Demography and life history of Arabis fecunda in Ravalli and Beaverhead Counties, Montana

:fW^~

>83. 123
411dLha
1994

Lesica* Peter

Demography and
life history of
Arabis fecunda in
Ravalli and
Beaverhead
Counties* Montana

^Krk,^4^

563.123
N lldLha

IS94

MONTANA STATE LIBRARY

3 0864 0010 3928 1

DEMOGRAPHY AND LIFE fflSTORY OF ARABIS FECUNDA
IN RAVALLI AND BEAVERHEAD COUNTIES, MONTANA

Prepared by:

Peter Lesica

929 Locust

Missoula, MT 59802

and

J. Stephen Shelly

Montana Natural Heritage Program

1515 E. 6th Ave.

Helena, MT 59620

Prepared for:

USDA Forest Service

Beaverhead National Forest

610 N. Montana

Dillon, MT 59725

TON

January 1994

P

s;^k^i

i {■

I ir\

i Vl\

© 1994 Montana Natural Heritage Program

This report should be cited as follows:

Lesica, P. and J. S. Shelly. 19941 Demography and life history of Arabis fecunda in Ravalli and Beaverhead counties, Montana.
Unpublished report to the Beaverhead National Forest Montana Natural Heritage Program, Helena. 29 pp.

Summary
We monitored individuals of Arabis fecunda over four consecutive years
at three sites in order to gain knowledge of this rare plant's life history.
Arabis fecunda is a short-lived perennial with high fecundity. Recruitment is
high as is mortality of juveniles. Plants flower by bolting or producing
axillary inflorescences. Bolting plants produced 2.5 times as many seeds,
matured earlier but had much higher mortality compared to axillary-flowering
plants. Seeds germinate readily without stratification. Seed dormancy is
induced by cold/dark conditions at some sites but not others.

Recruitment rate, survivorship, age at maturity and fecundity varied
significantly among sites. Much of this difference in life history traits was
due to differential bolting frequencies among the three sites. These results
suggest that life history traits are locally adapted and that adaptive genetic
differences may exist between populations.

Populations in the southern portion of A_^ fecunda' s range appear to be
stable and will be most sensitive to changes that cause a reduction in
recruitment. On the other hand, populations in the north may be declining and
should be most sensitive to declines in adult survivorship.

Introduction
Passage of the Federal Endangered Species Act of 1973 and subsequent
recognition of the value of conserving biotic diversity (Wilson 1988) have
resulted in many government agencies becoming active in species conservation.
Surveys to determine the location and size of populations of rare species are
being conducted on public lands throughout the west. These surveys are
necessary in any species conservation program; however, knowing the location
and size of populations at any one point in time is only the first step in a
long-term protection strategy. (Sutter 1986). Extinction is a process
requiring an understanding of population dynamics (Menges 1986). Periodic

i

inventories can detect trends but will do little to determine causality or
help generate predictive hypotheses (Palmer 1987). Long-term conservation
requires a knowledge of many life history parameters including fecundity,
recruitment, survivorship, age structure, and population flux. Demographic
monitoring techniques can provide information on factors regulating population
density and persistence (Palmer 1987). This information, in turn, provides an
essential basis for management decisions.

Arabis fecunda is a candidate for listing as a threatened or endangered
species by the U.S. Fish and Wildlife Service (USDI-FWS 1993), is considered
sensitive in Region One of the U.S. Forest Service, and is considered
threatened in Montana (Lesica and Shelly 1991). Little is known about the
life history and demography of Arabis fecunda populations. The purpose of
this study is to determine demographic patterns and variability for this rare
species and to use this knowledge to recommend appropriate management
strategies for conservation.

METHODS
The Species

Arabis fecunda Rollins is a rosette-forming, perennial in the Mustard
Family (Brassicaceae) . This recently described species (Rollins 1984) is
endemic to highly calcareous soils in the foothills of the Sapphire Range in
Ravalli County and in the Pioneer Range in Beaverhead and Silver Bow counties
in southwest Montana. Arabis fecunda plants flower in April and May, and
fruits mature in June and July. Flowering occurs in one of two ways: (1)
axillary flowering - 1 to many decumbent inflorescence stems develop from
axillary buds among the tightly clustered leaves of the rosette or (2) bolting
- a single inflorescence stem is produced from the terminal bud in the center
of the rosette. Bolting inflorescences are generally larger and leafier than
axillary inflorescences. An individual rosette may produce axillary
inflorescences for numerous years, while bolting rosettes always die. Some

•

rosettes are iteroparous, producing axillary inflorescences for 1-many years
before either dying or bolting and then dying. Others bolt once and are
essentially semelparous. Individuals may branch at the root crown to form
multi-rosette plants at any time during the life cycle. This is not
vegetative reproduction as individual rosettes from multi-rosette plants never
become independent plants. If only a portion of the rosettes in a multiple-
rosette plant bolt, the whole plant may or may not die.

Study Sites

We conducted our study at Charleys Gulch in Ravalli County and Lime
Gulch and Vipond Park in Beaverhead County, Montana. The Charleys Gulch site
is on a moderate southwest-facing slope, at 1525 m. At Hamilton, ca. 8 km
southwest and 300 m lower, mean temperatures for July and January are 19.4°
and -3.8° C respectively, and mean annual precipitation is 32 cm. Vegetation
surrounding the sites is foothills Aqropyron-Festuca grasslands with scattered
Pinus ponderosa Dougl. and Pseudotsuqa menziesii (Mirb. ) Franco. The Lime
Gulch site occurs on moderate east- and west-facing slopes above a small
drainage on the east side of the Pioneer Range at ca 1890 m. The Vipond Park
site is on a moderate south-facing slope at 2195 m at the north end of the
Pioneer Range. The two sites are separated from each other by ca. 32 km and
from the Ravalli County site by ca. 130 km. For Divide, at 1675 m and north
and east of the two sites, mean temperatures for July and January are 17.2°
and -7.2° C respectively, and mean annual precipitation is 31 cm. Vipond Park
is appreciably higher than the recording station, and thus likely experiences
colder temperatures and greater precipitation. Vegetation around Lime Gulch
is Juniperus /Cercocarpus woodland, while it is Artemisia-Festuca-Aqropyron
steppe at Vipond Park.

