Seed dispersal of southwestern white pine (pinus strobiformis)

Material Information

Seed dispersal of southwestern white pine (pinus strobiformis)
Samano, Sheridan
Publication Date:
Physical Description:
41 leaves : ; 28 cm


Subjects / Keywords:
White pine -- Seeds -- San Juan Mountains (Colo. and N.M.) ( lcsh )
Seeds -- Dispersal ( lcsh )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 39-41).
Statement of Responsibility:
by Sheridan Samano.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
48460682 ( OCLC )
LD1190.L45 2001m .S35 ( lcc )

Full Text
Sheridan Samano
B.S., Texas A&M University, 1995
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

This thesis for the Master of Arts
degree by
Sheridan Samano
has been approved by
Leo P. Bruederle
C j


Samano, Sheridan (M.A., Biology)
Seed Dispersal of Southwestern White Pine (Pinus strobiformis)
Thesis directed by Professor Diana F. Tomback
I studied seed dispersal of southwestern white pine (Pinus strobiformis) in
the San Juan Mountains of southern Colorado. Dispersal by wind and dispersal
by animals are the two alternative modes for seed dispersal of pines. The bird-
dispersed pines, in particular, have large, wingless seeds that cannot be carried
effectively by wind. Southwestern white pine, a close relative of limber pine, has
large, wingless seeds (occasional remnant wings) characteristic of bird dispersal,
but its modes of seed dispersal are poorly known.
Clarks Nutcracker (Nucifraga columbiana) is known to be an effective
seed disperser of several western North American pines. In the San Juan
Mountains, Clarks Nutcracker was found to be the most important primary
dispersal agent for southwestern white pine seeds. Stellers Jay (Cyanocitta
stelleri) played a role in primary and secondary seed dispersal of southwestern
white pine. Stellers Jay is the only corvid that occurs throughout the entire range
of southwestern white pine. Stellers Jays are less specialized than nutcrackers
for seed dispersal, but the role of Stellers Jay may become more important in
areas without nutcrackers. Small mammals (e.g., Tamias quadrivittatus) also play
a role in secondary seed dispersal of southwestern white pine. Pine squirrels
(Tamiasciurus hudsonicus) were found to be the most important pre-dispersal
seed predator of southwestern white pine. Seed and cone morphology of

southwestern white pine reflects selective pressures by Clarks Nutcracker
(primary seed disperser) and pine squirrels (pre-dispersal seed predator).
I also reevaluated the cone opening phenology of southwestern white pine
through visual observations by evaluating the degree of cone scale separation. I
found an asynchronous ripening pattern among and within individual trees, a
pattern predicted to promote avian seed dispersal. I propose that future studies
should use visual cues to determine cone opening phenologies for pines because
dispersers probably use these cues to ascertain seed ripeness.
This abstract accurately represents the content
recommend its publication.
of the candidates thesis. I

I dedicate this thesis to my grandfather, Henry Lyles Zabriskie, for sharing his
love and respect for the natural wonders that surround us each day.

My thanks to my advisor, Diana F. Tomback, for her guidance and support
throughout the development of this project. I offer my gratitude to Elizabeth
Weber for volunteering to gather valuable field observations, and to the staff of
Indian Head Lodge in Pagosa Springs, Colorado for providing refuge.
I also wish to thank Leo P. Bruederle and Greg Cronin for their words of
encouragement and constructive comments.

1. Introduction........................................................1
2. Study Area..........................................................6
3. Field Methods.......................................................8
3.1 Fo li age and Cone Measurements.....................................8
3.2 Cone Opening Phenology..............................................9
3.3 Survey of Cone Orientation.........................................10
3.4 Seedling/Sapling Transect.........................................12
3.5 Focal Observations of Animals Foraging on Southwestern White Pine
4. Results............................................................13
4.1 Foliage and Cone Measurements......................................13
4.2 Cone Opening Phenology............................................15
4.3 Survey of Cone Orientation........................................16
4.4 Seedling/ Sapling Transect........................................20
4.5 Foraging by Clarks Nutcracker....................................20


4.6 Foraging by Stellers Jay............................................26
4.7 Foraging by Other Birds..............................................27
4.8 Foraging by Pine Squirrels...........................................28
4.9 Foraging by Chipmunks................................................28
5. Discussion............................................................30

2.1 Location of Study Areas A, B, and C along Williams Creek Trail in San Juan
National Forest and adjacent Weminuche Wilderness Area in Hinsdale
County, Colorado......................................................7
3.1 Key for identifying cone positions on trees..........................11
4.1 Cone Opening Phenology for Study Area A..............................17
4.2 Cone Opening Phenology for Study Area C.............................18
4.3 Number of cones removed by squirrels................................19

4.1 Comparison of cone and foliage measurements for typical southwestern white
and limber pine with southwestern white pine from study areas in the San
Juan Mountains of southern Colorado..............................14
4.2 Cone position in a southwestern white pine population as determined on 11
September 1999 with evidence of foraging activity by Clarks Nutcracker in
the San Juan Mountains of southern Colorado......................21
4.3 Average seed extraction rates by Clarks Nutcracker for each stage of cone
openness and cone position for a population of southwestern white pine.23
4.4 Southwestern white pine cone orientation preferences for Clarks Nutcrackers
throughout the entire study period.....................................24

1. Introduction
In pines (Pinus, Pinaceae), dispersal by wind and dispersal by animals are
the two major alternative modes for primary seed dissemination the initial
movement of seeds away from the parent tree (Pijl 1972). For each mode, there
are associated seed and cone traits that facilitate movement of seeds away from
parent trees (Vander Wall and Baida 1977, Tomback and Linhart 1990, Lanner
The bird-dispersed pines (subgenus Strobus, the soft or white pines) in
particular, have large seeds that cannot be carried by wind effectively. The avian
dispersers for these pines are specialized members of the family Corvidae (jays
and nutcrackers) that have an annual cycle dependent on fresh and stored pine
seeds (Vander Wall and Baida 1977, Ligon 1978, Tomback 1978, 1982). Bird
dispersal is associated with the following pine characteristics: seeds that are larger
than those of wind-dispersed pines, seeds with absent or rudimentary seed wings,
seed retention in cones either by cone indehiscence or restraining flanges, cones
that lack terminal spines, and horizontally-directed cones placed on tips of
vertically-oriented branches. Winglessness and seed retention in cones reduce the
probability of seed dispersal by wind. These features are also likely to be
attractive to birds because they increase harvesting efficiency (Lanner 1980,
Tomback and Linhart 1990). Dropped or fallen seeds from the canopy of pines

