 during the 3-to-5 year period after clearcutting, but
were 2-to-5 kg N haꢀ1 lower than in the old-growth stand during the
6-to-12 year period. DON was the largest vector for N in both the
forest floor and mineral soil leachates, whereas NH4+ was the predom-
inant ionic form of DIN. An apparent large increase in upper mineral
soil N storage four years after clearcutting suggested that the mineral
soil could have been the major sink for the N released from the forest
floor. However, a number of missing data and inconsistencies made us
suspicious of this interpretation. We conclude that the pattern of min-
eral soil N probably represents pre-study site differences, and that at
this point in the study we had failed to provide an adequate explana-
tion for the fate of the forest floor N.
For the present study, it was necessary to make a number of as-
sumptions about such values as the hydrological flux and the N con-
tent of roots. One of the ways to assess the accuracy of these assump-
tions would be to synthesize the data acquired in the present study and
merge these with other data previously gathered from the same sites to
prepare an overall budget for forest floor N capital. In a subsequent
paper, we will construct a comprehensive N-budget model describing
the inputs and outputs to the forest floor. This will allow us to evaluate
the degree to which we have accounted for N losses, and to reveal
possible sinks and processes that were not considered in the earlier
part of the study.

JOURNAL OF SUSTAINABLE FORESTRY
66
REFERENCES
Aber, J.D., Melillo, J.M. and C.A. Federer. 1982. Predicting the effects of rotation
length, harvest intensity and fertilization on fiber yield from northern hardwood
forests in New England. For. Sci. 28: 31-45.
Aber, J.D., Melillo, J.M., McClaugherty, C.A. and K.N. Eshleman. 1983. Potential
sinks for mineralized nitrogen following disturbance in forest ecosystems. In: R.
Hallberg (ed.) Environmental Biogeochemistry. Proc. 5th Int. Symp. Env. Biogeo-
chemistry, Ecol. Bull. 35:179-192.
Bradley, R.L., Kimmins, J.P. and W.L. Martin. (2002). Post-clearcutting chronose-
quence in the B.C. coastal western hemlock zone: II. Tracking the assart flush. J.
Sust. For. 14(1):23-43.
Campbell, G.S., 1977. An Introduction to Environmental Biophysics. Springer Ver-
lag, NY.
Cole, D.W. and H. Van Miegroet. 1989. Chronosequences: A technique to assess
ecosystem dynamics. Pp. 5-23 In: W.J. Dyck and C.A. Mees (Eds.) Research
Strategies for Long-Term Site Productivity. Proceedings, IEA/BE A3 Workshop,
Seattle, WA. IEA/BE A3 Report No. 8. Forest Research Institute, New Zealand,
Bulletin 152.
CSSC. 1978. The Canadian System of Soil Classification. Can. Dept. Agric., Publ.
1646, Ottawa, pp.164.
Environment Canada. 1982. Canadian Climate Normals. Vol. 1. Canadian Climate
Program, Env. Can., Atm. Env. Serv., Ottawa.
Fahey, T.H. and J.W. Hughes. 1994. Fine root dynamics in a northern hardwood
forest ecosystem, Hubbard Brook Experimental Forest, NH. J. Ecol. 82: 533-548.
Feller, M.C., Kimmins, J.P. and K.A. Scoullar. 1983. FORCYTE-10: Calibration data
and simulation of potential long-term effects of intensive forest management on
site productivity, economic performance, and energy benefit/cost ratio. In: R.
Ballard and S.P. Gessel (eds.), IUFRO Symposium on Forest Site and Continuous
Productivity. Univ. Washington, Seattle, Aug. 1982.
Fredrikson, R.L. 1971. Forest Land Uses and Stream Environment: Proceedings of a
Symposium. J.T. Kryger and J.D. Hall (eds). Oregon State University Press,
Corvallis. Pp. 125
Gholtz, H.L., Grier, C.C., Campbell., A.G. and A.T. Brown. 1979. Equations for
estimating biomass and leaf areas of plants of the Pacific Northwest. Oregon St.
For. Res. Lab. Publ., Res. Pap. 41.
Graham, R.T. and J.L. Kingery. 1990. Seedling damage and mortality of conifer
plantations on transitory ranges in northern and central Idaho. Proc. Vertebr. Pest
Conf., Davis, CA. July, 1990. Pp. 209-213.
Grehan, J.R., Paricer, B.L., Wagner, D.L., Rosovsky, J., and J. Aleong. 1992. Root
damage by the conifer swift moth: a mortality factor in montane red spruce
regeneration. For. Sci. 38: 611-622.
Grier, C.C., Vogt, K.A., Keyes, M.R., and R.L. Edmonds. 1981. Biomass distribution
and above- and belowground production in young and mature Abies amabilis
zone ecosystems of the Washington Cascades. Can. J. For. Res. 11: 155-167.
Hedin, L.O., Armesto, J.J. and A.R. Johnson. 1995. Patterns of nutrient loss from

