Plant Ecology 144: 93–101, 1999.

© 1999 Kluwer Academic Publishers. Printed in the Netherlands.

93

Differential responses to nitrogen fertilization in native shrubs and exotic

annuals common to mediterranean coastal sage scrub of California

Pamela E. Padgett & Edith B. Allen

Department of Botany and Plant Sciences, University of California, Riverside, CA 92521-0124, USA
(E-mail: ppadgett@mail.ucr.edu)

Received 16 February 1998; accepted in revised form 15 February 1999

Key words: Avena, Brassica, Bromus, Luxury consumption, N deposition, Nitrophilous

Abstract
This study examined the growth responses of exotic annuals and native shrubs to elevated N levels to test the
hypothesis that increased N availability favors nitrophilous annuals over the slower-growing shrubs. The vege-
tation structure of the coastal sage scrub ecosystems in southern California is shifting from shrubland to annual
grasslands. Over the last 30 years large tracts of wildlands, particularly those adjacent to urban centers, have lost
significant native shrub cover, which has been replaced by exotic annuals native to the Mediterranean Basin. During
this same time, air pollution has led to increased terrestrial eutrophication by atmospheric deposition. Changes in
vegetation are often the result of changes in resource availability. The results of our experiments showed the three
native shrubs tested to be more nitrophilous than the three annuals tested, which contrasts with most models of
perennial species’ adaptation to stressful environments. Under greenhouse conditions the annual grasses exhibited
yield depression at the highest N treatments of 80 µg g−1 in soil. The three shrub species evaluated continued
to increase shoot biomass at 80 µg g−1 N in soil. The grasses also exhibited increased tissue N concentrations
with increased soil N in contrast with the shrubs where there was little difference in tissue N concentrations with
increasing availability. Although the differential yield responses to elevated N do not explain the success of the
annual vegetation in replacing shrubs, the inability of the shrubs to regulate growth under elevated N levels may
explain the poor survival of mature individuals.

Introduction impoverishment of species diversity, particularly in

areas adjacent to urban development (Freudenberger
Changes in vegetation structure are often related to et al. 1987; Minnich & Dezzani 1998). At the same
changes in resource availability (McLendon & Re- time, N deposition from atmospheric pollution has
dente 1991; Tilman & Olff 1991; Tilman 1993; Keeley increased terrestrial, inorganic N loads. Soil NO−3 con-
& Swift 1995). It is widely believed that shifts in centrations under high deposition conditions has been
vegetation composition in response to shifts in nutri- measured as high as 90 µg N g−1 soil during the sum-
ent resources are the result of differential growth or mer dormant period, as compared with 1 to 2 µg N g−1
plasticity among the plant species present (Westman (as NO− 3 ) in soils collected from cleaner locations at
1981b; Tilman 1987; McLendon & Redente 1991). the same time of the year (Allen et al. 1997; Padgett
Species that are best able to grow under the new con- et al. 1999).
ditions tend to replace the less adapted species. The Other studies suggest that the species best able to
coastal sage scrub (CSS) plant community of south- increase biomass in response to increased N sources
ern California appears to be undergoing a shift from (i.e., nitrophilous) will become dominant under higher
native shrubby vegetation to exotic annual grassland. N conditions. Because the N response of the species
The encroachment of exotic annuals in CSS corre- native to CSS is unknown, this study was undertaken
sponds to serious losses in native shrub densities and to determine whether differential N responses could
94