may end up secondarily dispersed by birds and/ or small mammals (Vander
Wall 1992,1993).
The western, North American pines known to be associated with primary
avian seed dispersal are whitebark (Pinus albicaulis), limber (P. flexilis),
southwestern white (P. strobiformis), and two pinon pines (P. monophylla and P.
edulis) (Vander Wall and Baida 1977, Benkman et al. 1984, Tomback 1994,
1998). Of these pines, southwestern white pine is the least studied. To date, there
is no published information on whether the seeds of southwestern white pine are
cached by birds, and if tree recruitment occurs by this means.
Southwestern white pine is closely related to limber pine but the two
species differ in cone, seed, and needle characteristics (Steinhoff and Andresen
1971). Limber pine is known to be dispersed by birds (Lanner 1980, Lanner and
Vander Wall 1980, Tomback and Kramer 1980, Benkman et al. 1984, Vander
Wall 1988), and southwestern white pine has large, wingless seeds (occasional
remnant wings) (Steinhoff and Andresen 1971) suggesting bird dispersal.
Southwestern white pine has greater values than limber for cone length, needle
length, peduncle length, and total seed weight (Steinhoff and Andresen 1971).
Reflexed cone scale apophyses are also characteristic of southwestern white pine
and not limber pine (Steinhoff and Andresen 1971).

Benkman et al. (1984) studied the cone and seed morphology and cone
ripening phenology of southwestern white pine in relation to selective seed
harvest by red squirrels (Tamiasciurus hudsonicus) and Clarks Nutcrackers
(Nucifraga columbiana) in north-central Arizona; however, caching and
subsequent seedling establishment of southwestern white pine were not observed.
They described the cone ripening phenology of southwestern white pine based on
the relationship between seed number and cone mass. As cones ripen, cone mass
increases, but the number of seeds remains the same. They suggested that in ripe
cones, the number of seeds per gram of cone is nearly constant, and that there
should be a strong correlation between cone mass and seed number. If cones
ripen synchronously, the correlation of cone mass to seed number should be very
high, and if asynchronously the correlation should be weaker. They concluded
that southwestern white pine cones ripened synchronously among and within
individual trees, a pattern that is associated with wind-dispersed conifers
(Benkman et al. 1984). Based on seed morphology, however, this pattern appears
non-adaptive for southwestern white pine (Benkman 1995a). These findings raise
questions about (1) the methodology to determine cone ripening phenology: what
if seed mass per cone varied among and/ or within trees, (2) what cues animals
actually use to determine seed ripeness/ availability, and (3) whether cone
ripening patterns vary geographically. For example, nutcrackers begin to cache

seeds before cones open, or in the case of whitebark pine, seeds before obtaining
maximum seed weights (Hutchins and Lanner 1982). Cues for seed caching may
relate to seed coat development and seed extraction rates rather than complete
ripeness. For some seed dispersers, scale opening may be more important for
access to seeds than complete maturity of seeds, i.e. maximum seed weights.
Southwestern white pine is found from northern Arizona and southern
Colorado south through northern Mexico (Critchfield and Little 1966), and it is an
important forest component in the San Juan Mountains of southern Colorado.
Anecdotal evidence and the work of Benkman et al. (1984) suggest that Clarks
Nutcracker disperses southwestern white pine seeds where it is sympatric with the
pine. Nutcrackers are known as effective dispersers of limber pine, whitebark
pine, Colorado pinon (P. edulis), and single-leaf pinon (P. monophylla) (Tomback
1998). Each year, individual nutcrackers harvest tens of thousands of pine seeds,
bury them in small clusters called caches, and retrieve seeds from stores during
winter, spring, and parts of the following summer until fresh seeds are available
again (Vander Wall and Baida 1977, Tomback 1978, 1982, Vander Wall and
Hutchins 1983, Tomback and Linhart 1990). In favorable years, seeds not
recovered by nutcrackers may germinate, leading to tree recruitment (Tomback
1982). By caching conifer seeds, nutcrackers can maintain climax communities,

colonize previously unforested sites, or initiate succession (Lanner and Vander
Wall 1980, Tomback et al. 1990, Tomback and Linhart 1990).
The reliable range of Clarks Nutcracker effectively ends at the southern
United States border (Tomback 1998), with the exception of a population in Cerro
Potosi, Nuevo Leon, Mexico. As such, nutcrackers are not present throughout all
of southwestern white pines range. Stellers Jay (Cyanocitta stelleri) is the only
corvid that occurs throughout the entire range of southwestern white pine, and has
been shown to be an important disperser of pinon pine seeds along with Pinyon
Jays (Gymnorhinus cyanocephalus) and nutcrackers (Vander Wall and Baida
1981, Greene et al. 1998). While less specialized than Clarks Nutcracker at
harvesting pine seeds (Green et al.1998), the Stellers Jay is a possible seed
disperser of southwestern white pine.
Secondary dispersal is further movement of seeds following primary
dispersal. Only animals can accomplish secondary seed dispersal. Evidence is
accumulating that reveals small mammals to be secondary seed dispersers in
primary bird-dispersed pines (Vander Wall 1992, 1993). For example,
Tomback and Schoettle (pers. comm.) have provided clear evidence that nocturnal
rodents secondarily disperse the seeds of limber pine.
The primary purpose of this study was to reevaluate the morphology and
cone opening phenology of southwestern white pine, and to identify the most

important diurnal animal dispersal agents for this pine. Another objective of this
study was to determine the relative importance of primary seed dispersal by
Clarks Nutcracker and/ or Stellers Jay in seed dissemination of southwestern
white pine. Finally, I wanted to identify the role of corvids and1 or small
mammals in secondary seed dispersal of this pine.
2. Study Area
I studied seed dispersal and cone ripening phenology of southwestern
white pine daily from 25 August through 30 September 1999 in the San Juan
Mountains of southern Colorado. The study was conducted along the Williams
Creek Trail in the San Juan National Forest and adjacent Weminuche Wilderness
Area (N 3732.761, W 107o3H.546; elev. approx. 2650 m) (Figure 2.1).
Southwestern white pine is a major forest component for the first 1.5 km of the
Williams Creek Trail. I partitioned the study area into three sites to track animal
usage of cones in individual trees throughout the study. Study Area A consisted
of 0.04 km2 to the west of the Williams Creek Trail within San Juan National
Forest boundaries. Study Areas B and C were farther north up the Williams
Creek Trail in the Weminuche Wilderness Area, with Study Area B on the west
side of the Williams Creek Trail and Study Area C on the east side of the trail
(Figure 2.1). These latter two study areas were approximately 60 m higher in
elevation than Study Area A, and ran the length of the trail until southwestern

