Directory UMM :Data Elmu:jurnal:T:Tree Physiology:Vol15.1995:
Tree Physiology 15, 427--432
© 1995 Heron Publishing----Victoria, Canada
Effects of irrigation, spacing and fertilization on flowering and growth
in young Alnus rubra
CONSTANCE A. HARRINGTON and DEAN S. DEBELL
Pacific Northwest Research Station, Forestry Sciences Laboratory, 3625 93rd Ave. S.W., Olympia, WA 98512-9193, USA
Received May 31, 1994
Summary Flowering and vegetative growth were assessed in
19 half-sib families of Alnus rubra Bong. planted in a replicated
field trial near Olympia, Washington, USA. The trial consisted
of three square spacings (0.5, 1.0 and 2.0 m), two irrigation
regimes (low and high), and two fertilization treatments (0 and
300 kg P ha −1). Male and female flowers were surveyed in all
plots for all families at plantation ages 4 and 5 years. Female
strobili were surveyed for seven families in the 2-m spaced
plots at plantation age 6 years. The percentage of trees flowering and the number of flowers per tree were always greatest,
and height and diameter growth were always least, in the
low-irrigation regime. Phosphorus fertilization had no effect
on the percentage of trees flowering or on 5-year height or
diameter growth; it had a positive but small effect on the
number of female flowers per tree at age 5 years. Wider spacing
resulted in larger trees, higher rates of flowering, and higher
tree survival. Within each irrigation regime, the percentage of
trees flowering increased as tree size increased. There was
substantial variation in flowering among families, with positive
but low correlations between tree size and flowering attributes.
At ages 4 and 5 years, the ratio of number of trees flowering in
the low-irrigation regime to number of trees flowering in the
high-irrigation regime differed among families. By age 6 years,
many more trees flowered than in previous years, and differences between irrigation regimes were reduced. Early growth
rates were rapid and resulted in substantial crown recession and
mortality in the closer spacings by age 5 years. We conclude
that spacings less than 2 m should only be used in seed
production areas if roguing can be done by age 2 to 3 years.
Keywords: diameter growth, flower production, height growth,
phosphorus.
Introduction
Alnus rubra Bong. occurs naturally from central California to
southeastern Alaska and is the most abundant hardwood tree
species in western Oregon, Washington and British Columbia.
Although considered to have no value for many decades, Alnus
rubra is now appreciated for its unique ecological attributes
(e.g., N2-fixation and immunity to Phellinus root rot) and its
contribution to the forest products economy of the region.
Several operational plantations of A. rubra have recently been
established, and the biology and management of the species
have been summarized (Hibbs et al. 1994). General guidelines
are available for collection and treatment of A. rubra seed
(Hibbs and Ager 1989), but detailed information on reproductive processes, including variability in flowering within and
among Alnus stands and the effects of management practices,
is limited (cf. Brown 1986 and Ager et al. 1994). We have
investigated the effects of spacing, fertilization and irrigation
on vegetative growth and flowering in 19 half-sib families of
A. rubra under short-rotation intensive culture regimes.
Materials and methods
The study reported here is one of several trials undertaken to
evaluate short-rotation intensive culture regimes for wood,
fiber or biomass production. This replicated trial, planted on a
site with uniform conditions and good access, provided the
opportunity to make repeated observations on reproductive
development and vegetative growth of A. rubra in response to
various treatments. The trial was installed in spring 1986 on a
xeric site near Olympia, Washington (47°00′ N, 122°45′ W).
The soil is a somewhat excessively drained, loamy fine sand
formed in sandy glacial outwash. Slope is 0--1%; elevation is
50 m. During the study, precipitation and temperature were
monitored with an on-site weather station. Based on a longterm weather station located 12.5 km from the study area,
mean annual precipitation is 129 cm with only 19 cm falling
from May 1 through September 30 (US Dept. of Commerce
1961).
The experimental design was a randomized complete block
design on three adjacent blocks. Tested were three square
spacings (0.5, 1.0 and 2.0 m), two irrigation regimes (low and
high), and two phosphorus fertilization treatments (0 and 300
kg P ha −1 as triple superphosphate). Irrigation regimes were
applied to whole plots, fertilization treatments to split plots,
and spacings to split-split plots. The low-irrigation regime was
intended to provide just enough supplemental water to ensure
survival and tree health in this regime; approximately 15 cm of
water per year was applied during June, July and August. The
high-irrigation regime was intended to increase tree growth
and thus accelerate the rate of stand development on this dry
site; approximately 55 cm of water per year was applied in this
regime. In most years, the high-irrigation treatment began
428
HARRINGTON AND DEBELL
between late May and mid-June; however, in 1991, irrigation
was not begun until July 5 because of equipment failures.
Phosphorus fertilization was chosen because A. rubra is a
nitrogen-fixing species whose growth is closely linked with P
status (Radwan and DeBell 1994). The P fertilizer was applied
with a spreader and disked into the surface soil. All plots were
maintained in a weed-free condition throughout the experiment.
Each measurement plot contained 100 trees and was surrounded by a minimum of three buffer rows. Within each
measurement plot, there were five to seven container-grown
seedlings from each of 19 half-sib families of A. rubra. Families were randomly assigned to planting spots. The families
were from locations that provided a range in latitude and
elevation (Table 1). All trees were measured annually for total
height and basal diameter (0.3 m above the ground); a selected
subsample of trees in each plot was measured periodically for
height growth during the growing seasons of 1986--1988 and
for diameter growth during the growing seasons of 1987-1991.
In spring 1990 and 1991, all plots (approximately 3000
trees) were surveyed for the presence of reproductive structures. In both years, each tree was coded as not flowering or as
having a low, medium or high number of flowers. For some
analyses, the categorical values were transformed to numerical
values based on the following conversion: none = 0, low = 5,
medium = 35, and high = 70 (the numerical values were
midpoints of categories based on ocular estimates of the number of inflorescences per tree). In spring 1993, seven of the 19
Table 1. Parent location and relative rank based on height of half-sib
families of Alnus rubra.
Geographic location,
tree number
Latitude
(N)
Elevation
(m)
Rank1
Code2
Telegraph Cove, BC, #3
Telegraph Cove, BC, #4
Nanaimo, BC, #3
Concrete, WA, #33
Forks, WA, #9
Forks, WA, #15
John’s River, WA I, #1
John’s River, WA II, #3
John’s River, WA II, #83
John’s River, WA II, #93
McCleary, WA, #10
Nisqually Delta, WA #402
Nisqually Delta, WA #4043
Nisqually Delta, WA #405
Nisqually Delta, WA #407
Elbe, WA, #4143
Elbe, WA, #4153
Carson, WA, #23
Otis, OR, #4
50°45′
50°45′
49°15′
48°30′
48°00′
48°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
46°45′
46°45′
46°00′
45°00′
n/a
n/a
340
65
50
50
75
150
150
150
100
0
0
0
0
380
380
365
120
15
19
10
14
18
12
8
16
17
2
11
5
6
9
1
4
3
7
13
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
1
2
3
Rank based on fall 1992 height in 2-m plots in the high-irrigation
regime (1 = tallest).
Code used for plotting family values in Figures 3 and 4.