explain the shift in vegetation structure by impacting sources. In the CSS ecosystems of southern California
early establishment of seedlings. a similar process seems to be occurring.
The physiological basis for plant responses to Coastal sage scrub is a low-productivity ecosys-
changes in soil nutrient status is poorly understood. tem native to the coastal foothills and inland valleys
In general, increasing the availability of a limiting soil of southern California (Westman 1981b). Southern
nutrient or other resource results in an increase in bio- California’s Mediterranean climate limits the rain-
mass. But the specific growth responses vary broadly fall to the winter months of October through March.
across the plant kingdom (Chapin 1980; Chapin et al. To cope with the 6-month annual drought, most of
1986). Some species, particularly the ruderal and the shrub species have adopted a drought deciduous,
annual species, exhibit nitrophilous behavior when ex- summer-dormant habit. Although some work on the
posed to increasing N availability. In these species water relations of these species has been conducted
biomass may more than double at the higher treat- (da Silva & Bartolome 1984; Davis & Mooney 1985;
ment levels (Garnier et al. 1989; Muller & Garnier D’Antonio & Mahall 1991; Eliason & Allen, 1997),
1990). Among the perennial species, the individual little is known about the nutrient requirements or N
growth responses tend to correspond to the native fer- responses of the CSS native species. This study was
tility (Grime 1979; Chapin 1980). Species adapted to undertaken to evaluate the early growth responses of
low-fertility ecosystems tend to be slow growing and three native shrubs compared to three exotic annual
exhibit limited responses to increased N supply. These seedlings to test the hypothesis that the success of
species generally absorb and store the nutrients rather the invasive annuals can be explained by N-enhanced
than synthesize new tissues in response to increased growth.
availability.
The magnitude of nitrophilly exhibited by individ-
ual plant species correlates well with survival in spe- Materials and methods
cific ecosystems (Grime & Hunt 1975; Tilman 1987).
Where N availability is abundant, the rapidly grow- Seed source
ing species tend to out compete the slower-growing
Seeds for the native shrubs Artemisia californica
species for light, water and nutrients. Where N avail-
Less., Eriogonum fasciculatum Benth. and Encelia
ability is low, the slower-growing species are better
farinosa Torrey & Gray were collected from stable
able to take advantage of flushes in availability by
stands of mature shrubs. Seeds were stored without
regulating growth and storing reserves for periods of
cleaning at 5 ◦ C but were separated from chaff just
scarcity. Thus, the slower-growing plants are better
prior to planting. All seeds were no older than one
able to persist under impoverished conditions but do
growing season. Seeds of the exotic annuals Bromus
not survive under nutrient-rich conditions. And the
rubens L. and Brassica geniculata L. were harvested
rapidly growing species do not grow well under defi-
from a highly weedy site. The fruit structures were left
cient conditions but flourish where resources are high.
intact for storage at 5 ◦ C. Seeds for the annual Avena
A natural or anthropogenic shift in the inherent fertil-
fatua L. were purchased from S & S seed (Carpinteria,
ity of an ecosystem is often accompanied by a shift
CA, USA). Genera and species names are as identified
in plant community structure (Westman, 1981a, b;
in Munz & Keck (1959).
D’Antonio & Vitousek 1992; Keeley & Swift 1995).
Several studies have linked N deposition to
Potting media and protocols
changes in the composition of shrub and grassland
plant community (Heil & Bobbink 1993; Pearson An artificial potting mix ‘UC mix #3’ (75% fine
& Stewart 1993; Dueck & Elderson 1992). Re- quartz sand, 25% ground peat moss), was used. Ear-
search conducted in the nutrient-poor heathlands and lier attempts to use native soils collected for field
chalk grasslands of the Netherlands have shown that sites were abandoned because of difficulties in regu-
increased N availability correlates with significant lating nitrogen concentrations. Seven hundred grams
changes in species composition (Bobbink & Willems of steam-sterilized UC mix were used for each pot.
1987; Bobbink 1991). The most common interpreta- Six replicate pots (6.4 × 25 cm ‘Deepots’, Stuewe
tion is that the additional N resources have enabled & Sons, Corvallis, OR, USA) were established for
the nitrophilous grass species to out-compete the low- each treatment. Pots were filled with potting mix and
nutrient-adapted shrubs and forbs for other soil re- leached with approximately 1 l of distilled, deionized
 95
Table 1. Mineral nutrient composition of coast sage scrub soil
solution evaluated during the spring growing season and com-
nary studies that indicated only small differences in
position of nutrient solution used for pot studies. Soil solution yield responses of the annuals at concentrations above
concentration was determined by saturated paste extract. Data 80 µg g−1 . These concentrations also reflected the
shown are the Means of 5 replications. The nutrient solutions range of soil inorganic N measured under polluted
were added to the potting medium before and during the growth
period depending on seedling growth rate. No erogenous Mo,
conditions in the field.
Ni or was added. Final solution pH was 6.5. Background N in the potting medium was approx-
imately 2 µg g−1 N as NH+ −
4 . No NO3 was detected
Soil solution Nutrient solution Specific after the first leaching. The NH+ 4 was assumed to
Element (mM) (mM) compound be derived from mineralization of the peat moss and
Ca 0.63 1.2 CaCO3 appeared to be firmly bound and largely inaccessible
Mg 0.32 0.6 MgO to the seedlings. In preliminary experiments where
Na 0.53 1 NaOH seedlings were treated with the nutrient solution only,
K 0.07 0.14 KCl without additional N, seedlings did not survive for
Cl 0.15 3 HCl more than 2 or 3 weeks. Therefore, the lowest N
S 0.06 1.2 MgSO4 treatment was maintained at 2 µg g−1 by exogenous
P 0.0008 0.16 KH2 PO4 application. This concentration provided just enough
Mn n/a 0.0001 MnSO4 N for seedlings to survive the 3-month experiments.
Zn n/a 0.001 ZnCl2 Soils were sampled every 1 to 2 weeks depending
Cu n/a 0.0001 CuSO4 on the size of the seedlings. One or two 5 × 100 mm
B n/a 0.003 H3 BO3 cores were removed from each pot and analyzed for
5 mg l−1
Fe n/a Fe EDTA
NO− +
3 and NH4 . When soil concentrations fell below
−1
10 µg g of the targeted concentrations, N solutions
were added to re-establish soil concentrations.