Williams Creek
Figure 2.1. Location of Study Areas A, B, and C along Williams Creek Trail in
San Juan National Forest and adjacent Weminuche Wilderness Area in Hinsdale
County, Colorado.

white pine trees thinned. Combined, these two areas covered approximately 0.04
The dominant species at the study areas were southwestern white pine,
white fir {Abies concolor), Douglas-fir {Pseudotsuga menziesii), and ponderosa
pine {P. ponderosa). Ponderosa pine was mainly limited to Study Area A and the
southwestern white pine trees at Study Area A were old-growth trees. Dominant
understory species at the study areas included the shrubs: Oregon grape {Mahonia
repens), Gambels oak {Quercus gambelii), wild rose {Rosa spp.), serviceberry
{Amelanchier alnifolia), and snowberry {Symphoricarpus spp.), and herbs:
strawberry {Fragaria spp.) and meadow rue (Thalictrum spp.).
3. Field Methods
3.1 Foliage and Cone Measurements
Limber pine and southwestern white pine differ in a number of foliage and
cone characteristics. To confirm the identity of the study trees as southwestern
white pine and document any hybridization or introgression, I collected 25 fallen
cones from the present years cone crop and measured the following traits: cone
length (total length from basal to distal scales), presence or absence of reflexed
cone scales, presence or absence of a peduncle, number of seeds per cone, and
pine foliage length, if needles were attached. The 25 cones were collected from

all areas throughout the study period, with most of the cones collected after being
cut down by pine squirrels (Tamiasciurus hudsonicus), as the cones were too high
to reach with a tree pruner. After removing all seeds from the cones, I dried the
seeds for 96 hours at 56 C. I then obtained total seed weight and endosperm
weight, and seed coat thickness measurements for a total of 50 seeds from 10
I also collected 200 fallen southwestern white pine cones in good
condition from previous years cone crops to determine the percentage of cones
with peduncles and/ or reflexed cone scales. One hundred cones were collected
within 3 m of the Williams Creek Trail at Study Areas B and C. An additional
100 cones were collected along a transect near the ten marked trees at Study Area
A (see 3.2 Cone Opening Phenology). I also measured foliage lengths and DBH
(diameter at breast height) for 19 trees distributed throughout the study areas.
Foliage measurements were taken for each needle fascicle, with the longest needle
length recorded. The data gathered here were compared to those data presented
by Steinhoff and Andresen (1971) from their research on limber and southwestern
white pine.
3.2 Cone Opening Phenology
I defined a ripe cone as one that had open scales and thus made its seeds
available to dispersers, as opposed to changes in seed mass (e.g., Benkman et al.

1984). I marked ten trees at Study Area A to evaluate cone opening phenology in
southwestern white pine at this site. Three additional trees were monitored at
Study Area C to observe any changes in opening phenology based on elevational
or other differences at this site. All cones visible on each tree from a designated
viewpoint were counted and mapped. I also took DBH measurements for the ten
phenology trees at Study Area A. The three trees at Study Area C grew on steep
cliffs, which prevented me from recording DBH measurements. I documented the
progression of opening on 30 August, 2 September, 7 September, 11 September,
16 September, 25 September, and 30 September 1999.
I separated the stages of openness for each cone into four distinct
categories: closed, scales separating, partly open, and open. Closed
cones exhibited completely closed cone scales, and were bright green and highly
resinous. Cones with scales separating revealed scale separation at the basal
end of the cones. Partly open cones had visible seeds on the basal and central
portions of each cone. Seeds were visible on the basal, central, and distal portions
of each open cone. I also noted the presence or absence of reflexed cone scales
on each monitored cone.
3.3 Survey of Cone Orientation
In order to determine whether Clarks Nutcrackers preferred to harvest seeds
from cones with a certain orientation with respect to the branch tip, e.g.

horizontally-directed cones, I surveyed cone orientations relative to supporting
branches for 100 southwestern white pine trees along the Williams Creek trail.
Frequencies of each cone position were estimated throughout the population using
the following methodology. I stood at a single point on the trail and counted all
of the visible cones for a given tree from that point. Counted cones were
categorized as: upright, angled at about 90 from above the branch tip; above
horizontal, angled about 45 or 135 above the branch tip; horizontal, directed
straight out from the branch tip; below horizontal, angled at about 45 or 135
below the branch tip; and pendulous, angled at about 90 below the branch tip
(Figure 3.1). Any damage to cones from seed harvest by Clarks Nutcrackers
(e.g. frayed scales) was also noted.
Figure 3.1. Key for identifying cone positions on trees. Positions one & five are
horizontal, two & four are above horizontal, three is upright, six & eight are
below horizontal, and seven is pendulous.
270 1
5 90

3.4 Seedling/ Sapling Transect
Multi-genet tree clusters are one signature of seed dispersal by animals,
with each trunk deriving from a single seed in a cache (Tomback and Linhart
1990). On 26 September, I conducted a survey of these growth forms along a
transect by following the Williams Creek trail for 1.5 km through the study area.
I noted all seedlings and saplings within 3 m to the left and right of the trail, and
the number of stems for each seedling or sapling site. I also counted each tree
within 3 m of the trail and noted the number of stems per tree site. For
seedlings, I determined if root systems were separate, or if produced by branching
from a single trunk, by probing around each stem. Separate individuals confirmed
the occurrence of a tree cluster.
3.5 Focal Observations of Animals Foraging on Southwestern White Pine
Daily observations were conducted using 10x50 binoculars to determine
which diurnal animal species were foraging on seeds from southwestern white
pine cones. I conducted observations for approximately eight hours per day,
beginning approximately one hour after sunrise (08:00) and continuing until late
afternoon when decreasing light conditions hindered observations. For each
observation, the following data were recorded: species, length of foraging bout,
position of cone on tree, presence or absence of reflexed cone scales, stage of

openness of each cone, tree DBH, and slope aspect of tree. I also noted whether
seeds were eaten or transported from the area in the animals mouth (e.g., rodent),
sublingual pouch (e.g., nutcracker), or esophagus (e.g., Stellers Jay). If
transported seeds were cached, I noted the following cache site characteristics:
location in study area, number of seeds per cache, ground substrate, slope aspect,
slope steepness, and presence or absence of southwestern white pine seedlings in
the area.
4. Results
4.1 Foliage and Cone Measurements
I collected data from seven characters for 13 trees (n = 23 cones), from
which means and standard deviations were calculated (Table 4.1). Two cones
collected early in the study had undeveloped seeds, and were eliminated from the
measurements. Cone length, wing length, and total seed weight for the trees at
my study areas exceeded those measurements reported by Steinhoff and Andresen
(1971) for southwestern white pine (Table 4.1). Average needle lengths for
southwestern white pine were less than those reported by Steinhoff and Andresen
(1971) for southwestern white pine, but greater than for limber pine. Shorter pine
needle lengths at my study areas suggest some genetic introgression with limber