Families surveyed for 1992 female strobili.
families in the 2-m plots were selected for an additional assessment of the number of woody ‘‘cone’’ clusters (i.e., female
inflorescences that had opened and shed seed during the preceding months) per tree. At least 15% of the trees in each
selected family had female flowers in the 2-m spacing +
low-irrigation treatment in 1991; in addition, the selected
families exhibited a range in apparent sensitivity of flowering
to irrigation. Because the 1993 survey provided a measure of
the reproductive output for 1992, the information is reported
here as 1992 female flowers. The estimates for 1992 are probably conservative because flowers may have aborted or been
destroyed between the time of flowering and when they were
surveyed.
Total oven-dry, aboveground woody biomass, foliar biomass
and leaf area were predicted for each tree from site-specific
equations based on cultural regime, diameter and total height
(DeBell et al. 1991). Statistical tests of the effects of the
irrigation, fertilization and spacing treatments on tree size and
flower production were done by ANOVA for a split-split plot
experiment (Snedecor and Cochran 1980). Treatment effects
were judged significant at P ≤ 0.05; however, actual probability levels are provided. In 1990, almost all trees that flowered had similar quantities of male and female flowers, so
separate analyses were not made for each flower type. Because
the frequency of male and female flowering differed in 1991,
separate analyses were made for each sex and for both types
combined for that year. Basic data on flowering by family were
summarized by block and treatment and examined graphically.
Correlations among flowering variables and tree size attributes
were calculated based on values for individual trees. Use of
values from individual trees in the correlation analyses assumes each tree was an independent observation and ignores
the basic experimental design; thus, results from these analyses were only used to describe the general relationships present
in the data and not to test for significant associations between
variables.
Results
By the end of the 1990 growing season, both the spacing and
irrigation treatments had markedly influenced tree size and
survival (Table 2). Tree size was significantly greater in the
1.0- and 2.0-m spacing treatments than in the 0.5-m spacing
treatment, and it was also significantly greater in the high-irrigation regime than in the low-irrigation regime. Tree size was
not affected by the fertilization treatments, and there were no
significant interactions between the fertilization and other cultural treatments (Table 3).
The high-irrigation regime altered the seasonal pattern of
growth. For example, in 1987, 1989 and 1990 in the 2-m
spacing treatment, diameter growth of the trees in the high-irrigation regime peaked later in the growing season and high
growth rates occurred over a longer period than in the low-irrigation regime (Figure 1). In a study of the same trees, DeBell
and Giordano (1994) noted that seasonal height growth continued for a longer period in the 1986--1988 growing seasons in
the high-irrigation regime than in the low-irrigation regime.
FLOWERING AND GROWTH IN YOUNG ALNUS RUBRA
429
Table 2. Effects of spacing and irrigation treatment (Low-I and High-I) on tree size and survival at the end of the 1990 growing season (5 years
after planting). Values are means ± 1 SE.
Spacing (m)
0.5
1.0
2.0
Height (m)
Diameter (cm)
Survival (%)
Low-I
High-I
Low-I
High-I
Low-I
High-I
4.5 ± 0.5
5.5 ± 0.3
6.1 ± 0.2
7.3 ± 0.1
8.2 ± 0.1
8.3 ± 0.2
3.3 ± 0.3
4.7 ± 0.2
7.5 ± 0.3
5.1 ± 0.1
7.1 ± 0.1
9.5 ± 0.1
53 ± 3.1
94 ± 0.4
99 ± 0.3
33 ± 2.1
67 ± 2.2
98 ± 0.7
Table 3. Significance of ANOVA model components on tree size 5 years after planting and on number of inflorescences by year and type (B =
both, F = female, M = male).
Source of variation
df
Probability of > F-value occurring
Fall 1990
Block
Irrigation1
Fertilization1
Irrigation × fertilization1
Spacing
Irrigation × spacing
Fertilization × spacing
Irrigation × fertilization × spacing
1
2
1
1
1
2
2
2
2
Number of inflorescences by year and type
Diameter
Height
90 B
91 B
91 F
91 M
0.01
0.01
0.62
0.55
< 0.01
< 0.01
0.81
0.12
< 0.01
0.02
0.48
0.83
< 0.01
0.44
0.32
0.15
0.44
0.24
0.95
0.99
< 0.01
0.03
0.81
0.86
0.05
0.08
0.59
0.94
< 0.01
< 0.01
0.92
0.81
0.03
0.07
0.05
0.13
< 0.01
< 0.01
0.74
0.45
0.22
0.11
0.80
0.59
< 0.01
< 0.01
0.75
0.67
Irrigation tested using block × irrigation as the error term (df = 2); fertilization and irrigation × fertilization tested using block × fertilization
(irrigation) as the error term (df = 4); the split-split error term had 16 df.
associated with poor flowering, and fertilization had no effect
on the percentage of trees flowering or on the mean number of
mature female flower clusters per tree (Table 6). The percentage of trees with flowers was greater in 1991 than in 1990 and,
for the families sampled, greater in 1992 than in 1991 (Table 4). Although different stages and types of flowers were
assessed in the various surveys, it is clear that the percentage
of trees flowering increased each year.
Within each irrigation regime, as tree size increased, there
was an increase in the percentage of trees within a family with
Table 4. Effects of irrigation and spacing on the percentage of trees
flowering.
Figure 1. Periodic diameter increment in 2-m plots of Alnus rubra by
irrigation regime and year.
Flowering was enhanced by the 1.0- and 2.0-m spacing
treatments. Spacing and the spacing × irrigation interaction
were significant in the ANOVA of all measures of flowering in
1990 and 1991 (Table 3). The high-irrigation regime suppressed flowering, and the difference between irrigation regimes was greatest in the 2-m spacing treatment (Table 4).
Fertilization with P did not significantly affect the percentage
of trees with inflorescences in any year, but it significantly
increased the number of female inflorescences per tree in 1991
(Tables 3 and 5). In 1992, the high-irrigation regime was again
Year
Strobilus type Irrigation
1990
Both
1991
Both
1991
Male
1991
Female
19921
Female
1
Low
High
Low
High
Low
High
Low
High
Low
High
Percent of trees with strobili
0.5 m
1.0 m
2.0 m
2.5
0.7
9.9
2.6
7.3
2.2
5.3
0.7
---
3.4
0.2
15.5
9.0
12.1
7.1
8.0
4.5
---
17.6
7.2
49.4
21.1
41.8
12.0
31.2
15.2
63.2
43.9
Only one spacing treatment and seven families surveyed (see text
for details).
430
HARRINGTON AND DEBELL
Table 5. Effects of irrigation, fertilization and spacing on number of female inflorescences per tree in spring 1991.
Spacing (m)
Low irrigation
0 kg P ha
0.5
1.0
2.0
High irrigation
−1
300 kg P ha
0.1
0.8
4.4
0.9
0.7
5.9
Table 6. Significance of ANOVA model components on percentage of
trees with 1992 female strobili and mean number of inflorescences
(only 2-m spacing sampled).