(DDI) water prior to planting. Field capacity was mea- Planting and harvesting
sured, established at 30% (240 g water pot−1 ) and
used to calculate soil solution volume for application Pots were planted with 10 seeds each. Final germi-
of nutrient solutions. nation was recorded at 10 to 14 days after seeding.
To duplicate the chemical and nutrient conditions Pots were thinned to one seedling per pot, and one
of CSS, the mineral nutrient content of native soils pot per treatment was maintained with no seedlings
was analyzed by saturated paste extractions (Soon & for evaluation of soil- versus plant-mediated changes
Warren 1993). A multiple-nutrient solution (excluding in N characteristics. Seedlings of all species were
N) was developed to mimic natural spring growing- harvested 3 months after initiation. The experimental
season soil solutions (Table 1). Pots were fertilized duration was based on time to flower for the annuals
with 125 ml pot−1 double-strength multiple-nutrient and avoidance of pot-bound conditions for the shrubs.
solutions (which represented half the volume of water These conditions were determined in preliminary ex-
held in the pots at field capacity) just prior to planting periments.
and at 1- to 4-week intervals depending upon seedling At the conclusion of the experiments the intact soil
size and growth rate. Soluble phosphorus was periodi- mass was separated from roots by soaking in water.
cally monitored to maintain 2 µg g−1 KCl extractable Even with care, species with fragile roots, especially
phosphorus as determined by continuous flow analysis A. californica lost fine root mass, so root weight data
(O’Halloran 1993). were not complete for all species but can be compared
across treatments within a species. Intact plants were
N treatments oven dried at 60 ◦ C for 48 h, separated into roots and
shoots and weighed.
Nitrogen was applied as solutions of NH4 Cl (5.4 g l−1 )
or KNO3 (10 g l−1 ) to achieve final soil N concen- Tissue and soil analysis
trations of 2, 20, 40 and 80 µg g−1 . Following N
applications, pots were watered with approximately Dried leaves were separated from stems and ground in
100 ml DDI water to distribute solubilized N. Soil a ball mill to a fine powder. Approximately 10 mg of
concentrations were chosen on the basis of prelimi- ground tissue was weighed into tin capsules, and %C
96