Soils at all sites are highly calcareous sandy loams derived from
outcrops of metamorphosed calc-silicates or limestone. These soils have a
tendency to slump on moderate to steep slopes. Vegetation at these sites is

sparse compared to surrounding grasslands and woodlands. Cryptogamic soil
crusts are coiranon at Charleys Gulch and Lime Gulch (Lesica and Shelly 1992a) .
Soils at Charleys Gulch have a lighter albedo than those at the Beaverhead
County sites.

Field methods

In 1987 we established two permanent transects, one of 5 and one of 12
contiguous 1-m- plots at Charleys Gulch. In 1989 we established two permanent
transects of 12 contiguous 1-m^ plots each at both Lime Gulch and Vipond Park.
Transects were located to be representative of the populations a whole. We
censused Arabis fecunda in 1988-93 at all three sites. Sampling was conducted
in late May at Charleys Gulch, mid-June at Lime Gulch and late June or early
July at Vipond Park. We chose these times because A_^ fecunda fruits were
mature or nearly so, but dispersal had not yet occurred. Plants smaller than
0.5 cm in diameter were not recorded because they could not be reliably
distinguished from other species.

Individual A^ fecunda plants were mapped and recorded following methods
outlined in Lesica (1987) and using the following life history stage
classification system:

Small (S) = single vegetative rosette < 2 cm in diameter
Juvenile (J) = single vegetative rosette > 2 cm in diameter
Multiple-rosette (M) = multiple vegetative rosettes
Reproductive (R) = plants producing 1-many inflorescences

In addition, for each reproductive plant we recorded the number of
inflorescences and the number of fruits matured. We recorded which plants
bolted in 1990-93.

A plants 's demographic properties are often more closely correlated with
size and life-history stage rather than age (Werner and Caswell 1977, Caswell
1989), although both may be important in predicting an individual's fate
(Young 1985). We chose these classes because they are correlated with age as
well as size and because they also represent a reasonable compromise between
having many categories with too few observations each and few categories with
many observations (Vandermeer 1978).

In each year we collected one fruit from the middle of the inflorescence
of each of 25 randomly chosen plants growing near the transects at each site.
We counted the number of mature or nearly mature seeds in each fruit to obtain
an estimate of seeds per fruit for each site.

In 1993 we collected one fruit each from 25 randomly selected axillary
flowering plants and 25 bolting plants from each site. After counting the
number of mature or developing seeds, those from Charleys Gulch and Vipond
ParJc were used in germination tests. Seeds from Lime Gulch were not mature
enough to be used. Seeds were stored dry at room temperature for 4 months
prior to the tests. Two treatments were tested: (1) warm/light - constant
20°C with 14 hours of constant light per day and (2) cold/dark - constant 5°C
in the dark. Seeds were placed on moist filter paper in petri dishes, 20
seeds from a single parent per dish with 6 bolting and 6 axillary flowering
replicates from each site for each treatment. The warm/light and cold/dark
treatments were given for 8 and 20 days respectively. Germinated seeds were
recognized by the radicle emerging at least 1 mm from the seed coat.

We estimated canopy cover to the nearest 5% of all vascular plants as
well as cover of rock, bare soil and basal vegetation in each plot (Daubenmire
1959) .

>?

Data analysis

Stage-structured transition matrix projection models summarize the way
in which survival, growth and reproduction at various life-history stages
interact to determine population growth (Caswell 1989, van Groenendael et al.
1988). Matrix projections assume fixed transition probabilities between
stages in a population through time (Lefkovitch 1965, Menges 1990). They
assume density-independent population growth and thus do not give an accurate
projection of long-term population future. Nonetheless, they can be used to
summarize short-term population dynamics or compare the dynamics of two
populations (Caswell 1989). One-year transition probabilities were estimated
as the number of plants in life-stage class i moving into class j. over the
course of one year divided by the number of plants in stage i at the beginning
of the year. This method assumes that an individual's transition depends only
on its life-stage class at the beginning of the period and is independent of
its transition the previous year. The equilibrium growth rate {k) is the
dominant eigenvalue of the transition matrix (Caswell 1989, Lefkovitch 1965).
X > 1.0 indicates population increase, while X < 1.0 indicates decrease. X
integrates the effects of survival, growth and fecundity of the different
life-history stages into a single parameter. Details on the construction and
use of matrix population models can be found in Caswell (1989) and Menges
(1990) .

Elasticity measures the relative change in the value of X in response to
changes in the value of a transition matrix element. Elasticity matrices
allow comparison of relative importance to population growth and fitness among
the various life history transitions (de Kroon et al. 1986). Elasticities sum
to unity and regions of the matrix may be summed to compare the importance of
growth and survival to recruitment (Caswell 1986) .

When the majority of seeds pass directly from production to germination
in less than one year, seeds should not appear as a separate stage in matrix

models (Caswell 1989, Silvertown et al. 1993). In most cases, the majority of
seeds probably germinate without a dormant period (see Results), and we have
used matrices with reproductive transition and recruitment columns combined to
calculate X. We calculated separate elasticities for reproductive transitions
and recruitment by dividing the reproductive+recruitment elasticities
proportionately between their two components.