Source of variation
Block
Irrigation1
Fertilization
Irrigation × fertilization
1
df
2
1
1
1
Probability of > F-value
% With strobili
No. of clusters
0.98
0.01
0.28
0.83
0.09
0.06
0.26
0.78
Irrigation tested using block × irrigation as the error term (df = 2);
the split-plot error term had 4 df.
male strobili (Figure 2). For each family, the relationship
between mean tree diameter (averaged across trees within one
irrigation, spacing and fertilization combination) and percent
flowering was positive. Within all tree sizes but the smallest
(which did not flower), there was a substantial range in flowering as a result of differences among families in propensity to
flower.
Correlations between number of female strobili in 1991 and
woody biomass, leaf biomass, leaf area and basal diameter
were positive but low (r ≈ 0.20). When the trees were separated
by irrigation regime, the correlations increased to 0.33--0.36
for trees in the low-irrigation regime but were unchanged for
trees in the high-irrigation regime (0.22--0.24). Correlations
between height and number of female strobili in 1991 were
Figure 2. Percentage of trees per family with 1991 male strobili versus
basal diameter at the end of the 1990 growing season. Each point is
based on approximately 15 trees and represents the mean value for a
family by irrigation, spacing, and fertilization regime.
−1
0 kg P ha −1
300 kg P ha −1
0.0
0.2
1.4
0.1
0.6
1.4
lower than those for biomass, leaf area or diameter (r = 0.05
for both regimes, r = 0.20 for trees in the low-irrigation regime,
r = 0.13 for trees in the high-irrigation regime). In the low-irrigation regime, the highest correlations were between total
number of flowers produced in 1990 + 1991 and leaf area (r =
0.40) or leaf biomass (r = 0.40). A similar correlation value
was observed between flowering and stem diameter, because
both leaf area and leaf biomass were highly correlated with
basal diameter (r = 0.96--0.97). Correlations between numbers
of clusters per tree and height and diameter growth the year the
flowers were initiated (1991) were low, but were indicative of
a stronger relationship between flower production and diameter growth (r = 0.18) than between flower production and
height growth (r = −0.06).
Although some families were more likely to flower than
others, individual trees did not exhibit a tendency to produce a
consistently high or low number of flowers (or to alternate high
and low flower production). There were also differences
among families in their flowering response to irrigation regime. In 1991 at the 2-m spacing, the percentage of trees with
female flowers ranged from 2.8% for John’s River II, #8 in the
high-irrigation regime to 63.6% for Nisqually Delta, #402 in
the low-irrigation regime (Figure 3). The ratio of the percentage of trees flowering in the low-irrigation treatment to the
percentage flowering in the high-irrigation treatment ranged
by family from 0.5 to 8.5, with an overall mean of 2.05.
Differences between irrigation regimes in percentage of trees
with female strobili was less in 1992 than in 1991 (range for
Figure 3. Percentage of trees in 2-m plots with 1991 female flowers by
irrigation regime and family. Dashed line indicates equivalence of
irrigation regimes. The greater the vertical distance from the line, the
more a family differed in flowering response between irrigation regimes. Letters identify families (see Table 1).
FLOWERING AND GROWTH IN YOUNG ALNUS RUBRA
1992 of 1.22 to 2.00, range for same families in 1991 of 1.3 to
6.5) (Figure 4); however, the 1992 ratio of the mean number of
strobili clusters per tree between irrigation regimes varied from
1.0 to 4.7 among families, indicating that family differences in
flowering response to irrigation were still being expressed.
Discussion
Flower initiation occurs during the summer (Furlow 1979),
flowering takes place the following February or March, and
seeds are ripe in the fall. Mature female strobili are woody and
cone-like in appearance, and remain intact and attached to the
plant during seed dispersal and for a time after dispersal is
completed. Previous-year female strobili were observed on
some trees in spring 1990, indicating that female flower initiation occurred in 1988 when the trees were 3 years old from
seed (2 years since planting). At the widest spacing, approximately half of the trees in seven families produced female
strobili in 1992, indicating female flower initiation in 1991
(i.e., 6 years from seed). These results are in agreement with
Stettler’s (1978) observation that A. rubra begins to flower at
age 3 to 4 years for individual trees and age 6 to 8 years for
most dominant trees in a stand. Several species of Betula that
naturally flower at young ages have been induced to flower
sooner when grown continuously, suggesting that Betula must
attain a minimum size before flowering (Longman 1984).
Thus, it may be possible to accelerate flowering in A. rubra
with growth-promoting treatments similar to those used with
Betula.
Crown size or volume is an important factor affecting flowering and seed production. Treatments such as wide initial
spacing or thinning are known to increase flowering or seed
production because they increase crown size and promote tree
vigor (Matthews 1963). We found positive correlations between flowering and variables that quantify crown size (e.g.,
branch biomass and leaf area). We conclude that variables that
quantify crown size are better predictors of flowering than
stem diameter if trees are growing in stands with considerable
variation in age, stocking or site quality.
Although N fertilization promotes flowering (Ross and
Figure 4. Mean number of 1992 female strobili clusters per tree in 2-m
plots by irrigation regime and family. Line and symbols as in Figure 3.
431
Pharis 1985), we did not include an N fertilization treatment in
this study because root nodules on A. rubra fix atmospheric N.
The P fertilization treatment was chosen because of its potential to increase tree growth rather than to enhance flowering;
however, the P fertilization did not increase 5-year height or
diameter growth. The lack of response of most flowering
variables to P fertilization is consistent with suggestions by
Matthews (1963) and Sedgley and Griffin (1989) that the
primary purpose of applying nutrients other than N to enhance
flowering or seed production should be to correct nutrient
deficiencies. Flowering of A. rubra may be responsive to P
fertilization on P-deficient sites because growth of the species
is strongly linked to P status (Radwan and DeBell 1994).
Although long-term seed production is enhanced by maintaining tree vigor, it has been suggested that flower initiation
is promoted by conditions that restrict shoot growth or cause
stress during the period when flower initiation can occur (Ross
and Pharis 1985, Owens 1991). Vegetative growth in Alnus is
primarily monopodial (the apical bud is a persistent leader, and
new branches arise laterally below the apex; Swartz 1971);
however, shoots that produce flowers exhibit sympodial
growth (the terminal bud withers, floral inflorescences are in
terminal and subterminal positions, and the main axis is made
up of a series of lateral branches). Additional primary shoot
extension does not occur after a reproductive structure has
been determined. Flower initiation probably occurs in late
June or early July for both A. rubra and A. glutinosa (L.)
Gaertn. (McVean 1955, Brown 1986, Ager et al. 1994); however, anatomical studies have not been reported. Species such
as Alnus with prolonged periods of shoot growth may have
greater leeway in the timing of floral initiation than species
with short periods of shoot extension (Longman 1985). However, if floral initiation in Alnus only occurs under certain
environmental conditions, it is possible that the extended period of shoot growth in the high-irrigation regime reduced or
even precluded floral initiation in some trees. The importance
of differences in genetic control of timing of flowering is
indicated by the study of O’Reilly and Owens (1988) showing
that initiation of seed-cone buds occurs on different dates in
different provenances of Pinus contorta Dougl. ex Loud.