and %N was determined by flash combustion chro- no specific trends. Results of the 2-µg g−1 treatments
matography (Carlo-Erba Instruments, Fisons, Dear- compared directly with those of the 80-µg g−1 treat-
born, MI, USA). ments showed that fertilization with NO− 3 resulted in
Soil NO− +
3 and NH4 content was determined by 1-g significantly (P < 0.05) increased R:S for A. cal-
extraction in 1 M KCl by standard technique (Maynard ifornica, but NH+ 4 fertilization caused a significant
& Kalra 1993). Soil samples were weighed into plastic decrease. The R:S for both E. farinosa and E. fascic-
centrifuge tubes, and 10 ml KCl was added. The soil ulatum was highly variable and yielded no discernible
slurries were either shaken on a wrist shaker for 30 min trends.
or overnight on a reciprocal shaker. Soil was separated
from the extractant by centrifugation at 2000 × g for Relative yield responses
10 min, and 2 to 3 ml of the extractant was transferred
into autosampler cups and stored at −20 ◦ C until an- The difference in relative yield response to NO− 3 com-
alyzed by continuous flow analyzer. Solutions were pared with NH+ 4 fertilization was not significant for
simultaneously analyzed for NH+ 4 by the indophenol
either exotic annuals (Figure 1A, 1C) or native shrubs
− (Figure 1B, 1D). However, the pattern of response to
blue procedure and NO3 by the cadmium reduction
procedure (Maynard and Kalra 1993) increasing N availability was distinctly different be-
tween the grasses and the shrubs, with B. geniculata
Statistical analysis responding more like the shrubs (Figure 1). With both
of the N source treatments A. fatua and B. rubens
Yield and tissue N content were analyzed separately reached maximum biomass at 40 µg g−1 N. Yield was
for each species by one-way ANOVA and t-tests using depressed with the 80-µg g−1 treatment (Figure 1A,
SigmaStat, version 2.0 by Jandel Scientific Software 1C) for these two species but not for the shrubs or
(San Rafael, CA USA). B. geniculata (Figure 1B, 1D). For all three shrubs,
the 40-µg g−1 treatment resulted in approximately 70
to 80% of the maximum yield.
Results
Shoot N content
Biomass
The predisposition of these species to accumulate N
After 3 months of growth, a trend toward larger plants in leaf tissue differed between life forms (Figure 2).
under NH+ −
4 fertilization as compared to NO3 fertil- All of the annuals had increased tissue N with increas-
ization was noted for A. fatua, A. californica and ing application rate (Figure 2A, 2C). As with the yield
E. fasciculatum; the opposite trend of smaller plants rate, there was no difference in tissue N between the
under NH+ −
4 as compared with NO3 fertilization was NO− +
3 and NH4 treatments. For the shrubs, the pattern
observed for B. rubens, B. geniculata and E. farinosa of tissue accumulation differed from that observed in
(Table 2). Since the yield response to the individual N the annuals. Artemisia californica contained signifi-
sources was significant only for A. fatua (P < 0.05) cantly (P < 0.01) more N than any of the other shrubs
and there was no consistency between native shrub and and accumulated relatively large quantities of N in the
exotic annual species, it does not appear that pot cul- N-starved seedlings (Figure 2B, 2D). Encelia farinosa
tures revealed a clear species-specific preference for tended to have the lowest percent leaf N, but the dif-
one N source over another, nor can any generalizations ferences between E. farinosa and E. fasciculatum were
be drawn regarding N source and the origin of these six not significant. All of the shrub species indicated little
plant species. propensity to accumulate large quantities of tissue N
Changes in root:shoot ratio (R:S) often accom- under high fertility conditions, especially as compared
pany changes in N availability (Levin et al. 1989). with the annuals. A small increase in tissue N con-
This predisposition was demonstrated by the R:S re- centration was noted for E. fasciculatum at the highest
sponses of A. fatua (Table 2). Changes in R:S were NO− +
3 and NH4 concentrations and for E. farinosa at
not so clear for the other species, however. Ammo- 80 µg g−1 N as NO− 3.
nium fertilization resulted in a trend of decreased R:S
in B. rubens but not in the NO− 3 treatments. The R:S
in B. geniculata was highly variable and also indicated
 97
Table 2. Biomass after 3 months of growth. Seedlings were harvested and oven dried. Data shown are the means with SE indicated by parentheses
(n = 5). All species demonstrated significant (P < 0.01) biomass differences with increasing nitrogen application. Differences between NH+ 4
and NO− 3 treatments were significant (P < 0.05) only for Avena fatua.

NO−3 NH+4
2 µg g−1 20 µg g−1 40 µg g−1 80 µg g−1 2 µg g−1 20 µg g−1 40 µg g−1 80 µg g−1

Avena fatua
Shoots 0.17 0.91 1.35 0.99 0.03 0.85 1.57 1.37
(0.05) (0.16) (0.26) (0.21) (0.02) (0.15) (0.37) (0.34)
Root:Shoot 13.30 5.28 8.58 4.99 13.30 4.80 4.85 3.45
(2.74) (0.98) (1.40) (1.09) (3.55) (0.50) (1.05) (0.73)

Bromus rubens
Shoots 0.25 0.64 1.68 1.37 0.19 0.56 1.39 1.15
(0.03) (0.06) (0.34) (0.26) (0.03) (0.11) (0.25) (0.07)
Root:Shoot 1.78 1.53 1.42 1.59 1.07 1.12 0.99 0.83
(0.39) (0.34) (0.48) (0.34) (0.20) (0.18) (0.19) (0.19)

Brassica geniculata
Shoots 0.01 0.39 0.89 1.32 0.01 0.23 0.42 0.81
(0.00) (0.06) (0.07) (0.10) (0.00) (0.06) (0.26) (0.21)
Root:Shoot 0.78 1.81 1.29 1.21 0.54 1.58 1.12 1.16
(0.45) (0.17) (0.15) (0.23) (0.15) (0.21) (0.40) (0.90)