We used the ratio of new recruits to survivors to compare annual
recruitment rates among populations. Growth was measured as the ratio of
plants in each population that grew into a larger size class to those that
remained in the same class or became smaller. We examined differences in age
at maturity by comparing ratios of plants that flowered during the first two
years to those that flowered later. Differences in recruitment, growth, new
recruit survival, survival of bolting plants, age at maturity and proportion
of bolting plants were assessed with an overall chi-square goodness of fit
test. If a 3 X 3 test showed a significant result, I used 2X2 tests to
determine which pairs of sites were different. Probability values were not
corrected for multiple tests.

We compared survivorship of the uneven-age sample population present in
1989 and the 1990 cohort among the sites using the nonparametric logrank test
(Pyke and Thompson 1986, Hutchings et al. 1991). Survivorship curves were
constructed following methods outlined in Hutchings et al. (1991).
Probability values were not adjusted for multiple tests.

The effects of site (population), year and bolting on number of fruits
per plant and number of seeds per fruit were analyzed using analysis of
variance (ANOVA) followed by contrast tests. Dependent variables were log-
transformed prior to analysis. The effects of treatment, site and bolting on
the arcsine-transformed proportion of germinating seed were also analyzed by
ANOVA.

Results

Vegetation

Mean canopy cover estimates for common vascular plant species are
presented in Table 1, Total basal vegetation cover was lower at Charley's
Gulch compared to Lime Gulch or Vipond Park (Table 1). Graminoids were common
at Lime Gulch, but forbs were more common at the other two sites. Amounts of
bare soil were highest at Charleys Gulch, intermediate at Lime Gulch, and
lowest at Vipond Park, while rock was more abundant at Vipond Park (Table 1).

Population Growth

Density of Arabis fecunda varied among sites and years (Fig. 1).
Population size was more variable at Lime Gulch and Vipond Park than at
Charleys Gulch. The coefficient of variation for density for 1989-93 was 22%
at Lime Gulch and Vipond Park but was 18% at Charleys Gulch.

Equilibrium population growth rate (X) also varied among sites and years
(Table 2). X was lowest and least variable at Charleys Gulch. In 1989 there
were no reproductive plants at Lime Gulch, but there were many in 1990. Thus,
1989-90 was a year of exceptional growth at Lime Gulch, but X was nearly
constant in the three ensuing year. X showed the most consistent high
variation at Vipond Park.

Recruitment

The ratio of new Arabis fecunda recruits to number of survivors was
significantly greater at Vipond Park compared to Charleys Gulch for all four
transition years (Fig. 2). In most years Vipond Park had higher recruitment
than Lime Gulch, and Lime Gulch had higher recruitment than Charleys Gulch
(Fig. 2). When all four years are pooled, the ratio of new recruits to number
of survivors is 0.31 for Charleys Gulch, 0.55 for Lime Gulch, and 0.95 for
Vipond Park, and these differences are significantly different between all
possible pairs of sites (P<0.001).

" Survivorship

Survivorship of the 1990 Arabis fecunda cohort over 1990-93 was
significantly lower at Vipond Park than at either Lime Gulch (LR=9.22, P<0.01)
or Charleys Gulch (LR=3.96, P=0.05; Fig. 3). Survivorship at Lime Gulch and
Charley's Gulch was not different (LR=0.01, P=0.91; Fig. 3). Analysis of the
depletion curve for the 1989 uneven-age populations gave similar results.
Arabis fecunda populations at Charleys Gulch and Lime Gulch have type II
survivorship curves where number of deaths is a constant with time, while the
Vipond Park population's survivorship fits more closely a type III curve,
where probability of death is a constant (Deevey 1947).

From 1991 through 1993, the proportion of new recruits that survived was
67% at Charleys Gulch, 74% at Lime Gulch and 57% at Vipond Park. The ratio of
survivors to deaths of new recruits at Vipond Park was significantly lower
than either Lime Gulch (x"=35.3, P<0.01) or Charleys Gulch {x"=4.14, P=0.04),
A while Lime Gulch and Charleys Gulch were not different (x"=1.90, P=0.17).

Growth

In two out of four years, significantly more Arabis fecunda plants moved
into larger size classes at Vipond Park compared to Charleys Gulch, and in
three out of four years growth was significantly greater at Lime Gulch
compared to Charleys Gulch (Fig. 4). When summed over all four years, there
were significantly fewer plants moving into larger size classes at Charleys
Gulch (x^=24.761, df=2, P<0.001), but there was no difference between Lime
Gulch and Vipond Park (x-=l-281, P=0.26).

Fecundity

Over the course of the study the ratio of the number of Arabis fecunda
plants that bloomed at an early age (<2 yr) to later (>2 yr) was 1.1 at
Charleys Gulch, 1.75 at Lime Gulch and 4.0 at Vipond Park. Vipond park was
^ significantly greater than both Lime Gulch (x^=17.05, P=0.001) and Charleys

r^

C

♦

Gulch (x^=7.72, P=0.005), but the latter two sites were not different
(X'=0.89, P=0.345). Thus, A_^ fecunda plants at Vipond Park matured earlier in
life than those at the other two sites.

The number of fruits per reproductive plant varied significantly among
sites and years (Table 3). Over the course of the study, the mean for Lime
Gulch was 10.6, significantly lower than Charleys Gulch and Vipond Park which
had means of 14.6 and 14.5 respectively (Table 3). The number of seeds per
fruit also varied significantly among sites and years although differences
were not large (Table 4). Over the course of the study, the means were 30.9,
32.4 and 34.0 for Charleys Gulch, Lime Gulch and Vipond Park respectively.
Only Charleys Gulch and Vipond Park were significantly different (Table 4).
For both number of seeds per fruit and fruits per plant, there was a
significant interaction between the site and year effects, possibly due to
different weather conditions at the sites.