Character Limber Pine1 Southwestern White Pine1 Study Area Southwestern White Pine
Avg. SE Avg. SE Avg. SE
Needle length (mm) < 552.6 > 703.2 65.11.1
Cone Length (mm) < 1003.4 >1104.3 129.14.9
Wing Length (mm) 2-6 2-6 4.30.2
Total Seed Weight (mg) < 14010.8 > 14020.9 150.00.0
Additional Data Endosperm Weight (mg) Seed Coat Thickness (mm) Number of Seeds per Cone 60.00.0 0.40.0 65.07.0
Table 4.1. Comparison of cone and foliage measurements for typical southwestern
white and limber pine with southwestern white pine from study areas in the San
Juan Mountains of southern Colorado.
Steinhoff and Andresen (1971)

Of the 100 fallen cones examined along the Williams Creek Trail, 63 had
reflexed cone scales; all of the cones had peduncles. Of the other 100 cones
examined at Study Area A, 98 had peduncles and 66 had reflexed cone scales.
All of the 25 cones collected from the 1999 cone crop had peduncles and 18
(72.00%) had reflexed cone scales. Thus, the majority of cones at my study sites
had reflexed cones scales, and only two cones out of the 225 cones (0.89%)
examined did not have peduncles.
DBH measurements for southwestern white pine trees ranged from 10.5
cm to 58.6 cm. The average DBH per tree was 34.7 cm (n = 90 trees, std. dev. =
10.4). Slope aspects obtained for 111 trees ranged from 90 to 248 at the study
4.2 Cone Opening Phenology
On 30 August when the cone opening survey was initiated, only 11 out of
207 cones showed scale separation. These 11 cones were on two different trees at
Study Area C. Cones with separating scales did not occur at Study Area A until
the third survey on 7 September, when 16 out of 199 cones showed evidence of
scale separation; eight cones were missing from the trees, cut down by pine
squirrels. I attributed missing cones to removal by pine squirrels and not
nutcrackers, because pine squirrels were observed daily removing cones from
trees throughout the study. Nutcrackers had difficulty handling southwestern

white pine cones due to cone weight, and I saw nutcrackers cut down only two
cones throughout the entire study. On 11 September, one cone at Study Area C
was partly open, 35 cones showed scale separation, 136 were still tightly
closed, and 35 cones had been cut down from the trees. I did not see open
cones until 25 September, when a total of eight cones were open on four
different trees, thirteen cones were partly open, 55 were classified as scales
separating, and 33 were still tightly closed. Only one cone was open at
Study Area A. Three of the open cones at Study Area C showed evidence of
foraging by Clarks Nutcracker. By 25 September, 94 of the original 207 cones
had been cut down from the monitored trees. On 30 September, the last survey
period, 14 cones were open, eight were partly open, 27 showed scale
separation, 4 cones were still tightly closed, and 154 cones had been cut down
from the trees. Figures 4.1, 4.2, and 4.3 show the progression of ripening at both
study areas, and the percentage of cones removed by pine squirrels throughout the
4.3 Survey of Cone Orientation
On 11 September, I examined the position of 637 cones on 100
southwestern white pine trees along the Williams Creek Trail. Below
horizontal cones comprised the largest percentage of cones, with horizontal

Study Area A
Figure 4.1. Cone Opening Phenology for Study Area A.

Study Area C
Closed cones
Scales separating
Scales partly open
Cones fully open
16 Sep
25 Sep 30 Sep
Figure 4.2. Cone Opening Phenology for Study Area C.

Figure 4.3. Number of cones removed by squirrels.

cones having the second highest frequency in the population. Upright was the
least frequent/ common cone position in the population (Figure 3.1, Table 4.2).
Frayed cone scales were counted as evidence of foraging by Clarks Nutcrackers.
Approximately 10 percent of the counted cones showed evidence of foraging
activity by nutcrackers, and horizontal cones were preferred for foraging before
cones opened (Table 4.2).
4.4 Seedling/ Sapling Transect
Of 95 tree sites surveyed, 79 trees had a single stem, 15 had two stems,
and one had three stems. I surveyed a total of 85 seedling and sapling sites. I
noted 71 single-stemmed, six double-stemmed, and five triple-stemmed seedlings
or saplings. I also found three seedlings with five, six, and seven stems at the
base. Stem clusters at my study areas provide strong evidence of seed
germination as a result of animal caches (Linhart and Tomback 1985, Fumier et
al. 1987, Carsey and Tomback 1994).
4.5 Foraging by Clarks Nutcracker
On 25 August, there was already evidence of foraging by
nutcrackers on closed cones. Clarks Nutcracker was the most frequently
observed species foraging in southwestern white pine trees. I observed up to 20
individuals at one time foraging within study areas, but group size usually

Cone Position Number of Cones Percentage of Population Number of cones Percentage of with nutcracker cones with damage nutcracker damage
Upright 4 0.6 2 3.2
Above Horizontal 19 3.0 7 11.3
Horizontal 230 36.1 46 74.2
Below Horizontal 248 38.9 7 11.3
Pendulous 136 21.4 0 0
Total number of cones 637 100% 62 100%
Table 4.2. Cone position in a southwestern white pine population as determined
on 11 September 1999 with evidence of foraging activity by Clarks Nutcracker in
the San Juan Mountains of southern Colorado.