We postulate that the increase in the percentage of trees
flowering each year was due to increases in the number of trees
attaining a minimum size that permits flowering (cf. Longman
1984). It is possible that differences in weather from year to
year also influenced flowering; however, in most tree species,
abundant seed crops do not occur in consecutive years even if
environmental conditions are favorable. Owens (1991) suggested that heavy seed crops alter endogenous conditions and
thereby inhibit reproductive bud development. We found no
indication that occurrence or relative number of flowers in one
year affected the occurrence or relative number of flowers in
the succeeding year. This finding contrasts with the report by
LaBastide and Vredenburch (1970) that seed crops for mature
A. glutinosa follow an annually alternating pattern. The number of flowers produced by the young trees in our study was
substantially lower than that observed for older trees in the
area; thus, it is possible that alternating years of high and low
432
HARRINGTON AND DEBELL
flower production could occur when these trees begin producing large seed crops. No long-term records of flowering or seed
production in A. rubra are available (see Harrington et al. 1994
for summary of short-term records). McVean (1955) concluded that the size of seed crops of A. glutinosa could vary
substantially from year to year, but that ‘‘boom-and-bust’’ patterns of seed production were not typical.
Genetic differences in flowering or seed production have
been observed in many species (Sedgley and Griffin 1989). We
observed differences among families in their propensity to
flower that could not be attributed solely to differences in mean
tree size (cf. Figure 2); however, we used too few members per
family in each treatment combination to test for differences
among families in flowering. Even though family differences
in flowering were evident, overall treatment responses were
clear, indicating that the results should be generally applicable
to other genotypes.
We conclude that if A. rubra stands are being managed for
seed production, the trees should be widely spaced: ≥ 2 m at
the time of the first seed crop and wider spacings as trees get
larger. Because crown recession and competition-related mortality occur very rapidly in dense stands of A. rubra, use of
spacings narrower than 2 m at planting is only feasible if
roguing is done in the first 2 to 3 years. Establishment on dry
sites may be ideal if supplemental irrigation is minimal during
June and July to promote flowering and is limited to amounts
needed to maintain long-term survival and health during the
rest of the year.
Acknowledgments
This research was supported in part by funding from US Dept. of
Energy, Biofuels Feedstock Development Program, Interagency Research Agreement DE-AI05-810R20914. We thank J. Hawks, S.
Bailey and M. Paschke for assistance in conducting the study and
personnel at the Washington State Department of Natural Resources,
Meridian Seed Orchard for their cooperation.
References
Ager, A., Y. Tanaka and J. McGrath. 1994. Biology, ecology, and
utilization of red alder seed. In The Biology and Management of
Red Alder. Eds. D.E. Hibbs, D.S. DeBell and R.F. Tarrant. Oregon
State Univ. Press, Corvallis, OR, pp 159--169.
Brown, S.M. 1986. Sexual allocation patterns in red alder (Alnus rubra
Bong.) along three elevational transects. M.S. Thesis. Univ. Washington, 242 p.
DeBell, D.S., C.A. Harrington and G.W. Clendenen. 1991. Increasing
the productivity of biomass plantations of alder and cottonwood in
the Pacific Northwest. USDA For. Serv., Pac. NW Res. Stn. and US
Dept. Energy, Woody Crops Program, Interagency Agreement DEAI05-810R20914, Annu. Tech. Rep., 60 p.
DeBell, D.S. and P.A. Giordano. 1994. Growth patterns of red alder.
In The Biology and Management of Red Alder. Eds. D.E. Hibbs,
D.S. DeBell and R.F. Tarrant. Oregon State Univ. Press, Corvallis,
OR, pp 116--130.
Furlow, J.J. 1979. The systematics of the American species of Alnus
(Betulaceae). Rhodora 81:1--21.
Harrington, C.A., J.C. Zasada and E.A. Allen. 1994. Biology of red
alder (Alnus rubra Bong.). In The Biology and Management of Red
Alder. Eds. D.E. Hibbs, D.S. DeBell and R.F. Tarrant. Oregon State
Univ. Press, Corvallis, OR, pp 3--22.
Hibbs, D.E. and A.A. Ager. 1989. Red alder: guidelines for seed
collection, handling, and storage. For. Res. Lab Spec. Pub. 18, Col.
For., Oregon State Univ. Press, Corvallis, OR, 6 p.
Hibbs, D.E., D.S. DeBell and R.F. Tarrant. 1994. Biology and management of red alder. Oregon State Univ. Press, Corvallis, OR, 256 p.
LaBastide, J.G.A. and C.L.H. van Vredenburch. 1970. The influence
of weather conditions on the seed production of some forest trees in
the Netherlands. Mededeling, Stichting Bosbouwproefstation ‘De
Dorschkamp,’ Wageningen, No. 102, 12 p.
Longman, K.A. 1984. Physiological studies in birch. Proc. Roy. Soc.
Edinburgh 85B:97--113.
Longman, K.A. 1985. Variability in flower initiation in forest trees. In
Attributes of Trees as Crop Plants. Eds. M.G.R. Cannell and J.E.
Jackson. Inst. Terrestrial Ecol., Nat. Environ. Res. Council, Huntingdon, U.K., pp 398--408.
Matthews, J.D. 1963. Factors affecting the production of seed by forest
trees. For. Abstr. 24:1--13.
McVean, D.N. 1955. Ecology of Alnus glutinosa (L.) Gaertn. I. Fruit
formation. J. Ecol. 44:219--222.
O’Reilly, C. and J.N. Owens. 1988. Reproductive growth and development in seven provenances of lodgepole pine. Can. J. For. Res.
18:43--53.
Owens, J.N. 1991. Flowering and seed set. In Physiology of Trees. Ed.
A.S. Raghavendra. John Wiley and Sons Inc., New York, pp 247-271.
Radwan, M.S. and D.S. DeBell. 1994. Fertilization and nutrition of red
alder. In The Biology and Management of Red Alder. Eds. D.E.
Hibbs, D.S. DeBell and R.F. Tarrant. Oregon State Univ. Press,
Corvallis, OR, pp 216--228.
Ross, S.D. and R.P. Pharis. 1985. Promotion of flowering in tree corps:
different mechanisms and techniques, with special reference to
conifers. In Attributes of Trees as Crop Plants. Eds. M.G.R. Cannell
and J.E. Jackson. Inst. Terrestrial Ecol., Nat. Environ. Res. Council,
Huntingdon, U.K., pp 383--397.
Sedgley, M. and A.R. Griffin. 1989. Sexual reproduction of tree crops.
Academic Press, London, 378 p.
Snedecor, G.W. and W.G. Cochran. 1980. Statistical methods, 7th Edn.
The Iowa State University Press, Ames, Iowa, pp 325--329.
Stettler, R.F. 1978. Biological aspects of red alder pertinent to potential breeding programs. In Utilization and Management of Alder.
Eds. D.G. Briggs, D.S. DeBell and W.A. Atkinson. USDA For.
Serv., Pac. NW For. and Range Exp. Stn., Portland, OR, Gen. Tech.
Rep. PNW-70, pp 209--222.
Swartz, D. 1971. Collegiate dictionary of botany. Ronald Press Co.,
New York, 520 p.
US Department of Commerce Weather Bureau. 1961. Local climatological data with comparative data, 1960, Olympia, Washington.