Kimmins, Martin, and Bradley
67
unpolluted, old-growth temperate forests: Evaluation of biogeochemical theory.
Ecology 76: 493-509.
Hillel, D. 1980. Fundamentals of Soil Physics. Academic Press, NY.
Howard, P.J.A. and D.M. Howard. 1994. Ammonification of complexes prepared
from gelatin and aqueous extracts of leaves and freshly-fallen litter of trees on
different soil types. Soil. Biol. Biochem. 25: 1249-1256.
Kimmins, J.P. 1993. Scientific foundations for the simulation of ecosystem function
and management in FORCYTE-11. Information Report NOR-X-328. Forestry
Canada--Northwest Region, Northern Forestry Centre, Edmonton. 88 pp.
Kimmins, J.P. and K. Scoullar. 1984. The role of modelling in tree nutrition research
and site nutrient management. In: G.D. Bowen and E.K.S. Nambiar (eds.), Nutri-
tion of Plantation Forests. Academic Press, NY, pp-463-487.
Klinka, K., Green, R.N., Trowbridge, R.L., and L.E. Lowe. 1981. Taxonomic classi-
fication of humus forms in ecosystems of British Columbia. First Approximation.
Land Management Report Number 8. Ministry of Forests, Victoria, B.C., 54 pp.
Krajina, V.J. 1969. Ecology of forest trees in British Columbia. Ecol. West. N. Am. 2:
1-146.
Kronzucker, H.J., Siddiqi, M.Y. and A.D.M. Glass. 1997. Conifer root discrimination
against nitrate and the ecology of forest succession. Nature 385: 59-61.
Krumlik, G.J. 1979. Comparitive study of nutrient cycling in the subalpine Mountain
Hemlock zone of British Columbia. Unpublished Ph.D. thesis. Faculty of Forest-
ry, Univ. British Colombia, Vancouver.
Likens, G.E., Bormann, F.H., Johnson, N.M., Fisher, D.W. and R.S. Pierce. 1970.
Effects of clearcutting and herbicide treatment on nutrient budgets in the Hubbard
Brook watershed-ecosystem. Ecol. Mono. 40: 23-47.
McInnis, B.G. and M.R. Roberts. 1995. Seedling microenvironment in full-tree and
tree-length logging slash. Can. J. For. Res. 25: 128-136.
Marks, P. 1974. The role of pin cherry (Prunus pensylvanica L.) in the maintenance
of stability in northern hardwood ecosystems. Ecol. Monogr. 44: 73-88.
Martin, W.L., Bradley, R.L. and J.P. Kimmins. (2002). Post-clearcutting chronose-
quence in the B.C. coastal western hemlock zone: I. Changes in forest floor C and
N storage. J. Sust. For. 14(1):1-22.
Morris, D.M., Kimmins, J.P. and D.R. Duckert. 1997. The use of soil organic matter
as a criterion of the relative sustainability of forest management alternatives: A
modelling approach using FORECAST. For. Ecol. Manag. 94: 61-78.
Nadelhoffer, K., Aber, J., Downs, M., Fry, B., and J. Melillo. 1992. Biological sinks
for N additions to a forested catchment. CEC Report of the Proceedings of the
International Symposium on Experimental Manipulations of Biota and Biochemi-
cal Cycling in Ecosystems. 18-20 May, 1992, Copenhagen.
Northup, R.R., Yu, Z., Dahlgren, R.A., and K.A. Vogt. 1995. Polyphenol control of
nitrogen release from pine litter. Nature 377: 227-229.
Pojar, J., Klinka, K. and D.A. Demarchi. 1991. Coastal Western Hemlock zone.
Pp. 95-111 In: D. Meidinger and J. Pojar (Eds.), Ecosystems of British Columbia.
Research Branch, Ministry of Forests, Victoria, B.C.
Qian, H., and K. Klinka. 1995. Spatial variability of humus forms in some coastal
forest ecosystems of British Columbia. Ann. Sci. For. 52: 653-666.

JOURNAL OF SUSTAINABLE FORESTRY
68
Seely, B., Lajtha, K., and G.D. Salvussi. 1998. Transformation and retention of
nitrogen in a coastal forest ecosystem. Biogeochem. 42: 325-344.
Stevens. P.A., Norris, D.A., Williams, T.G., Hughes, S., Durrant, D.W.H., Anderson,
M.A., Weatherley, N.S., Hornung, M. and C. Woods. 1995. Nutrient losses after
clearfelling in Beddgelert Forest: A comparison of the effects of conventional and
whole-tree harvest on soil water chemistry. Forestry 68: 15-131.
Taylor, B., Kremsater, L., and R. Ellis. 1997. Adaptive management of forests in
British Columbia. ISBN 0-7726-3333-9. B.C. Ministry of Forests, Forest Practic-
es Branch, Victoria, B.C.
Received: November 3, 1999
Accepted: August 28, 2000
for faculty/professionals with journal subscription recommendation
authority for their institutional library . . .
If youhavereadareprint orphotocopy of this article, would
you like to make sure that your library also subscribes to
this journal? If you have the authority to recommend sub-
scriptions to your library, we will send you a free sample
copy for review with your librarian. Just fill out the form below--and make
sure that you type or write out clearly both the name of the journal and
your own name and address.
( ) Yes, please send me a complimentary sample copy of this journal:
(please write in complete journal title here--do not leave blank)
I will show this journal to our institutional or agency library for a possible
subscription.
The name of my institutional/agency library is:
NAME:
INSTITUTION:
ADDRESS:
CITY:
STATE:
ZIP:
Return to: Sample Copy Department, The Haworth Press, Inc.,
10 Alice Street, Binghamton, NY 13904-1580