Artemisia californica
Shoots 0.07 0.67 0.96 1.18 0.08 0.85 1.25 1.73
(0.01) (0.12) (0.06) (0.21) (0.02) (0.16) (0.14) (0.18)
Root:Shoot 0.12 0.41 0.37 0.36 0.48 0.24 0.19 0.16
(0.02) (0.04) (0.03) (0.07) (0.14) (0.03) (0.05) (0.04)

Encelia fatinosa
Shoots 0.05 0.20 0.40 0.45 0.05 0.17 0.17 0.39
(0.01) (0.03) (0.02) (0.06) (0.01) (0.06) (0.02) (0.07)
Root:Shoot 1.58 1.85 2.16 2.82 1.18 1.90 1.64 1.94
(0.27) (0.27) (0.26) (0.49) (0.16) (0.51) (0.20) (0.08)

Eriogonum fasiculatum
Shoots 0.05 1.28 1.45 1.86 0.55 0.93 1.75 2.43
(0.03) (0.38) (0.26) (0.28) nd (0.18) (0.39) (0.29)
Root:Shoot 3.66 6.17 2.05 0.99 0.26 2.09 1.15 2.09
(3.29) (4.28) (0.36) (0.17) nd (0.50) (0.20) (3.34)

Discussion 2.5-fold increases in biomass between the 20-µg g−1

treatment and the treatment resulting in the highest
The results of these experiments indicate that as biomass accumulation. However, the grasses attained
seedlings, the three CSS shrub species showed a their highest yields at 40 µg g−1 N, whereas the shrubs
greater relative yield response to increased N avail- and B. geniculata grew significantly larger with the
ability than the three exotic annuals. This contrasted 80-µg g−1 treatment.
with predictions for N response based on native fer- No difference in the relative yield response pat-
tility, productivity and species life form, in which it terns between N as NO− +
3 or NH4 was detected. And in
was expected that the shrubs would engage in uptake only one case was the absolute biomass significantly
and storage of N rather than increased tissue synthe- different between the two sources. Determination of
sis. All of the six species tested demonstrated 1.5- to N preference was important to understanding plant
98

Figure 1. Relative shoot yield in response to two N sources, NO− −

3 and NH4 . Relative yield was calculated as a percentage of the biomass from
the highest yielding treatment. The responses were significantly different (P < 0.01) within but not between treatments.

response to deposition loads because in southern Cal- in this study because of the noted difference in bio-
ifornia more than 70% of the N deposited from pollu- mass allocation patterns to roots or shoots between
tion is as NO− 3 (Bytnerowicz & Fenn 1996; Allen et al. monocots and dicots (Lambers & Poorter 1992).
1997; Padgett et al. 1999). This is in contrast to many The significantly higher N content of A. califor-
of the European studies where most of the deposited N nica leaves is probably related to leaf morphology and
is derived from agricultural NH+4 (Bobbink 1991). For anatomy. The leaves are thread-like, and microscopic
these six species, however, the N source in a pot study investigation suggests that they consist of one layer
was of little importance. of epidermal cells, a row of palisade parenchyma and
Changes in R:S that frequently accompany a vascular bundle (Padgett unpubl. obser.). There is
changes in N availability are often used to explain the very little schlerophilous tissue, spongy parenchyma
outcome of interspecific competition (Garnier et al. or fibers. Although it might be tempting to conclude
1989; Gutschick 1993; Van der Werf et al. 1993). The that this species is particularly N inefficient because of
results of this study did not indicate any clear trends the high tissue content (sensu Killingbeck & Whitford
regarding root responses. For example, the decrease 1996), the N content is probably more related to the
in R:S with increased N by A. fatua suggest that at lack of cells and tissues devoted to structural mainte-
higher N rates, this species would be less compet- nance such as fibers. Thus, a greater proportion of the
itive below-ground because of reduced root surface dry weight is involved in physiological processes that
in relationship to shoot biomass. However, A. fatua would require nitrogenous compounds.
also demonstrated the highest R:S among the 6 species The apparent nitrophilous nature of A. califor-
under all N treatments. Comparisons between the an- nica, E. farinosa and E. fasciculatum observed in this
nuals and perennial shrubs is probably not appropriate study concur with observations by Gray & Schlesinger
 99

Figure 2. Mean N content (n = 5) in leaf tissue of six CSS species treated with four levels of NO− −
3 and NH4 . Bars = SE.