Bolting plants

Over the period of 1990-93 the mean percentage of reproductive plants
that produced a bolting inflorescence was 3%, 26% and 44% for Charleys Gulch,
Lime Gulch and Vipond Park respectively (Fig. 6). These differences were
statistically significant for all four years (x">7.3, df=2, P<0.05).

Bolting plants had a mean of 19.9 fruits, while axillary-flowering
plants had a mean of 9.8 fruits, and this difference was significant (Table
5). There were also significant effects of year, and site X year interaction,
probably due to different weather conditions at the sites over the course of
the study. There was also a strong interaction between bolting and year,
suggesting that the number of fruits per plant is partially under
environmental control. Bolting plants produced an average of 36.8 seeds per
fruit, while axillary-flowering plants produced 29.8 seeds per fruit, and the
difference was significant (ANOVA, F=26.4, P<0.001). Number of seeds per

10

\t

*

fruit also varied among sites in the same manner as described above. On
average, bolting plants produce 2.5 times as many seeds as axillary flowering
plants .

During the period of 1991-93 at Lime Gulch, 53% of axillary flowering
plants survived to the following year, while only 5% of bolting plants
survived. Results for the Vipond Park population were similar, with 68% of
axillary flowering and 16% of bolting plants surviving. Survivorship was
significantly greater for axillary flowering plants at both sites for all
three years (x->17.3, df=2, P<0.001).

Germination recruirements

Germination of Arabis f ecunda seed occurred readily at room temperature
in the light without stratification. Site (source of seed) had no effect on
germination; 89% and 86% of seed from Charleys Gulch and Vipond Park
respectively germinated within eight days (ANOVA F=0.048, df=24, £=0.828).
80% of seeds from Charleys Gulch germinated in the cold and dark after 14
days, but only 8% of seeds from Vipond Park germinated under the same
conditions, and this difference was highly significant {F=129.59, df=24,
P<0.001). Seeds from Vipond Park remained dormant after being placed in a
warm light environment for eight days. The significant site*treatment
interaction in the full ANOVA model (Table 6) indicates that seeds from the
two sites are genetically different. Seeds from bolting plants germinated
better than those from axillary flowering parents under warm-light conditions
but germinated more poorly under cold, dark conditions. The significant
bolting X treatment interaction term in the ANOVA model suggests that seeds
from bolting and axillary flowering plants are genetically different (Table
6).

11

.4
V

Elasticity analysis

Elasticities for the three sites for the four annual transitions are
presented in Table 7. Growth and survival of plants in the non-reproductive
stages accounted for ca. 50% of equilibrium population growth (X) at all three
sites. Growth and survival of reproductive plants was responsible for 36% of
X at Charleys Gulch but less than 20% at Lime Gulch and Vipond Park. On the
other hand, recruitment from seed accounted for 34% and 36% of X at these
latter two sites but only 16% at Charleys Gulch (Table 7). Adult
(reproductive) growth and survival was the most important transition at
Charleys Gulch, while recruitment was predominant at Lime Gulch and Vipond
Park.

Discussion
Life history

Arabis fecunda is a relatively short-lived perennial; only ca. half of
the plants that establish live for more than two years, and only ca. one-third
live for four years or more. Annual recruitment is generally high; the ratio
of new recruits to survivors varied from 0.09 to 2.05 with means for 1989-93
between 0.31 and 0.95. Mortality of new recruits is also high; in 1991-93, it
varied from ca. 20-50%. Fecundity is generally high; reproductive A^ fecunda
plants produced an average of 340-500 seeds per year. Plants that bolted
produced ca. 2.5 times as many seeds per year as axillary flowering plants but
had much higher mortality. Seeds become ripe in late spring or early summer
and germinate readily without stratification. These results suggest that most
seeds germinate in the fall, of the same year that they are produced. Seeds
from Charleys Gulch also show high germination in the cold and dark,
suggesting that at this site only a transient type II seed bank is formed
(sensu Thompson and Grime 1979). On the other hand, cold/dark conditions
induce dormancy in seeds from Vipond Park; thus, A_^ fecunda probably does have
a long-term seed bank at this site.

12

Variability in life histories

Life history theory predicts tradeoffs between traits that will maximize
fitness for a particular environment. In particular, there is thought to be a
negative relationship between reproduction and growth and survival. Some
environments favor slower growth, greater age at first reproduction, smaller
output per reproductive bout, and greater longevity. Other environments
select for shorter lifespan, early maturity and larger reproductive output per
bout. Early maturing, highly fecund populations have a higher intrinsic
population growth rate. Early maturity is favored in environments where adult
mortality is relatively high or highly variable (Stearns 1992). In plants,
the extreme case is the annual habit.

There was great variation in life history traits among the three
populations studied. In most cases, the Charleys Gulch and Vipond Park
populations occupied the two extremes of life history trait continua with the
Lime Gulch population intermediate. For the purpose of this discussion, we
will compare the former two populations, bearing in mind that the Lime Gulch
population was similar to Charleys Gulch for some traits but more similar to
Vipond Park for most.

The Arabis fecunda population at Charleys Gulch had a lower recruitment
rate but higher overall as well as new recruit survivorship. On average,
plants grew more slowly, were older at first reproduction, and had lower
annual fecundity as a result of producing fewer seeds per fruit. The Vipond
Park population had higher recruitment, faster growth, and higher mortality.
Annual fecundity was higher and plants became fecund at an earlier age.
Population size was more stable at Charleys Gulch than at Vipond Park. The
Vipond Park population demonstrated germination traits that make a long-term
seed bank more likely than at Charleys Gulch.