US Dept. Com., Weather Bureau, Asheville, NC, 4 p.
© 1995 Heron Publishing----Victoria, Canada
Effects of irrigation, spacing and fertilization on flowering and growth
in young Alnus rubra
CONSTANCE A. HARRINGTON and DEAN S. DEBELL
Pacific Northwest Research Station, Forestry Sciences Laboratory, 3625 93rd Ave. S.W., Olympia, WA 98512-9193, USA
Received May 31, 1994
Summary Flowering and vegetative growth were assessed in
19 half-sib families of Alnus rubra Bong. planted in a replicated
field trial near Olympia, Washington, USA. The trial consisted
of three square spacings (0.5, 1.0 and 2.0 m), two irrigation
regimes (low and high), and two fertilization treatments (0 and
300 kg P ha −1). Male and female flowers were surveyed in all
plots for all families at plantation ages 4 and 5 years. Female
strobili were surveyed for seven families in the 2-m spaced
plots at plantation age 6 years. The percentage of trees flowering and the number of flowers per tree were always greatest,
and height and diameter growth were always least, in the
low-irrigation regime. Phosphorus fertilization had no effect
on the percentage of trees flowering or on 5-year height or
diameter growth; it had a positive but small effect on the
number of female flowers per tree at age 5 years. Wider spacing
resulted in larger trees, higher rates of flowering, and higher
tree survival. Within each irrigation regime, the percentage of
trees flowering increased as tree size increased. There was
substantial variation in flowering among families, with positive
but low correlations between tree size and flowering attributes.
At ages 4 and 5 years, the ratio of number of trees flowering in
the low-irrigation regime to number of trees flowering in the
high-irrigation regime differed among families. By age 6 years,
many more trees flowered than in previous years, and differences between irrigation regimes were reduced. Early growth
rates were rapid and resulted in substantial crown recession and
mortality in the closer spacings by age 5 years. We conclude
that spacings less than 2 m should only be used in seed
production areas if roguing can be done by age 2 to 3 years.
Keywords: diameter growth, flower production, height growth,
phosphorus.
Introduction
Alnus rubra Bong. occurs naturally from central California to
southeastern Alaska and is the most abundant hardwood tree
species in western Oregon, Washington and British Columbia.
Although considered to have no value for many decades, Alnus
rubra is now appreciated for its unique ecological attributes
(e.g., N2-fixation and immunity to Phellinus root rot) and its
contribution to the forest products economy of the region.
Several operational plantations of A. rubra have recently been
established, and the biology and management of the species
have been summarized (Hibbs et al. 1994). General guidelines
are available for collection and treatment of A. rubra seed
(Hibbs and Ager 1989), but detailed information on reproductive processes, including variability in flowering within and
among Alnus stands and the effects of management practices,
is limited (cf. Brown 1986 and Ager et al. 1994). We have
investigated the effects of spacing, fertilization and irrigation
on vegetative growth and flowering in 19 half-sib families of
A. rubra under short-rotation intensive culture regimes.
Materials and methods
The study reported here is one of several trials undertaken to
evaluate short-rotation intensive culture regimes for wood,
fiber or biomass production. This replicated trial, planted on a
site with uniform conditions and good access, provided the
opportunity to make repeated observations on reproductive
development and vegetative growth of A. rubra in response to
various treatments. The trial was installed in spring 1986 on a
xeric site near Olympia, Washington (47°00′ N, 122°45′ W).
The soil is a somewhat excessively drained, loamy fine sand
formed in sandy glacial outwash. Slope is 0--1%; elevation is
50 m. During the study, precipitation and temperature were
monitored with an on-site weather station. Based on a longterm weather station located 12.5 km from the study area,
mean annual precipitation is 129 cm with only 19 cm falling
from May 1 through September 30 (US Dept. of Commerce
1961).
The experimental design was a randomized complete block
design on three adjacent blocks. Tested were three square
spacings (0.5, 1.0 and 2.0 m), two irrigation regimes (low and
high), and two phosphorus fertilization treatments (0 and 300
kg P ha −1 as triple superphosphate). Irrigation regimes were
applied to whole plots, fertilization treatments to split plots,
and spacings to split-split plots. The low-irrigation regime was
intended to provide just enough supplemental water to ensure
survival and tree health in this regime; approximately 15 cm of
water per year was applied during June, July and August. The
high-irrigation regime was intended to increase tree growth
and thus accelerate the rate of stand development on this dry
site; approximately 55 cm of water per year was applied in this
regime. In most years, the high-irrigation treatment began
428
HARRINGTON AND DEBELL
between late May and mid-June; however, in 1991, irrigation
was not begun until July 5 because of equipment failures.
Phosphorus fertilization was chosen because A. rubra is a
nitrogen-fixing species whose growth is closely linked with P
status (Radwan and DeBell 1994). The P fertilizer was applied
with a spreader and disked into the surface soil. All plots were
maintained in a weed-free condition throughout the experiment.
Each measurement plot contained 100 trees and was surrounded by a minimum of three buffer rows. Within each
measurement plot, there were five to seven container-grown
seedlings from each of 19 half-sib families of A. rubra. Families were randomly assigned to planting spots. The families
were from locations that provided a range in latitude and
elevation (Table 1). All trees were measured annually for total
height and basal diameter (0.3 m above the ground); a selected
subsample of trees in each plot was measured periodically for
height growth during the growing seasons of 1986--1988 and
for diameter growth during the growing seasons of 1987-1991.
In spring 1990 and 1991, all plots (approximately 3000
trees) were surveyed for the presence of reproductive structures. In both years, each tree was coded as not flowering or as
having a low, medium or high number of flowers. For some
analyses, the categorical values were transformed to numerical
values based on the following conversion: none = 0, low = 5,
medium = 35, and high = 70 (the numerical values were
midpoints of categories based on ocular estimates of the number of inflorescences per tree). In spring 1993, seven of the 19
Table 1. Parent location and relative rank based on height of half-sib
families of Alnus rubra.
Geographic location,
tree number
Latitude
(N)
Elevation
(m)
Rank1
Code2
Telegraph Cove, BC, #3
Telegraph Cove, BC, #4
Nanaimo, BC, #3
Concrete, WA, #33
Forks, WA, #9
Forks, WA, #15
John’s River, WA I, #1
John’s River, WA II, #3
John’s River, WA II, #83
John’s River, WA II, #93
McCleary, WA, #10
Nisqually Delta, WA #402
Nisqually Delta, WA #4043
Nisqually Delta, WA #405
Nisqually Delta, WA #407
Elbe, WA, #4143
Elbe, WA, #4153
Carson, WA, #23
Otis, OR, #4
50°45′
50°45′
49°15′
48°30′
48°00′
48°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
47°00′
46°45′
46°45′
46°00′
45°00′
n/a
n/a
340
65
50
50
75
150
150
150
100
0
0
0
0
380
380
365
120
15
19
10
14
18
12
8
16
17
2
11
5
6
9
1
4
3
7
13
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
1
2
3
Rank based on fall 1992 height in 2-m plots in the high-irrigation
regime (1 = tallest).
Code used for plotting family values in Figures 3 and 4.