(1983) of the N response of Salvia leucophylla, an- obtained in our work had we used N concentrations
other semideciduous shrub native to CSS. Like A. cal- above 80 µg g−1 .
ifornica, E. farinosa and E. fasciculatum, S. leuco- The response of the CSS shrubs to increasing N
phylla demonstrated a linear growth response with is contrary to observations of other perennial species
increased N availability up to the highest treatment as compared to ruderal annuals (e.g., Chapin et al.
level, of 168 µg g−1 N. The N response of S. leuco- 1986; Chapin & Moilanen 1991). All plants do have
phylla differed substantially from that of a comparison some ability to respond to changes in resource levels
sclerophyllous evergreen shrub, Ceanothus megacar- either by regulating growth or by regulating nutrient
pus. This shrub is an evergreen nonleguminous N absorption (Glass et al. 1985; Aslam et al. 1993). In
fixer native to the physiographically similar chaparral these experiments analysis of the potting mix during
ecosystems. Maximum yield for C. megacarpus was the growth experiments indicated little difference in
achieved at 84 µg g−1 N, which was half the N uptake rate among the six species suggesting that the
concentration required for maximum yield of S. leu- differential yield response was not due to differences
cophylla. Although the Gray & Schlesinger study was in uptake rates among these species (data not shown).
conducted in sand culture using flowing nutrient solu- The apparent nitrophilous nature of the CSS shrubs
tions, the targeted concentrations of 21 to 168 µg g−1 might be expected to give them a competitive advan-
in solution were similar to the 2 to 80 µg g−1 in soil tage under fertile conditions such as those occurring
used in this study. The results of Gray & Schlesinger in areas of high N deposition. However, the field ob-
(1983) suggest that higher biomass would have been servations indicate that the native shrubs fare poorly
under high N deposition, particularly once grasses are
100