13

A

The frecjuency of bolting was much higher at Vipond Park, and this is
likely the source of much of the difference between Arabis fecunda life
histories at the two sites. Bolting plants have higher annual fecundity and
much higher mortality than axillary flowering plants. Axillary flowering
plants are iteroparous (perennial or polycarpic), while bolting plants
approach the semelparous (annual or monocarpic) life history.

Discussion of the environmental characters that cause these differences
in Arabis fecunda life history can only be speculative. Charleys Gulch is
warmer and likely has lower precipitation. Bare soil was more common and
vegetation cover was lower. The bleached color of the mineral soil at
Charleys Gulch may indicate a more extreme edaphic environment. These
conditions may result in slower growth but lower density-dependent mortality
which, in turn, should provide more stable population sizes and favor the
iteroparous habit. The high elevation of the Vipond Park site may provide an
more unstable habitat in which the semelparous habit and a long-term seed bank
are favored (van Groenendael and Slim 1988).

The differences in life history traits exhibited among the populations
studied could be the result of genetic differentiation, phenotypic plasticity
(one genotype that produces different phenotypes under different conditions)
or both. Quantitative genetics studies are required to determine the basis of
the variation. Leeper et al. (in press) used starch gel electrophoresis to
investigate apportionment of genetic variation in Arabis fecunda populations,
including the three that we studied. Of 18 putative loci scored, 17 were
invariant; however, the one polymorphic locus had different frequencies among
the populations, suggesting a fair degree of differentiation. Results of the
germination studies indicate that there is genetic differentiation between the
Charleys Gulch and Vipond Park populations. Furthermore, they suggest that
there is a genetic difference between plants that bolt and those that do not.

14

i

i

i

A Together these results provide evidence that differences in life history
traits between the two sites have a genetic basis.

Population growth and viability

Sample populations of Arabis fecunda at Lime Gulch and Vipond Park, the
two study sites in the southern portion of the range became larger between
1989 and 1993. Equilibrium population growth rates (A.) at these sites were
generally greater than or equal to one. Thus, our study provided no evidence
that these populations are in decline. On the other hand, sample populations
at Charleys Gulch and Birch Creek (Lesica and Shelly 1993) became smaller in
number since 1987. Furthermore, X at Charleys Gulch was appreciably less than
one in two out of the four years that it was measured. Our study was designed
to elucidate demographic and life history characters and may not provide a
robust assessment of trend. Nonetheless, our results suggest that populations
in Ravalli County may be declining. Populations of A_^ fecunda in Ravalli
^ County have been invaded by the aggressive exotic, Centaurea maculosa, and

Lesica and Shelly (manuscript submitted) provide evidence that the invader has
a negative impact on population growth rates of A_^ fecunda. Furthermore, all
Ravalli County sites are subject to livestock grazing (Lesica 1985,
Schassberger 1988) which may have adverse effects on A_^ fecunda (Lesica and
Shelly 1992). Taken together, these observations suggest that A_^ fecunda
populations in the northern portion of its range may be in jeopardy.

Analysis of elasticity matrices indicates that Lime Gulch and Vipond
Park populations of Arabis fecunda are heavily dependent on recruitment from
seed to maintain population growth, while the Charleys Gulch population
depends most on survivorship of mature individuals. This is consistent with
the presence of germination responses promoting a long-term seed bank at
Vipond Park but not at Charleys Gulch. Thus, populations at Lime Gulch and
Vipond Park will be most sensitive to changes that reduce seedling
establishment such as damping-off diseases or the introduction of aggressive

15

i

<

i

exotics. The Charleys Gulch population should be most affected by
disturbances that destroy adults, such as trampling or herbicide application.
The differences in life history traits are at least partly controlled by the
frequency of bolting and axillary flowering, and the fact that both types of
flowering occurred in all populations suggests that there is probably ample
variation, genetic or plastic, to compensate for any changes that may occur if
they are not too drastic and do not occur too quickly.

Management Considerations
Results of our studies suggest that Arabis f ecunda populations at Lime
Gulch and Vipond Park are stable or growing. Population growth at these sites
depends heavily on recruitment from seed, a life history stage that is
probably buffered by the presence of a long-term seed bank. Weed infestations
could pose a serious problem as they can reduce recruitment of A_^ fecunda
(Lesica and Shelly submitted) . Furthermore, weed infestations are most
frequent in the mesic grassland and xeric forest zones in western Montana
(Forcella and Harvey 1983), the same habitats where A_^ fecunda is most common.
At this time, there are no serious weed infestations near any known
populations in Beaverhead or Silver Bow counties. Nonetheless, encroachment
by exotics is a very real potential problem. Populations of A^ fecunda should
be regularly monitored for exotics, and roads and other disturbances that
promote weed infestations should be minimized in these areas.

Populations of Arabis fecunda at Charleys Gulch and Birch Creek (Lesica
and Shelly 1993) may be declining. Results of our studies indicate that the
Charleys Gulch population will be most sensitive to declines in the survival
of mature plants. Centaurea maculosa is present at all Ravalli County sites,
and this aggressive exotic does have a negative effect on A_^ fecunda
population growth (Lesica and Shelly, submitted) . However, the main negative
effect of C^ maculosa on A^ fecunda is to reduce recruitment, so the two
species may be able to coexist (Lesica and Shelly, submitted) . On the other

16

hand, livestock are also present at the Ravalli county A_^ fecunda sites, and
trampling by livestock or large ungulates can have an adverse effect on adult
survival (Lesica and Shelly 1992). Negative impacts resulting from heavy
livestock trampling and Centaurea maculosa encroachment taken together may be
enough to result in declines of A_^ fecunda populations.