consisted of small flocks of three to eight birds. Birds usually foraged singly in
trees, but I did observe up to three individuals on the same tree without any
aggressive behavior. Nutcrackers were most active mid-morning, and again in
late afternoon. Nutcracker activity decreased in the middle of the day. When
foraging on closed cones, individual nutcrackers jabbed their bill between cone
scales, loosening scales to expose seeds. On scales separating, partly open,
and open cones, nutcrackers were able to pull seeds from the cones without
having to jab into them, thereby fraying scales.
I observed Clarks Nutcrackers foraging on southwestern white pine cones
for a total of 40,438 sec (11.2 hrs) throughout the study period. I observed
Clarks Nutcrackers as they ate 169 seeds and placed a total of 957 seeds in their
sublingual pouches, presumably for transport to caching sites. Ten seeds were
discarded by nutcrackers after bill clicking (Tomback 1998). I never saw
nutcrackers pull winged seeds from southwestern white pine cones, even though
winged seeds comprised 11.6 percent (175 out of 1511 seeds) of the collected
Seed extraction rates varied significantly among the four stages of cone
openness. Average extraction rates were highest (i.e. slowest) for closed cones,
and lowest (i.e. fastest) for partly open cones (Table 4.3). Increased extraction

Cone Position Closed Mean (S.D.) Scales Separating Partly Open Mean (S.D.) Mean (S.D.) Open Mean (S.D.)
Upright N = 3 43.5 (0.0) 48.6 (7.0) * *
Above Horizontal N = 7 92.8 (62.6) 32.1 (18.7) * *
Horizontal N = 69 68.8 (40.1) 53.1 (21.4) 7.6 (4.6) 7.9 (2.7)
Below Horizontal N = 27 80.3 (32.4) 29.8(18.1) 9.5 (4.7) 15.9(12.7)
Pendulous N= 14 123.0 (0.0) 7.7 (0.0) 6.7 (2.5) 7.2 (3.4)
Table 4.3. Average seed extraction rates by Clarks Nutcrackers for each stage of
cone openess and cone position for a population of southwestern white pine. In
each category, the first value is the mean (seconds per seed) for each seed
extracted. (ANOVA overall: df = 3, P = 0.000).

rates for open cones can be attributed to seed depletion, resulting in decreased
harvesting efficiency.
Clarks Nutcrackers spent a total of 19,016 sec (5.28 hrs) foraging on
closed cones, 16,119 sec (4.48 hrs) on scales separating cones, 560 sec (9.33
min) on partly open cones, and 2,194 sec (36.57 min) on open cones. The
remaining 2,549 sec (42.48 min) were spent on cones of unknown openness.
Clarks Nutcrackers preferentially foraged on horizontal cones throughout the
study (Table 4.4). Nutcrackers spent more than 15% of the observed time
foraging on pendulous cones. However, only two seeds were successfully
obtained from pendulous cones before the partly open stage of openness.
Cone Position Frequency of cones in population Frequency of foraging by nutcrackers
Upright 0.6% 2.2%
Above Horizontal 3.0% 4.5%
Horizontal 36.1% 54.2%
Below Horizontal 38.9% 22.9%
Pendulous 21.4% 16.2%
Table 4.4. Southwestern white pine cone orientation preferences for Clarks
Nutcrackers throughout the entire study period (x2 40.7, df = 4).

On two separate occasions, 25 August and 8 September, I saw individual
nutcrackers cut down a closed, pendulous cone from a southwestern white
pine tree. On 27 August and 9 September, I saw an individual nutcracker set a
fallen closed cone up against a log for support. The nutcracker then proceeded
to jab into the cones with its bill, fraying cone scales, before pulling out seeds.
On 26 September, I observed nutcrackers dropping seeds while removing
them from open cones. These seeds were dropped before any bill manipulation
or clicking occurred (Tomback 1998). I also noted that seeds fell from open
cones, while nutcrackers were foraging on other cones on nearby branches. On
27 September, I observed nutcrackers having difficulty foraging on pendulous
cones. Two different nutcrackers fell from pendulous cones while attempting to
take seeds from them.
I documented seven seed caches made by Clarks Nutcrackers throughout
the study. Cache sizes ranged from one to five seeds per cache. The first cache
was made on 26 August at Study Area B. One seed was cached in rocky substrate
overlying moist dirt (slope aspect 150). Three caches, totaling eight seeds, were
made in succession by one bird on 10 September at Study Area B. The seeds were
removed from an above horizontal, scales separating cone. All three caches
were on a rocky slope (slope aspect unknown, slope steepness 26) with heavy

forest litter. Southwestern white pine seedlings grew near the cache sites. I
observed two caches in Study Area A on 13 September, on a steep, rocky ridge
(slope steepness 34, aspect 198) by the same individual. Mature trees on the
slope were predominantly southwestern white pine, but no southwestern white
pine seedlings occurred anywhere near the caches. The majority of seeds
pouched by nutcrackers were transported to other cache sites across Williams
During my study, there was a pair of Sharp-shinned Hawks (Accipiter
striatus) nesting on a steep ridge at Study Area A. Whenever nutcrackers were in
the area, the hawks exhibited aggressive behavior, resulting in decreased activity
by nutcrackers. Nutcrackers ceased foraging and calling activity when alarm
calls were given by the hawks. On 28 September, I observed two nutcrackers and
a Stellers Jay on a single southwestern white pine tree without either species
displaying agonistic behavior.
4.6 Foraging by Stellers Jays
Stellers Jays were observed on 14 separate occasions for a total of 632
sec (10.5 min) foraging for seeds on southwestern pine trees. On two different
occasions, 5 and 29 September, I observed a jay eating a single pine seed. On 5
September, a Stellers Jay displaced a Clarks Nutcracker foraging on a closed,
horizontal pine cone. The cone had frayed scales from previous foraging by a

nutcracker. The Stellers Jay then took an exposed seed from the closed cone
and ate it. One jay took three seeds lying on foliage on a southwestern white pine
tree on 28 September. I did not observe the initial placement of the seeds in the
foliage. The jay did not eat the seeds, but placed all three in its mouth for
transport. On 28 September, I observed a Stellers Jay eating several
southwestern white pine seeds, although I did not observe the jay extracting the
seeds from a cone. On 28 September, after a nutcracker foraged on cones of a
southwestern white pine tree, a Stellers Jay flew in and searched the ground
under the canopy, presumably for seeds.
4.7 Foraging by Other Birds
I observed a Hairy Woodpecker (Picoides villosus) attempting to open a
southwestern white pine cone for seeds for 4.6 min on 8 September. The cone
was scales separating and horizontal: no seeds were extracted by the
woodpecker. A Red-breasted Nuthatch (Sitta canadensis) spent a total of 15 sec
on three horizontal, open cones on 27 September; all three cones showed
evidence of previous nutcracker activity. One seed was extracted and eaten in 15
sec. Williamsons Sapsuckers (Sphyrapicus thyroideus) were observed on four
occasions, two on 10 September and two on 11 September, probing between cone
scales of horizontal, scales separating cones. No seeds were removed by