Families surveyed for 1992 female strobili.
families in the 2-m plots were selected for an additional assessment of the number of woody ‘‘cone’’ clusters (i.e., female
inflorescences that had opened and shed seed during the preceding months) per tree. At least 15% of the trees in each
selected family had female flowers in the 2-m spacing +
low-irrigation treatment in 1991; in addition, the selected
families exhibited a range in apparent sensitivity of flowering
to irrigation. Because the 1993 survey provided a measure of
the reproductive output for 1992, the information is reported
here as 1992 female flowers. The estimates for 1992 are probably conservative because flowers may have aborted or been
destroyed between the time of flowering and when they were
surveyed.
Total oven-dry, aboveground woody biomass, foliar biomass
and leaf area were predicted for each tree from site-specific
equations based on cultural regime, diameter and total height
(DeBell et al. 1991). Statistical tests of the effects of the
irrigation, fertilization and spacing treatments on tree size and
flower production were done by ANOVA for a split-split plot
experiment (Snedecor and Cochran 1980). Treatment effects
were judged significant at P ≤ 0.05; however, actual probability levels are provided. In 1990, almost all trees that flowered had similar quantities of male and female flowers, so
separate analyses were not made for each flower type. Because
the frequency of male and female flowering differed in 1991,
separate analyses were made for each sex and for both types
combined for that year. Basic data on flowering by family were
summarized by block and treatment and examined graphically.
Correlations among flowering variables and tree size attributes
were calculated based on values for individual trees. Use of
values from individual trees in the correlation analyses assumes each tree was an independent observation and ignores
the basic experimental design; thus, results from these analyses were only used to describe the general relationships present
in the data and not to test for significant associations between
variables.
Results
By the end of the 1990 growing season, both the spacing and
irrigation treatments had markedly influenced tree size and
survival (Table 2). Tree size was significantly greater in the
1.0- and 2.0-m spacing treatments than in the 0.5-m spacing
treatment, and it was also significantly greater in the high-irrigation regime than in the low-irrigation regime. Tree size was
not affected by the fertilization treatments, and there were no
significant interactions between the fertilization and other cultural treatments (Table 3).
The high-irrigation regime altered the seasonal pattern of
growth. For example, in 1987, 1989 and 1990 in the 2-m
spacing treatment, diameter growth of the trees in the high-irrigation regime peaked later in the growing season and high
growth rates occurred over a longer period than in the low-irrigation regime (Figure 1). In a study of the same trees, DeBell
and Giordano (1994) noted that seasonal height growth continued for a longer period in the 1986--1988 growing seasons in
the high-irrigation regime than in the low-irrigation regime.
FLOWERING AND GROWTH IN YOUNG ALNUS RUBRA
429
Table 2. Effects of spacing and irrigation treatment (Low-I and High-I) on tree size and survival at the end of the 1990 growing season (5 years
after planting). Values are means ± 1 SE.
Spacing (m)
0.5
1.0
2.0
Height (m)
Diameter (cm)
Survival (%)
Low-I
High-I
Low-I
High-I
Low-I
High-I
4.5 ± 0.5
5.5 ± 0.3
6.1 ± 0.2
7.3 ± 0.1
8.2 ± 0.1
8.3 ± 0.2
3.3 ± 0.3
4.7 ± 0.2
7.5 ± 0.3
5.1 ± 0.1
7.1 ± 0.1
9.5 ± 0.1
53 ± 3.1
94 ± 0.4
99 ± 0.3
33 ± 2.1
67 ± 2.2
98 ± 0.7
Table 3. Significance of ANOVA model components on tree size 5 years after planting and on number of inflorescences by year and type (B =
both, F = female, M = male).
Source of variation
df
Probability of > F-value occurring
Fall 1990
Block
Irrigation1
Fertilization1
Irrigation × fertilization1
Spacing
Irrigation × spacing
Fertilization × spacing
Irrigation × fertilization × spacing
1
2
1
1
1
2
2
2
2
Number of inflorescences by year and type
Diameter
Height
90 B
91 B
91 F
91 M
0.01
0.01
0.62
0.55
< 0.01
< 0.01
0.81
0.12
< 0.01
0.02
0.48
0.83
< 0.01
0.44
0.32
0.15
0.44
0.24
0.95
0.99
< 0.01
0.03
0.81
0.86
0.05
0.08
0.59
0.94
< 0.01
< 0.01
0.92
0.81
0.03
0.07
0.05
0.13
< 0.01
< 0.01
0.74
0.45
0.22
0.11
0.80
0.59
< 0.01
< 0.01
0.75
0.67
Irrigation tested using block × irrigation as the error term (df = 2); fertilization and irrigation × fertilization tested using block × fertilization
(irrigation) as the error term (df = 4); the split-split error term had 16 df.
associated with poor flowering, and fertilization had no effect
on the percentage of trees flowering or on the mean number of
mature female flower clusters per tree (Table 6). The percentage of trees with flowers was greater in 1991 than in 1990 and,
for the families sampled, greater in 1992 than in 1991 (Table 4). Although different stages and types of flowers were
assessed in the various surveys, it is clear that the percentage
of trees flowering increased each year.
Within each irrigation regime, as tree size increased, there
was an increase in the percentage of trees within a family with
Table 4. Effects of irrigation and spacing on the percentage of trees
flowering.
Figure 1. Periodic diameter increment in 2-m plots of Alnus rubra by
irrigation regime and year.
Flowering was enhanced by the 1.0- and 2.0-m spacing
treatments. Spacing and the spacing × irrigation interaction
were significant in the ANOVA of all measures of flowering in
1990 and 1991 (Table 3). The high-irrigation regime suppressed flowering, and the difference between irrigation regimes was greatest in the 2-m spacing treatment (Table 4).
Fertilization with P did not significantly affect the percentage
of trees with inflorescences in any year, but it significantly
increased the number of female inflorescences per tree in 1991
(Tables 3 and 5). In 1992, the high-irrigation regime was again
Year
Strobilus type Irrigation
1990
Both
1991
Both
1991
Male
1991
Female
19921
Female
1
Low
High
Low
High
Low
High
Low
High
Low
High
Percent of trees with strobili
0.5 m
1.0 m
2.0 m
2.5
0.7
9.9
2.6
7.3
2.2
5.3
0.7
---
3.4
0.2
15.5
9.0
12.1
7.1
8.0
4.5
---
17.6
7.2
49.4
21.1
41.8
12.0
31.2
15.2
63.2
43.9
Only one spacing treatment and seven families surveyed (see text
for details).
430
HARRINGTON AND DEBELL
Table 5. Effects of irrigation, fertilization and spacing on number of female inflorescences per tree in spring 1991.
Spacing (m)
Low irrigation
0 kg P ha
0.5
1.0
2.0
High irrigation
−1
300 kg P ha
0.1
0.8
4.4
0.9
0.7
5.9
Table 6. Significance of ANOVA model components on percentage of
trees with 1992 female strobili and mean number of inflorescences
(only 2-m spacing sampled).