Anthropogenic global climate change is considered a potential cause of
species extinctions in the near future (Dobson et al. 1989, Peters 1988).
Populations of Arabis fecunda occur throughout a wide range of elevations and
habitats (Schassberger 1988), so it seems unlikely that climatic changes will
have adverse effects.

Acknowledgements
Peter Achuff, Anne Garde, Bonnie Heidel, Lisa Roe and Jim Vanderhorst
helped conduct field work. We are grateful to George Frost for allowing us to
conduct our study on his ranch. This study was funded by Beaverhead National
Forest, the U.S. Fish and Wildlife Service and The Nature Conservancy.

Literature Cited

Caswell, H. 1989. Matrix population models. Sinauer Associates, Sunderland,
Massachusetts, USA.

Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis.
Northwest Science 33: 43-64.

Deevey, E. S. 1947. Life tables for natural populations of animals.
Quarterly Review of Biology 22: 283-314.

Dobson, A., A. Jolly and D. Rubenstein. 1989. The greenhouse effect and
biological diversity. Trends in Ecology and Evolution 4: 64-68.

de Kroon, H. , A. Plaiser, J. M. van Groenendael and H. Caswell. 1986.
Elasticity: the relative contribution of demographic parameters to population
growth rate. Ecology 67: 1427-1431.

Forcella, F. and S. J. Harvey. 1983. Eurasian weed infestation in western
Montana in relation to vegetation and disturbance. Madrono 30: 102-109.

Hutchings, M. J., K. D. Booth, and S. Waite. 1991. Comparison of
survivorship by the logrank test: criticisms and alternatives. Ecology 72:
2290-2293.

17

Leeper, D., D. Pavek, R. Walsh and T. Mitchell-Olds. 1993. Preliminary
report of combined demographic and genetic analyses for management of Arabis
fecunda. In Plants and their environment in the Greater Yellowstone
Ecosystem. National Park Service Transactions.

Lefkovitch, L. P. 1965. The study of population growth in organisms grouped
by stage. Biometrics 21: 1-18.

Lesica, P. 1985. Report on the conservation status of Arabis fecunda, a
potential candidate species. Report to the U.S. Fish and Wildlife Service,
Office of Endangered Species, Denver, CO.

Lesica, P. 1987. A technique for monitoring nonrhizomatous perennial plant
species in permanent belt transects. Natural Areas Journal 7: 65-68.

Lesica, P. 1992. Vascular and sensitive plant species inventory for the
Highland Mountains, Deer lodge National Forest. Montana Natural Heritage
Program, Helena.

Lesica, P. and J. S. Shelly. 1991. Endangered, threatened and sensitive
vascular plants of Montana. Montana Natural Heritage Program, Occasional
Publication No. 1, Helena, Montana, USA.

Lesica, P. and J. S. Shelly. 1992. The effects of cryptogamic soil crust on
the population dynamics of Arabis fecunda (Brassicaceae) . American Midland
Naturalist 128: 53-60.

Lesica, P. and J. S. Shelly. 1993. Demographic monitoring of Arabis fecunda
populations in the Sapphire and Beaverhead ranges, Montana. 1992 progress
report. Unpublished report, Montana Natural Heritage Program, Helena.

Lesica, P. and J. S. Shelly. Submitted. Demographic analysis of competitive
effects of Centaurea maculosa on Arabis fecunda. Journal of Ecology.

Menges, E. S. 1986. Predicting the future of rare plant populations:
demographic monitoring and modeling. Natural Areas Journal 6: 13-25.

Menges, E. S. 1990. Population viability analysis for an endangered plant.
Conservation Biology 4: 52-62.

Palmer, M. E. 1987. A critical look at rare plant monitoring in the United
States. Biological Conservation 39: 113-127.

Peters, R. L. 1988. Effects of global warming on species and habitats: An
overview. Endangered Species Update 5(7): 1-8.

Pyke, D. A. and J. N. Thompson. 1986. Statistical analysis of survival and
removal experiments. Ecology 67: 240-245.

Rollins, R. C. 1984. Studies in the Cruciferae of western North America II.
Contributions to the Gray Herbarium 214: 1-18.

Schassberger, L. A. 1988. An update of the report on the conservation status
of Arabis fecunda, a candidate threatened species. Report to the U.S. Fish
and Wildlife Service, Office of Endangered Species, Denver, CO.

Silvertown, J., M. Franco, I. Pisanty and A. Mendoza. 1993. Comparative
plant demography - relative importance of life-cycle components to the finite
rate of increase in woody and herbaceous perennials. Journal of Ecology 81:
465-476.

18

Stearns, S. C, 1992. The evolution of life histories. Oxford University
Press, Oxford.

Sutter, R. D. 1986. Monitoring rare plant species and natural areas-ensuring
the protection of our investment. Natural Areas Journal 6: 3-5.

Thompson, K. and J. P. Grime. 1979. Seasonal variation in the seed banks of
herbaceous species in ten contrasting habitats. Journal of Ecology 67: 893-
921.

USDI-Fish and Wildlife Service. 1993. Endangered and threatened wildlife and
plants; review of plant taxa for listing as endangered or threatened species;
notice of review. Federal Register 58: 51144-51190.

Vandermeer, J. 1978. Choosing category size in a stage projection matrix.
Oecologia 32: 79-84.

van Groenendael, J, M. de Kroon and H. Caswell. 1988. Projection matrices in
population biology. Trends in Ecology and Evolution 3: 264-269.

van Groenendael, J. M. and P. Slim. 1988. The contrasting dynamics of two
populations of Plantaqo lanceolata classified by age and size. Journal of
Ecology 76: 585-599.