Williamsons Sapsuckers in the 681 sec (11.4 min) of observations. All of the
cones showed signs of previous nutcracker activity.
4.8 Foraging by Pine Squirrels
Pine squirrels were observed on southwestern white pine trees for a total
of 4,762 sec (1.3 hrs). I only documented pine squirrel activity while they were in
the canopy of trees. Once they cut down cones and left the canopy, I did not
continue observing them and did not document individual squirrel movement to
middens (extensive areas piled with cone debris of many years). I observed pine
squirrels removing 75 entire cones between 31 August and 27 September. Sixty-
four cones were closed, three cones were scales separating, one cone was
partly open, and seven cones were moved to middens before the stage of
openness was determined. Pine squirrels took an average of only 63.5 sec to cut
an entire southwestern white pine cone from a tree. Of the 75 cones cut down by
pine squirrels, 65 cones (86.67%) did not have reflexed cone scale apophyses. I
did not document any foraging activity or cutting by pine squirrels on closed
cones after 27 September. On several occasions, I saw pine squirrels cutting
down entire branch tips with intact cones and foliage attached.
4.9 Foraging by Chipmunks
I observed Colorado Chipmunks (Tamias quadrivittatus) foraging on
southwestern white pine cones throughout the study period. Typical foraging

behavior included the following activities: pulling off individual cone scales with
their teeth, and removing and subsequently eating the exposed seeds while still in
the tree canopy. The bare cone core was left on the tree after the chipmunk
removed all of the cone scales and seeds.
On 2 September, I observed a chipmunk eating seeds from one upright
cone at the top of a tree for 1,205 sec (20.1 min). Another chipmunk spent 377
sec (6.3 min) searching several pine cones with evidence of previous nutcracker
activity on 10 September; I saw the chipmunk eat only one seed that day. On 13
September, I observed a chipmunk foraging on a closed, horizontal cone.
The chipmunk was unable to extract any seeds, even though the cone showed
evidence of previous nutcracker activity.
By the end of the study, I noted seeds falling from open cones when
winds were strong, and when nutcrackers were foraging on cones. I observed
chipmunks foraging on the ground under the canopies of southwestern white
pines, possibly searching for fallen seeds. On three occasions, between 27 and 30
September, I saw chipmunks removing fallen seeds from the ground. I was
unable to verify the number of seeds gathered by the chipmunks, but seeds were
clearly held in the chipmunks cheek pouches, presumably for transport.

5. Discussion
Foliage, seed and cone measurements are consistent with expectations for
southwestern white pine. All measurements matched Steinhoff and Andresen
(1971). However, intermediate needle lengths suggest some gene flow between
limber and southwestern white pine at my study areas, a finding consistent with
those of previous studies in areas of sympatry (e.g., Steinhoff and Andresen
Clarks Nutcracker was the most important primary disperser of
southwestern white pine seeds at my study sites. I recorded hundreds of seeds
pouched by nutcrackers for transport. The nutcrackers then transported the seeds
across Williams Creek, presumably to a communal cache site that gave them good
access to seed stores. Previous studies have shown a mutualistic relationship
between nutcrackers and several species of large-seeded pines (Vander Wall and
Baida 1977, Hutchins and Lanner 1982, Tomback 1982, 1998, Tomback et al.
1990). In years with large cone crops, nutcrackers may harvest more seeds than
are energetically required (Vander Wall and Baida 1977, 1981, Tomback 1982).
Southwestern white pine seeds not recovered from nutcracker caches may
germinate, particularly after summer rains, leading to tree recruitment. I propose
that in areas of sympatry, nutcrackers are primarily responsible for seed dispersal
leading to tree recruitment of southwestern white pine.

Stellers Jays foraged on southwestern white pine cones, but less
frequently than nutcrackers. Stellers Jays were unable to open tightly closed
cones of southwestern white pine, and had to wait for cones to begin to open to
extract seeds. Thus, Stellers Jays are not able to disperse as many seeds as
nutcrackers. Stellers Jays may be responsible for primary dispersal of
southwestern white pine seeds, but to a lesser degree than nutcrackers where
Stellers Jays and nutcrackers co-occur.
Some pines experience both primary and secondary seed dispersal for tree
recruitment (Vander Wall 1992). Stellers Jays were observed on several
occasions searching for seeds through southwestern white pine canopies, as well
as on the ground below the trees; Stellers Jays had difficulty taking seeds directly
from pine cones. As cones opened, I noticed an increasing number of seeds
falling from cones to tree limbs and to the ground. I also saw a jay place seeds in
its mouth and transport them away from the parent tree, presumably to cache.
Although, I did not observe this jay caching seeds, transported seeds are usually
cached by Stellers Jays (Greene et al. 1998). Stellers Jay is the only species that
occurs throughout the entire range of southwestern white pine, and may play a
role in tree recruitment, both as a primary and secondary seed disperser.
Secondary dispersal by corvids is known to result in seedling establishment of
some pines (Tomback 1978, Vander Wall 1992). I suggest that Stellers Jays and

their role in secondary dispersal may become more important in areas without
Food-hoarding mammals have the capacity to quickly remove the many
wind-dispersed seeds that fall around a tree (Vander Wall 1992). Vander Wall
(1993) documented chipmunks (Tamias spp.) carrying pine seeds away from
source trees, and subsequently burying them in the ground in microhabitats where
they were more likely to establish new seedlings. Cache depths of scatter-
hoarding rodents are often within the range of depths suitable for plant
establishment (Vander Wall 1993). When foraging on southwestern white pine
cones in tree canopies, chipmunks were observed eating all of the exposed seeds.
However, chipmunks were also observed picking up seeds from the ground below
tree canopies, and subsequently moving them away from the seed source.
Chipmunks may cache some secondarily dispersed seeds. The number of cached
seeds that escape detection by animals is likely to be small (Vander Wall 1994),
but some cached seeds by chipmunks may lead to tree recruitment for
southwestern white pine.
The occurrence of tree cluster seedlings, saplings, and trees is a signature
of tree recruitment resulting from animal caches (Tomback and Linhart 1990).
The occurrence of three growth forms is characteristic of bird-dispersed pines: 1)
single-trunk tree, 2) multi-trunk tree (multiple trunks of a single genet), and 3)