Source of variation
Block
Irrigation1
Fertilization
Irrigation × fertilization
1
df
2
1
1
1
Probability of > F-value
% With strobili
No. of clusters
0.98
0.01
0.28
0.83
0.09
0.06
0.26
0.78
Irrigation tested using block × irrigation as the error term (df = 2);
the split-plot error term had 4 df.
male strobili (Figure 2). For each family, the relationship
between mean tree diameter (averaged across trees within one
irrigation, spacing and fertilization combination) and percent
flowering was positive. Within all tree sizes but the smallest
(which did not flower), there was a substantial range in flowering as a result of differences among families in propensity to
flower.
Correlations between number of female strobili in 1991 and
woody biomass, leaf biomass, leaf area and basal diameter
were positive but low (r ≈ 0.20). When the trees were separated
by irrigation regime, the correlations increased to 0.33--0.36
for trees in the low-irrigation regime but were unchanged for
trees in the high-irrigation regime (0.22--0.24). Correlations
between height and number of female strobili in 1991 were
Figure 2. Percentage of trees per family with 1991 male strobili versus
basal diameter at the end of the 1990 growing season. Each point is
based on approximately 15 trees and represents the mean value for a
family by irrigation, spacing, and fertilization regime.
−1
0 kg P ha −1
300 kg P ha −1
0.0
0.2
1.4
0.1
0.6
1.4
lower than those for biomass, leaf area or diameter (r = 0.05
for both regimes, r = 0.20 for trees in the low-irrigation regime,
r = 0.13 for trees in the high-irrigation regime). In the low-irrigation regime, the highest correlations were between total
number of flowers produced in 1990 + 1991 and leaf area (r =
0.40) or leaf biomass (r = 0.40). A similar correlation value
was observed between flowering and stem diameter, because
both leaf area and leaf biomass were highly correlated with
basal diameter (r = 0.96--0.97). Correlations between numbers
of clusters per tree and height and diameter growth the year the
flowers were initiated (1991) were low, but were indicative of
a stronger relationship between flower production and diameter growth (r = 0.18) than between flower production and
height growth (r = −0.06).
Although some families were more likely to flower than
others, individual trees did not exhibit a tendency to produce a
consistently high or low number of flowers (or to alternate high
and low flower production). There were also differences
among families in their flowering response to irrigation regime. In 1991 at the 2-m spacing, the percentage of trees with
female flowers ranged from 2.8% for John’s River II, #8 in the
high-irrigation regime to 63.6% for Nisqually Delta, #402 in
the low-irrigation regime (Figure 3). The ratio of the percentage of trees flowering in the low-irrigation treatment to the
percentage flowering in the high-irrigation treatment ranged
by family from 0.5 to 8.5, with an overall mean of 2.05.
Differences between irrigation regimes in percentage of trees
with female strobili was less in 1992 than in 1991 (range for
Figure 3. Percentage of trees in 2-m plots with 1991 female flowers by
irrigation regime and family. Dashed line indicates equivalence of
irrigation regimes. The greater the vertical distance from the line, the
more a family differed in flowering response between irrigation regimes. Letters identify families (see Table 1).
FLOWERING AND GROWTH IN YOUNG ALNUS RUBRA
1992 of 1.22 to 2.00, range for same families in 1991 of 1.3 to
6.5) (Figure 4); however, the 1992 ratio of the mean number of
strobili clusters per tree between irrigation regimes varied from
1.0 to 4.7 among families, indicating that family differences in
flowering response to irrigation were still being expressed.
Discussion
Flower initiation occurs during the summer (Furlow 1979),
flowering takes place the following February or March, and
seeds are ripe in the fall. Mature female strobili are woody and
cone-like in appearance, and remain intact and attached to the
plant during seed dispersal and for a time after dispersal is
completed. Previous-year female strobili were observed on
some trees in spring 1990, indicating that female flower initiation occurred in 1988 when the trees were 3 years old from
seed (2 years since planting). At the widest spacing, approximately half of the trees in seven families produced female
strobili in 1992, indicating female flower initiation in 1991
(i.e., 6 years from seed). These results are in agreement with
Stettler’s (1978) observation that A. rubra begins to flower at
age 3 to 4 years for individual trees and age 6 to 8 years for
most dominant trees in a stand. Several species of Betula that
naturally flower at young ages have been induced to flower
sooner when grown continuously, suggesting that Betula must
attain a minimum size before flowering (Longman 1984).
Thus, it may be possible to accelerate flowering in A. rubra
with growth-promoting treatments similar to those used with
Betula.
Crown size or volume is an important factor affecting flowering and seed production. Treatments such as wide initial
spacing or thinning are known to increase flowering or seed
production because they increase crown size and promote tree
vigor (Matthews 1963). We found positive correlations between flowering and variables that quantify crown size (e.g.,
branch biomass and leaf area). We conclude that variables that
quantify crown size are better predictors of flowering than
stem diameter if trees are growing in stands with considerable
variation in age, stocking or site quality.
Although N fertilization promotes flowering (Ross and
Figure 4. Mean number of 1992 female strobili clusters per tree in 2-m
plots by irrigation regime and family. Line and symbols as in Figure 3.
431
Pharis 1985), we did not include an N fertilization treatment in
this study because root nodules on A. rubra fix atmospheric N.
The P fertilization treatment was chosen because of its potential to increase tree growth rather than to enhance flowering;
however, the P fertilization did not increase 5-year height or
diameter growth. The lack of response of most flowering
variables to P fertilization is consistent with suggestions by
Matthews (1963) and Sedgley and Griffin (1989) that the
primary purpose of applying nutrients other than N to enhance
flowering or seed production should be to correct nutrient
deficiencies. Flowering of A. rubra may be responsive to P
fertilization on P-deficient sites because growth of the species
is strongly linked to P status (Radwan and DeBell 1994).
Although long-term seed production is enhanced by maintaining tree vigor, it has been suggested that flower initiation
is promoted by conditions that restrict shoot growth or cause
stress during the period when flower initiation can occur (Ross
and Pharis 1985, Owens 1991). Vegetative growth in Alnus is
primarily monopodial (the apical bud is a persistent leader, and
new branches arise laterally below the apex; Swartz 1971);
however, shoots that produce flowers exhibit sympodial
growth (the terminal bud withers, floral inflorescences are in
terminal and subterminal positions, and the main axis is made
up of a series of lateral branches). Additional primary shoot
extension does not occur after a reproductive structure has
been determined. Flower initiation probably occurs in late
June or early July for both A. rubra and A. glutinosa (L.)
Gaertn. (McVean 1955, Brown 1986, Ager et al. 1994); however, anatomical studies have not been reported. Species such
as Alnus with prolonged periods of shoot growth may have
greater leeway in the timing of floral initiation than species
with short periods of shoot extension (Longman 1985). However, if floral initiation in Alnus only occurs under certain
environmental conditions, it is possible that the extended period of shoot growth in the high-irrigation regime reduced or
even precluded floral initiation in some trees. The importance
of differences in genetic control of timing of flowering is
indicated by the study of O’Reilly and Owens (1988) showing
that initiation of seed-cone buds occurs on different dates in
different provenances of Pinus contorta Dougl. ex Loud.