Werner, P. A. and H. Caswell. 1977. Population growth rates and age versus
stage-distribution models for teasel (Dipsacus sylvestris Huds.). Ecology 58:
1103-1111.

Wilson, E. O. 1988. Biodiversity. National Academy Press, Washington D.C.

Young, T. P. 1985. Lobelia telekii herbivory, mortality and size at
reproduction: variation with growth rate. Ecology 66: 1879-1883.

19

<

#

Table 1. Mean ground cover and canopy cover of common vascular plant species
in Arabis fecunda monitoring transects at three study sites.

Charleys Gulch
East West

Rock
Soil
Basal vegetation

Agropyron spicatum
Aristida longiseta
Carex filifolia
Carex rossii
Oryzopsis hymenoides
Poa secunda
Stipa comata

5
77
19

2
59
40

<1
3 1

Lime

Gulch

North

South

6

6

52

58

43

38

6

6

—

3

—

5

8

—

9

2

7

15

Vipond
East

Park
West

14
43
43

15

40
48

8

12

—

2

Artemisia frigida
Centaurea maculosa
Chrysopsis villosa
Haplopappus acaulis
Oxytropis besseyi
Phlox muscoides
Physaria geyeri
Sedum lanceolatum
Senecio canus

30
5 <1

10

19

<1

8

13

2

11

<1

20

Table 2. Stage-based transition matrices for Arabis fecunda at three sites in 1989-93. The reproductive and
recruitment columns must be added together before solving for 1, the dominant eigenvalue (see Methods).

Charleys Gulch

1989-90

From

1991-92

From

Sm

Ro

Mu

Rep

Rec

Sm

Ros

Mul

Rep

Rec

To

To

Small

0

0

0

0

.377

Small

.455

.016

0

.036

.321

Rosette

.391

.375

.136

.076

.434

Rosette

.205

.492

0

.286

.071

Multiple

.044

.025

.182

.057

.076

Multiple

0

.064

.625

.179

0

Repro

.087

.325

.409

.660

.094

Repro

0

.286

.250

.286

0

J.= 1

.138

X=0

.898

1990-91

From

1992-93

From

Sm

Ro

Hu

Rep

Rec

Sm

Ros

Mul

Rep

Rec

To

To

Small

.350

.056

.077

0

.321

Small

.091

0

0

0

.031

Rosette

.150

.556

0

.286

.071

Rosette

.333

.306

0

.094

.156

Multiple

0

.074

.539

.159

0

Multiple

0

.020

.250

.031

.094

Repro

0

.148

.154

.286

0

Repro

0

.204

.250

.653

.281

X=0

.844

i=1

.050

Lime Gulch

1989-90

From

1991-92

From

Sm

Ro

Mu

Rep

Rec

Sm

Ros

Mul

Rep

Rec

To

To

Small

.193

.022

.023

0

8.42

Small

.236

.009

0

0

.675

Rosette

.518

.248

.046

0

8.57

Rosette

.382

.307

.021

.065

.398

Multiple

.024

.071

.341

0

1.29

Multiple

.016

.031

.333

.008

.073

Repro

.036

.495

.364

.714

3.00

Repro

.033

.425

.396

.301

.114

X=i,.

.909

X=1

.009

1990-91

From

1992-93

From

Sm

Ro

Mu

Rep

Rec

Sm

Ros

Mul

Rep

Rec

To

To

Small

.161

.001

.023

.044

.635

Small

.244

.017

.032

.012

1.11

Rosette

.543

.506

.046

.101

.522

Rosette

.435

.449

0

.089

.746

Multiple

.049

.056

.250

.050

.082

Multiple

.009

.017

.452

.012

.041

Repro

.025

.269

.523

.327

.044

Repro

.009

.298

.323

.420

.036

i=1.

.068

X=^.

.130

Vipond

Park

1989-90

From

1991-92

From

Sm

Ro

Hu

Rep

Rec

Sm

Ros

Mul

Rep

Rec

To

To

Small

.154

.036

.019

.049

.854

Small

.271

.031

.035

.010

1.75

Rosette

.289

.255

0

.037

.976

Rosette

.157

.245

.035

.087

.505

Multiple

.039

.042

.245

.037

.439

Multiple

.186

.063

.368

.049

.272

Repro

.173

.442

.491

.598

.622

Repro

.043

.453

.333

.447

.252

i=1.

815

i=1.

,357

1990-91

From

1992-93

From

Sm

Ro

Mu

Rep

Rec

Sm

Ros

Mul

Rep

Rec

To

To

Small

.157

.037

.016

0

.246

Small

.116

.025

.035

.006

.546

Rosette

.326

.244

.079

.100

.322

Rosette

.295

.280

.012

.081

.839

Multiple

.079

.089

.413

.043

.147

Multiple

.073

.059

.391

.040

.299

Repro

.056

.296

.175

.194

.043

Repro

.024

.271

.138

.333

.052

X=0.783

X= 1.044

21

Table 3. Effect of site (population), year and their interaction on log-
transformed number of fruits per reproductive Arabis fecunda plant in 1989-93
by ANOVA. Means (+SE) followed by different letters are significantly
different (P<0.001) by contrast test after ANOVA.

Source of Variation

df

MS

Site
Year

Site*Year
Error

2

4

8

1495

13,
5.
4.

39
21

61

0.81

16.45
6.40
5.66

<0.001
<0.001
<0.001

Charleys Gulch
14.6+0.8'

Lime Gulch
10.6+0.4''

Vipond Park
14.5+0.5'

Table 4. Effect of site (population), year and their interaction on log-
transformed number of seeds per fruit for Arabis fecunda in 1989-91 and 1993
by ANOVA. Means (+SE) followed by different letters are significantly
different (P<0.05) by contrast test after ANOVA.