well established (Schultz 1996; Eliason & Allen 1997; This research was supported by USDA NRI 95-37101-
Minnich & Dezanni 1998). The exotic grasses are well 1700, NSF DEB-9408079 and DEB-9526564.
adapted to the semi-arid, Mediterranean ecosystems
of California, and once established their populations
are self-sustaining, making eradication very difficult References
(da Silva & Bartolome 1984; Welker et al. 1991).
Although differential yield responses between the life Allen, E. B., Padgett, P. E., Bytnerowicz, A. & Min-
nich, R. A. 1997. Nitrogen deposition effects on coastal
forms were clearly evident, at first glance they do not sage vegetation of southern California. 16 pp. In: Byt-
seem to explain the success of the invasive grasses nerowicz, A. & Arbaugh, M. (eds), Proceedings, Sympo-
and forbs and the loss of native shrubs during early sium on Air Pollution and Climate Change Effects on Forest
seedling establishment. Ecosystems. http://www.rfl.psw.fs.fed.us/pubs/psw-gtr-164/ full-
text/ allen/allen.html#anchor14735.
Other hypotheses for the success of exotic grasses Aslam, M., Travis, R. T. & Huffaker, R. C. 1993. Comparative in-
have been tested, including greater seedling size and duction of nitrate and nitrite uptake and reduction systems by
growth rate resulting in shading of adjacent shrubs, ambient nitrate and nitrite in intact roots of barley (Hordeum
below-ground competition for other nutrients or water vulgare L.) seedlings. Plant Physiol. 102: 811–819.
Bobbink, R. 1991. Effects of nutrient enrichment in Dutch chalk
(Eliason & Allen 1997), poor shrub seedling emer- grassland. J. Appl. Ecol. 28: 24–41.
gence under dense grass litter (Schultz 1996) and gen- Bobbink, R. & Willems, J. H. 1987. Increases in dominance of
eral depletion of shrub seed bank reserves following Brachypodium pinnatum (L.) Beauv. in chalk grasslands: a threat
disturbances. If N were the only variable in the com- to a species-rich ecosystem. Biol. Cons. 40: 301–314.
Bytnerowicz, A. & Fenn, M. E. 1996. Nitrogen deposition in
petition between invasive annuals and native shrubs, California forests: A review. Environ. Pollut. 92: 127–146.
the results of this study indicate that neither group Chapin F. S. III. 1980. The mineral nutrition of wild plants. Annu.
has a particular advantage in the face of increasing N Rev. Ecol. Syst. 11: 233–260.
Chapin, F. S., Vitousek, P. M. & Van Cleave, K. 1986. The nature of
availability due to N deposition.
nutrient limitation in plant communities. Am. Nat. 127: 48–58.
One other explanation yet to be extensively ex- Chapin, F. S. & Moilanen, L., 1991. Nutritional controls over ni-
plored is that poor regulation of growth results in trogen and phosphorus resorption from Alaskan birch leaves.
shortened life spans. Poor long-term survival of arid- Ecology 72: 709–715.
D’Antonio, C. M. & Mahall, B. E. 1991. Root profiles and compe-
adapted plants grown under horticultural conditions is tition between the invasive, exotic perennial, Carpobrotus edulis
common (Keator 1994). Because these shrubs appar- and two native shrub species in California coastal scrub. Am. J.
ently lack the ability to restrict growth in the presence Bot. 78: 885–894.
of exogenous N, under high deposition conditions, D’Antonio, C. M. & Vitousek, P. M. 1992. Biological invasions
by exotic grasses, the grass/fire cycle, and global change. Annu.
they may not receive or respond to environmental cues Rev. Ecol. Syst. 23: 63–87.
that should initiate preparation for dormancy. Absorp- da Silva, P. G. & Bartolome, J. W. 1984. Interaction between a
tion of N pollutants is thought to occur through foliar shrub Baccaris pilularis subsp consanguinea (Asteraceae) and
means in addition to the normal root pathway (Vose an annual grass Bromus mollis (Poaceae) in coastal California.
Madrono 31: 93–101.
& Swank 1990; Hanson & Garten 1992; Nussbaum Davis, S. D. & Mooney, H. A. 1985. Comparative water relations of
et al. 1993). If this were to occur in CSS species, fo- adjacent California shrub and grassland communities. Oecologia
liar deposition and subsequent absorption could cause 66: 522–529.
a bypass in the normal environmental cues resulting Dueck, T. A., Van der Eerden, L. J., Beemsterboer, B. & Elderson,
J. 1991. Nitrogen uptake and allocation by Calluna vulgaris (L.)
in continued growth when these species should be Hull and Deschampsia flexuosa (L.) Trin. exposed to 15 NH3 .
preparing for summer dormancy. This hypothesis is Acta Bot. Neerl. 40: 257–267.
consistent with the observed pattern of loss of mature Dueck, T. A. & Elderson, J. 1992. Influence of NH3 and SO2 on the
growth and competitive ability of Arnica montana L. and Viola
shrubs in the field. Work is ongoing to test this and
canina L. New Phytol. 122: 507–514.
other hypotheses to develop a generally applicable un- Eliason, S. A. & Allen, E. B. 1997. Competition as a mechanism
derstanding of arid and semi-arid ecosystem responses for exotic grass persistence following conversion from native
to N deposition. California shrubland. Restor. Ecol. (in press).
Freudenberger, D. O., Fish, B. E. & Keeley, J. E. 1987. Distribution
and stability of grasslands in Los Angeles Basin. Bull. South.
Calif. Acad. Sci. 86: 13–26.
Acknowledgements Garnier, E., Koch, G. W. & Mooney, H. A. 1989. Responses of wild
plants to nitrate availability. Oecologia 79: 542–550.
Glass, A. D. M., Thompson, R. G. & Bordeleau, L. 1985. Reg-
We wish to thank Sheila Kee, Samantha Traylor, Lisa ulation of NO− 13 NO− . Plant
3 influx in barley. Studies using 3
Jones and Elise Focht for their technical assistance. Physiol. 77: 379–381.
 101