tree cluster (a cluster of two or more genets) (Tomback et al. 1990); all three
growth forms were present at my study areas. Future genetic analyses for
saplings and mature trees are necessary to determine whether a tree clump, or a
clump of stems, is a multi-trunk tree or a tree cluster (Carsey and Tomback 1994).
However, I was able to determine the presence of multiple individuals at seedling
sites throughout the study areas. These findings confirm tree recruitment
resulting from animal caches for southwestern white pine at my study areas.
Pine squirrels were observed removing intact southwestern white pine
cones at all stages of openness from trees. Cone removal by pine squirrels
eliminates access by primary seed dispersers. Pine cones removed by pine
squirrels are usually stored intact in middens, usually at the base of a tree.
Although seeds may remain viable for a year or longer, they remain in cones
(Finley 1969). Seeds from cone middens of Tamiasciurus squirrels rarely are in
an environment (e.g., substrate) suitable for germination (Hutchins and Lanner
1982). Pines that might get established in the midden are likely to die of root
damage due to the constant digging-up and churning of midden material by the
squirrel (Hutchins and Lanner 1982). Also, southwestern white pine seedlings
may not receive enough sunlight in middens (Finley 1969); furthermore, seed
dispersal distances from parent trees are restricted by squirrel home range size
(Smith 1970). Benkman et al. (1984) found that pine squirrels were the most

important seed predators at low elevations. Pine squirrel activity was greater at
low elevations at Study Area A, where I marked trees for the phenology study; I
rarely saw nutcrackers foraging on trees at these elevations. Pine squirrels are
definitely an important seed predator of southwestern white pine in areas of
Nutcrackers preferentially foraged on horizontal cones throughout the
study (Table 4.4). Average seed extraction rates decreased as cones opened,
exposing their seeds. Extraction rates later increased, when seed availability
dropped (Table 4.3). Nutcracker harvesting efficiency was actually greatest on
pendulous, partly open cones. Nutcrackers had difficulty foraging on
pendulous cones, before the partly open stage; decreased (i.e., faster)
extraction rates on these cones can be attributed to seed availability.
Benkman et al. (1984) described four possible patterns of cone ripening
phenology depending on combinations of synchrony and asynchrony within and
among tree canopies of conifers. Each pattern has different consequences for
animal dispersers. Synchrony within a tree and asynchrony among trees is a
pattern predicted by Benkman et al. (1984) to be the most beneficial to dispersers
such as nutcrackers. Synchronous ripening within a tree allows a disperser to
recognize profitable food trees, and to return to a given tree and remove the most
seeds while revisiting the fewest cones. Asynchronous ripening among trees

lengthens the harvest period, preventing saturation of disperser species, which
decreases the number of seeds that fall to the ground where germination is less
likely and predation is higher. Synchronous ripening within and among tree
canopies should be limited to wind-dispersed pines, due to the likely saturation of
the dispersal agent and increased seed loss to predators. According to Benkman
et al. (1984), total asynchrony, and synchronous ripening among trees and
asynchronous ripening within a tree should not be found in animal-dispersed
conifers. These two patterns increase the time for dispersers to harvest seeds, but
should also reduce harvesting efficiency. These patterns should only occur if
cones ripen in a predictable pattern on trees and ripe cones are readily
recognizable by dispersers (Benkman et al. 1984).
Benkman et al. (1984) evaluated cone ripening phenology in southwestern
white pine based on the relationship between seed number and cone mass. As
cones ripen, cone mass increases, but the number of seeds remains the same. In
ripe cones, the number of seeds per gram cone is nearly constant, and a strong
correlation between cone mass and seed number is expected. They suggested that
southwestern white pine cones ripened synchronously among and within
individual trees, a pattern characteristic of wind-dispersed conifers. I determined
the cone opening phenology for southwestern white pine by assessing the degree
of cone scale separation through visual observations, as opposed to changes in

cone mass (e.g., Benkman et al. 1984). Differences in phenology patterns within
a species may be found when assessing visual cues (e.g., degree of scale
separation) versus changes in cone mass. I chose to use cone characters because
seed dispersers probably use these kinds of visual cues to ascertain cone and/ or
seed ripeness. I found an asynchronous cone opening phenology among and
within tree canopies. As predicted, my findings differed from those of Benkman
et al. (1984).
Trees with seeds that are animal-dispersed should be expected to maintain
their seeds as long as possible, thereby providing dispersal agents with ample time
to harvest seeds (Vander Wall and Baida 1977). Contrary to Benkman et al.
(1984), I postulate that total asynchrony actually promotes avian seed dispersal.
Tomback and Kramer (1980) studied the cone opening phenology in limber pine
in the Sierra Nevada of California, and categorized limber pines cone opening
phenology based on degree of cone scale separation. They determined total
asynchrony as the pattern of cone ripening for limber pine, which is dispersed by
nutcrackers. Vander Wall and Baida (1977) also found an asynchronous opening
phenology in pinon pine (P. edulis), another bird-dispersed pine. Birds and small
mammals visually inspect cones before extracting seeds, with cone openness most
likely recognized by these animals. Therefore, determination of cone opening

phenology and seed availability in future studies may be more accurate if assessed
through cone characters, and not cone mass.
Cone and seed morphology of conifers is influenced by predation pressure
and mode of dispersal. In the San Juan Mountains of southern Colorado, primary
seed dispersal is accomplished mainly by Clarks Nutcrackers, while pine
squirrels are the main pre-dispersal seed predators. Cones and seeds of
southwestern white pine at my study areas had characteristics that facilitated
harvest by nutcrackers and deterred predation by pine squirrels.
Characteristics of southwestern white pine consistent with other bird-
dispersed pines include: large, wingless seeds (occasional remnant wings);
spineless cones; thin seed coats; and an asynchronous cone opening phenology.
Characteristics that deter squirrel predation include reflexed cone scale apophyses
and resinous cones (Smith 1970). Reflexed cone scales decrease cutting
efficiency and subsequent seed removal by pine squirrels, and resinous cones
increase handling times (Smith 1970).
However, not all cone and seed traits of southwestern white pine at my
study areas deter predation by pine squirrels. Thin seed coats also decrease
handling times by pine squirrels (Smith 1970), and not all cones had reflexed cone
scale apophyses. Pine squirrels preferred to cut down cones without reflexed

scale apophyses. Long peduncles, characteristic of cones at the study areas, also
facilitate cone cutting by pine squirrels (Smith 1970).
Pine squirrels do not occur throughout the range of southwestern white
pine. Tamiasciurus hudsonicus occurs south through the Rocky Mountains to
New Mexico, but is absent from southwestern New Mexico and southeastern
Arizona (see Smith 1970). Smith (1970) argued that cone morphology largely
functions to increase seed dispersal by decreasing seed predation. This study, as
well as others (Benkman 1995b), demonstrates that in areas of sympatry, pine
squirrels and nutcrackers have important and conflicting selective impacts on
conifer cone and seed morphology. Future studies should analyze cone and seed
traits of southwestern white pine in areas without nutcrackers and/ or pine
squirrels to ascertain any morphological differences due to changes in selection