Source of Variation

df

MS

Site
Year

Site*Year
Error

2

3
6

288

234.1

150.8

417.9

56.4

4.15 0.017
2.67 0.048
7.41 <0.001

Lime Gulch

Vipond Park

32.4+0.7'*'

34.0+1.0''

Charleys Gulch
30.9+0.6'

22

Table 5. Effect of bolting, site (population), year and their interactions on
log-transformed number of fruits per Arabia fecunda plant in 1990-93 by ANOVA.

Source of Variation

df

MS

Bolting

Site

Year

Bolting*Site

Bolting*Year

Site*year

Error

1

13.67

20.

.76

<0.001

2

1.05

1.

.60

0.202

3

6.08

9.

.24

<0.001

2

1.84

2.

,79

0.062

3

7.49

11.

.38

<0.001

6

3.03

4.

,60

<0.001

1347

0.66

ch

Lime Gulch

Vipond Park

18.3+0.

.9

20,

.9+0.8

8.1+0.

.3

9.

.8+0.5

Bolting

Axillary flowering

Charleys Gulch

17.3+3.9
14.3+0.9

Table 6. Effect of treatment, site (population), bolting and their
interactions on arcsine-transformed proportion of Arabis fecunda seeds
germinating by ANOVA.

Source of Variation

df

MS

Treatment

1

3.

.83

103.15

<0.001

Site

1

2,

.43

65.43

<0.001

Bolting

1

0,

.02

0.50

0.484

Treatment* Site

1

2,

.23

60.15

<0.001

Treatment* Bolting

1

0,

.18

4.90

0.033

Site*Bolting

1

0,

,01

0.28

0.602

Error

41

0,

,04

Warm-Light

Cold-Dark

Charleys Gulch

Axillary flowering

0.87+0.04

0.86+0.05

Bolting

0.91+0.04

0.72+0.07

Vipond Park

Axillary flowering

0.86+0.04

0.11+0.05

Bolting

0.87+0.07

0.05+0.02

23

Table 7 Mean elasticities for Arabis fecunda stage transition matrices at
JSieJltf fo..lS8,-,3^ T.eU«tJ„e colons «pr.se„t^.=„-^

seed.

Charley

■s Gulch

Small

Rosette

From

Multiple

Repro

Recruit

To

Small
Rosette
Multiple
Repro

.0219
.0340
.0010
.0041

.0030
.1481
.0163
.1146

.0017
.0015
.0806
.0541

.0011
.0518
.0287
.2752

.0333
.0467
.0113
.0714

Total

.0610

.2820

.1379

.3568

.1627

Lime

Gulch

Small

Rosette

From

Multiple

Repro

Recruit

TO

Small
Rosette
Multiple
Repro

.0207
.0740
.0040
.0186

.0013
.1156
.0111
.2005

.0004
.0012
.0131
.0340

.0019
.0136
.0043
.1327

.0832
.1242
.0162
.1396

Total

.1171

.3285

.0487

.1525

.3632

Vipond Park

Small

Rosette

From

Multiple

Repro

Recruit

To
Small
Rosette
Multiple
Retjro

.0173
.0380
.0160
.0389

.0055
.0572
.0135
.1580

,0026
.0074
.0483
.0695

.0012
.0167
.0073
.1651

.0836
.1149
.0428
.0965

Total .1102 .2342 .1278 .1903 .3378

24

Figure 1. Number of Arabis fecunda plants at three study sites in 1989-93.

700
600
500
400
300
200
100
0

• Charleys
V Lime
T Vipond

v-

V

J L

1989 1990 1991 1992 1993
Year

25

Figure 2. Annual recruitment in relationship to population size (survivors)
of Arabia fecunda at three study sites in 1989-93. Sites with different
letters had different recruitment rates (recruits/survivors) as determined by
Chi-square tests (P<0.05).

1990

1991

c

O

0)

E

3

700
600
500
400
300
200
100
0

Survivors
Recruits

Charleys Lime Vipond

Charleys Lime Vipond

1992

1993

Charleys Lime Vipond

Charleys Lime Vipond

26

Figure 3. Survive
study sites.

rship curves for the 1990 Arabis fecunda cohort at three

O
>

*>

13
in

c

CD

O
i_

O)

Q_

100

80 -

50

40

20 -

0

• Charleys
V Lime
T Vipond

1990 1991 1992 1993
Year

27

Figure 4. Arabis f ecunda plants moving into a larger size class or moving
into the same or a smaller class at three study sites in 1989-93. Sites with
different letters had different growth rates ( larger/smaller+same) as
determined by Chi-square tests (P<0.05).

1990

1991

500

400 -

Q- 300 -

E

500

Charleys Lime Vipond

Charleys Lime Vipond

500

« 400

o

°- 300
"o

-

1992
b

b

-

Numbe

-■ IS)

o o

3 O O

0

1

1

-

1993

500

Charleys Lime Vipond

Charleys Lime Vipond

28

Figure 5. Number of bolting and axillary-flowering Arabis fecunda plants at
three study sites in 1990-93. Sites with different letters had different
proportions of bolting/axillary plants as determined by Chi-square tests
(P<0.05) .

1990

1991

c

Q-

E

3

250

200

150

100

50

I I Axillary
F^^ Bolting

250

200 h

150

100

50 h

b
~ a

Charleys Lime Vipond

Charleys Lime Vipond

1992

1993

250

250
200
150
100
50

-

b

c

-

-

P^

P

-

o

Charleys Lime Vipond

Charleys Lime Vipond

29

Pl^j^,.., .,,.,, ..,,;,

f>;;tr^,,,fjif:j- ...

L