Gray, J. T. & Schlesinger, W. H. 1983. Nutrient use by evergreen Munz, P. A. & Keck, D. D. 1959. A California flora. University of
and deciduous shrubs in southern California II. Experimental California Press, Berkeley, CA.
investigations of the relationship between growth and nitrogen Nussbaum, S., von Ballmoos, P., Gfeller, H., Schlunegger, U. P.,
uptake and nitrogen availability. J. Ecol. 71: 43–56. Fuhrer, J., Rhodes, D. & Brunold, C. 1993. Incorporation of
Grime, J. P. 1979. Plant Strategies and Vegetation Processes. John atmospheric 15 NO2 -nitrogen into free amino acids by Norway
Wiley & Sons, New York. spruce Picea abies (L.) Karst. Oecologia 94: 408–414.
Grime, J. P. & Hunt, R. 1975. Relative growth rate: Its range and O’Halloran, I. P. 1993. Total and organic phosphorus. Pp. 213–229.
adaptive significance in local flora. J. Ecol. 63: 393–422. In: Carter, M. R. (ed.), Soil Sampling and Methods of Analysis.
Gutschick, V. P. 1993. Nutrient-limited growth rates: roles of Lewis Publishers, Boca Raton, FL.
nutrient-use efficiency and of adaptations to increased uptake Padgett, P. E., Allen, E. B., Bytnerowicz & Minnich, R. A. changes
rate. J. Exp. Bot. 44: 41–51. in soil inorganic nitrogen as related to atmospheric nitrogenous
Hanson, P. J. & Garten, C. T. 1992. Deposition of H15 NO3 vapour pollutants in southern California. Atmos Environ. 33: 769–781.
to white oak, red maple and loblolly pine foliage: experimental Pearson, J. & Stewart, G. R. 1993. The deposition of atmospheric
observations and a generalized model. New Phytol. 122: 329– ammonia and its effects on plants. New Phytol. 125: 283–305.
337. Riggan, P. J., Lockwood, R. N. & Lopez, E. 1985. Deposition and
Heil, G. W. & Bobbink, R. 1993. ‘Calluna’, a simulation model for processing of airborne nitrogen pollutants in Mediterranean-type
evaluation of impacts of atmospheric nitrogen deposition on dry ecosystems in Southern California. Environ. Sci. Technol. 19:
heathlands. Ecol. Model. 68: 161–182. 781–789.
Keator, G., 1994. Complete Garden Guide to the Native Shrubs of Soon, Y. K. & Warren, C. J. 1993. Soil solutions. Pp. 147–159.
California. Pp. 3–13. Chronicle Books, San Francisco. In: Carter, M. R. (ed.), Soil Sampling and Methods of Analysis.
Keeley, J. E. & Swift, C. C. 1995. Biodiversity and ecosystem Lewis Publishers, Boca Raton, FL.
functioning in Mediterranean-climate California. Pp. 121–183. Schultz, G. 1996. Seedling establishment and competition in coastal
In: Davis, G. W & Richardson, D. M. (eds), Mediterranean-Type sage scrub and annual grassland. Thesis, University of Califor-
Ecosystems. The Function of Biodiversity. Springer-Verlag, New nia, Riverside.
York. Tilman, D. 1987. Secondary secession and the pattern of plant dom-
Killingbeck, K. T. & Whitford, W. G. 1996. High foliar nitrogen inance along experimental nitrogen gradients. Ecol. Monogr. 57:
in desert shrubs: an important ecosystem trait or defective desert 189–214.
doctrine? Ecology 77: 1728–1737. Tilman, D. 1993. Species richness of experimental productivity gra-
Lambers, H. & Poorter, H. 1992. Inherent variation in growth rate dients: how important is colonization limitation. Ecology 74:
between higher plants: A search for physiological causes and 2179–2191.
ecological consequences. Adv. Ecol. Res. 23:187–261. Tilman, D. & Olff, H. 1991. An experimental study of the effects
Levin, S. A., Mooney, H. A. & Field, C. 1989. The dependence of of pH and nitrogen on grassland vegetation. Acta Ecol. 12: 427–
plant root:shoot ratios on internal nitrogen concentration. Ann. 441.
Bot. 64: 71–75. Van der Werf, A., Van Nuenen, M., Visser, A. J. & Lambers,
Maynard, D. G. & Kalra, Y. P. 1993. Nitrate and exchangeable H. 1993. Contribution of physiological and morphological plant
ammonium nitrogen. Pp. 25–38. In: Carter, M. R. (ed.), Soil traits to a species’ competitive ability at high and low nitrogen
Sampling and Methods of Analysis. Lewis Publishers, Boca supply. Oecologia 94: 434–440.
Raton, FL. Vose, J. M. & Swank, W. T. 1990. Preliminary estimates of foliar
McLendon, T. & Rendente, E. F. 1991. Nitrogen and phosphorus absorption of 15 N-labeled nitric acid vapor (HNO3 ) by mature
effects on secondary secession dynamic on a semi-arid sagebrush eastern white pine (Pinus strobus). Can. J. For. Res. 20: 857–860.
site. Ecology 75: 2016–2024. Welker, J. M., Gordon, D. R. & Rice, K. J. 1991. Capture and allo-
Minnich, R. A. & Dezzani, R. J. (1998) Historical decline of coastal cation of nitrogen by Quercus douglasii seedlings in competition
sage scrub in the Riverside-Perris Plain, California. Western with annual and perennial grasses. Oecologia 87: 459–466.
Birds 29: 366–391. Westman, W. E. 1981a. Diversity relations and succession in Cali-
Muller, B. & Garnier, E. 1990. Components of relative growth rate fornian coastal sage scrub. Ecology 62: 170–184.
and sensitivity to nitrogen availability in annual and perennial Westman, W. E. 1981b. Factors influencing the distribution of
species of Bromus. Oecologia 84: 513–518. species of Californian coastal sage scrub. Ecology 62: 439–455.