We postulate that the increase in the percentage of trees
flowering each year was due to increases in the number of trees
attaining a minimum size that permits flowering (cf. Longman
1984). It is possible that differences in weather from year to
year also influenced flowering; however, in most tree species,
abundant seed crops do not occur in consecutive years even if
environmental conditions are favorable. Owens (1991) suggested that heavy seed crops alter endogenous conditions and
thereby inhibit reproductive bud development. We found no
indication that occurrence or relative number of flowers in one
year affected the occurrence or relative number of flowers in
the succeeding year. This finding contrasts with the report by
LaBastide and Vredenburch (1970) that seed crops for mature
A. glutinosa follow an annually alternating pattern. The number of flowers produced by the young trees in our study was
substantially lower than that observed for older trees in the
area; thus, it is possible that alternating years of high and low
432
HARRINGTON AND DEBELL
flower production could occur when these trees begin producing large seed crops. No long-term records of flowering or seed
production in A. rubra are available (see Harrington et al. 1994
for summary of short-term records). McVean (1955) concluded that the size of seed crops of A. glutinosa could vary
substantially from year to year, but that ‘‘boom-and-bust’’ patterns of seed production were not typical.
Genetic differences in flowering or seed production have
been observed in many species (Sedgley and Griffin 1989). We
observed differences among families in their propensity to
flower that could not be attributed solely to differences in mean
tree size (cf. Figure 2); however, we used too few members per
family in each treatment combination to test for differences
among families in flowering. Even though family differences
in flowering were evident, overall treatment responses were
clear, indicating that the results should be generally applicable
to other genotypes.
We conclude that if A. rubra stands are being managed for
seed production, the trees should be widely spaced: ≥ 2 m at
the time of the first seed crop and wider spacings as trees get
larger. Because crown recession and competition-related mortality occur very rapidly in dense stands of A. rubra, use of
spacings narrower than 2 m at planting is only feasible if
roguing is done in the first 2 to 3 years. Establishment on dry
sites may be ideal if supplemental irrigation is minimal during
June and July to promote flowering and is limited to amounts
needed to maintain long-term survival and health during the
rest of the year.
Acknowledgments
This research was supported in part by funding from US Dept. of
Energy, Biofuels Feedstock Development Program, Interagency Research Agreement DE-AI05-810R20914. We thank J. Hawks, S.
Bailey and M. Paschke for assistance in conducting the study and
personnel at the Washington State Department of Natural Resources,
Meridian Seed Orchard for their cooperation.
References
Ager, A., Y. Tanaka and J. McGrath. 1994. Biology, ecology, and
utilization of red alder seed. In The Biology and Management of
Red Alder. Eds. D.E. Hibbs, D.S. DeBell and R.F. Tarrant. Oregon
State Univ. Press, Corvallis, OR, pp 159--169.
Brown, S.M. 1986. Sexual allocation patterns in red alder (Alnus rubra
Bong.) along three elevational transects. M.S. Thesis. Univ. Washington, 242 p.
DeBell, D.S., C.A. Harrington and G.W. Clendenen. 1991. Increasing
the productivity of biomass plantations of alder and cottonwood in
the Pacific Northwest. USDA For. Serv., Pac. NW Res. Stn. and US
Dept. Energy, Woody Crops Program, Interagency Agreement DEAI05-810R20914, Annu. Tech. Rep., 60 p.
DeBell, D.S. and P.A. Giordano. 1994. Growth patterns of red alder.
In The Biology and Management of Red Alder. Eds. D.E. Hibbs,
D.S. DeBell and R.F. Tarrant. Oregon State Univ. Press, Corvallis,
OR, pp 116--130.
Furlow, J.J. 1979. The systematics of the American species of Alnus
(Betulaceae). Rhodora 81:1--21.
Harrington, C.A., J.C. Zasada and E.A. Allen. 1994. Biology of red
alder (Alnus rubra Bong.). In The Biology and Management of Red
Alder. Eds. D.E. Hibbs, D.S. DeBell and R.F. Tarrant. Oregon State
Univ. Press, Corvallis, OR, pp 3--22.
Hibbs, D.E. and A.A. Ager. 1989. Red alder: guidelines for seed
collection, handling, and storage. For. Res. Lab Spec. Pub. 18, Col.
For., Oregon State Univ. Press, Corvallis, OR, 6 p.
Hibbs, D.E., D.S. DeBell and R.F. Tarrant. 1994. Biology and management of red alder. Oregon State Univ. Press, Corvallis, OR, 256 p.
LaBastide, J.G.A. and C.L.H. van Vredenburch. 1970. The influence
of weather conditions on the seed production of some forest trees in
the Netherlands. Mededeling, Stichting Bosbouwproefstation ‘De
Dorschkamp,’ Wageningen, No. 102, 12 p.
Longman, K.A. 1984. Physiological studies in birch. Proc. Roy. Soc.
Edinburgh 85B:97--113.
Longman, K.A. 1985. Variability in flower initiation in forest trees. In
Attributes of Trees as Crop Plants. Eds. M.G.R. Cannell and J.E.
Jackson. Inst. Terrestrial Ecol., Nat. Environ. Res. Council, Huntingdon, U.K., pp 398--408.
Matthews, J.D. 1963. Factors affecting the production of seed by forest
trees. For. Abstr. 24:1--13.
McVean, D.N. 1955. Ecology of Alnus glutinosa (L.) Gaertn. I. Fruit
formation. J. Ecol. 44:219--222.
O’Reilly, C. and J.N. Owens. 1988. Reproductive growth and development in seven provenances of lodgepole pine. Can. J. For. Res.
18:43--53.
Owens, J.N. 1991. Flowering and seed set. In Physiology of Trees. Ed.
A.S. Raghavendra. John Wiley and Sons Inc., New York, pp 247-271.
Radwan, M.S. and D.S. DeBell. 1994. Fertilization and nutrition of red
alder. In The Biology and Management of Red Alder. Eds. D.E.
Hibbs, D.S. DeBell and R.F. Tarrant. Oregon State Univ. Press,
Corvallis, OR, pp 216--228.
Ross, S.D. and R.P. Pharis. 1985. Promotion of flowering in tree corps:
different mechanisms and techniques, with special reference to
conifers. In Attributes of Trees as Crop Plants. Eds. M.G.R. Cannell
and J.E. Jackson. Inst. Terrestrial Ecol., Nat. Environ. Res. Council,
Huntingdon, U.K., pp 383--397.
Sedgley, M. and A.R. Griffin. 1989. Sexual reproduction of tree crops.
Academic Press, London, 378 p.
Snedecor, G.W. and W.G. Cochran. 1980. Statistical methods, 7th Edn.
The Iowa State University Press, Ames, Iowa, pp 325--329.
Stettler, R.F. 1978. Biological aspects of red alder pertinent to potential breeding programs. In Utilization and Management of Alder.
Eds. D.G. Briggs, D.S. DeBell and W.A. Atkinson. USDA For.
Serv., Pac. NW For. and Range Exp. Stn., Portland, OR, Gen. Tech.
Rep. PNW-70, pp 209--222.
Swartz, D. 1971. Collegiate dictionary of botany. Ronald Press Co.,
New York, 520 p.
US Department of Commerce Weather Bureau. 1961. Local climatological data with comparative data, 1960, Olympia, Washington.
US Dept. Com., Weather Bureau, Asheville, NC, 4 p.