Benkman CW (1995a) Wind dispersal capacity of pine seeds and the evolution of
different seed dispersal modes in pines. Oikos 73: 221-224.
Benkman CW (1995b) The impact of tree squirrels (Tamiasciurus) on limber pine
Seed dispersal adaptations. Evolution 49: 585-604.
Benkman CW, Baida RP, Smith CC (1984) Adaptations for seed dispersal and the
compromises due to seed predation in limber pine. Ecology 65: 632-642.
Carsey KS, Tomback DF (1994) Growth form distribution and genetic
relationships in tree clusters of Pinus flexilis, a bird-dispersed pine.
Oecologia 98: 402-411.
Critchfield WB, Little EL, Jr. (1966) Geographic Distribution of the Pines of the
World. U.S. Dept, of Agric., Misc. Publ. 991, Washington, D.C.
Finley RB, Jr. (1969) Cone caches and middens of Tamiasciurus in the Rocky
Mountain region. Univ. Kans. Mus. Nat. Hist., Misc. Publ. 51: 223-273.
Fumier GR, Knowles P, Clyde MA, Dancik BP (1987) Effects of avian seed
dispersal on the genetic structure of whitebark pine populations. Evolution
Greene E, Davison W, Muehter VR (1998) Stellers Jay (Cyanocitta stelleri). The
Birds of North America, No. 343 (Poole A, Gill F, eds.). Philadelphia:
Acad of Nat Sciences; Washington, D. C.: The American Ornithologists
Hutchins HE, Lanner RM (1982) The central role of Clarks Nutcracker in the
dispersal and establishment of whitebark pine. Oecologia 55: 192-201.
Lanner RM (1980) Avian seed dispersal as a factor in the ecology and evolution
of limber and whitebark pines. Proc Sixth North Am For Biol Workshop,
pp. 15-48. Univ. of Alberta, Edmonton, Alberta, Canada.
Lanner RM (1998) Seed dispersal in Pinus. Ecology and Biogeography of Pinus,
pp. 281-295. Cambridge Univ. Press, New York, NY.

Lanner RM, Vander Wall SB (1980) Dispersal of limber pine seed by Clarks
Nutcracker. J For 78: 637-639.
Linhart YB, Tomback DF (1985) Seed dispersal by Clarks Nutcracker causes
multi-trunk growth form in pines. Oecologia 67:107-110.
Ligon JD (1978) Reproductive interdependence of Pinyon Jays and pinon pines.
Ecol Monogr48: 111-126.
Pijl Van der L (1972) Principles of dispersal in higher plants. Second edition,
Springer Verlag, New York, NY.
Smith CC (1970) The coevolution of Pine Squirrels (Tamiasciurus) and conifers.
Ecol Monogr 40: 349-371.
Steinhoff RJ, Andresen JW (1971) Geographic variation in Pinus flexilis and
Pinus strobiformis and its bearing on their taxonomic status. Silv Genet
20: 159-167.
Tomback DF (1978) Foraging strategies of Clarks Nutcracker. Living Bird 16:
Tomback DF (1982) Dispersal of whitebark pine seeds by Clarks Nutcracker: a
mutualism hypothesis. J Anim Ecol 51: 451-467.
Tomback DF (1994) Ecological relationship between Clarks Nutcracker and four
wingless-seed Strobus pines of western North America. ProcIntern
workshop on subalpine stone pines and their environment: the state of our
knowledge (Schmidt WC, Holtmeier FK, comp.) USDA Forest Service
Intermountain Res. Sta. INT-6TR-309, Ogden, UT.
Tomback DF (1998) Clarks Nutcracker (Nucifraga Columbiana). The Birds of
North America, No. 331 (Poole A, Gill F, eds.). Philadelphia: Acad of
Nat Sciences; Washington, D.C.: The American Ornithologists Union.
Tomback DF, Kramer KA (1980) Limber pine seed harvest by Clarks Nutcracker
in the Sierra Nevada: timing and foraging behavior. Condor 82: 467-468.
Tomback DF, Linhart YB (1990) The evolution of bird-dispersed pines. Evol

Eccil 4: 185-219.
Tomback DF, Hoffman LA, Sund SK (1990) Coevolution of whitebark pine and
nutcrackers: Implications for forest regeneration. Pages 118-129.
(Schmidt WC, McDonald KJ, compilers). Proc-Symposium on whitebark
pine ecosystems: Ecol and Management of a high mountain resource.
USDA For Serv Interm Res Sta, Gen Tech Report INT-270, Ogden, Utah.
Vander Wall SB (1988) Foraging of Clarks Nutcrackers on rapidly changing pine
seed resources. Condor 90: 621-631.
Vander Wall SB (1992) The role of animals in dispersing a wind-dispersed
pine. Ecology 7: 614-621.
Vander Wall SB (1993) A model of caching depth: Implications for scatter
hoarders and plant dispersal. Am Nat 141: 217-232.
Vander Wall SB (1994) Removal of wind-dispersed pine seeds by ground-
foraging vertebrates. Oikos 69: 125-132.
Vander Wall SB, Baida RP (1977) Coadaptations of the Clarks Nutcracker and
pinon pine for efficient seed harvest and dispersal. Ecol Monogr 47: 89-
Vander Wall SB, Baida RP (1981) Ecology and evolution of food-storage
behavior in conifer-seed-caching corvids. Z Tierpsycol 56: 217-242.
Vander Wall SB, Hutchins HE (1983) Dependence of Clarks Nutcracker,
Nucifraga columbiana, on conifer seeds dining postfledging period. Can
Field Nat 97: 208-214.