HomeMy WebLinkAboutAPA3178._,
RED SQUIRREL (TAMIASCIURUS HUDSONICUS) ECOLbGY
DURING SPRUCE CONE FAILURE IN ALASKA
APPROVED: ''
APPROvru:---,-3 Anlg. ~
Dean of the College of Biological
Sciences and Renewable Resources
Vice President for Research and
Advanced Study
Chairman
Department Head
DATE:
RED SQUIRREL (TAMIASCIURUS HUDSONICUS) ECOLOGY
DURING SPRUCE CONE FAILURE IN ALASKA
A
THESIS
Presented to the Faculty of the
University of Alaska in Partial Fulfillment
of the Requirements
for the Degree of
MASTER OF SCIENCE
By
Michael Charles Smith, B.S.
May, 1967
-
ABSTRACT
Observations were made on a red squirrel (Tamiasciurus
hudsonicus) population in a mature white spruce (Picea glauca) forest
near Fairbanks, Alaska, during two years of spruce cone crop failure
(July, 1964, to April, 1966). An adequate supply of old spruce cones,
cached in previous years, was available during the first winter. A
67% drop in numbers of the squirrel population followed the second
crop failure with the remaining squirrels utilizing spruce buds as
their primary food during the winter. Stomach analyses revealed that
when present, spruce seed is the major constituent in the diet. In
its absence, heavy utilization of mushrooms in summer and spruce buds
in winter occurs. Feeding trials conducted with captive red
squirrels in March, 1965, and April, 1966, showed that about 194 old
cones per day were necessary to sustain a squirrel, approximately
35% more than for cones from the current year's crop. Three squirrels
survived for eight days on a diet of only white spruce buds. Analysis
of old spruce cones showed that 31% of the seed was potentially
viable (filled), but that only 1.4% of the seed germinated. Calori-
metrie determinations of old seed (minus coat), spruce buds, and
mushrooms yielded values of 5,976, 4,986, and 4,552 cal/g
respectively. Excavation of middens revealed up to 8,518 old, cached
cones per midden, despite a crop failure. In years of normal cone
production, squirrels may eut and cache 12,000 to 16,000 cones; the
excess accrues each year and eventually a sufficient supply exists to
maintain the squirrels through a winter following a cone crop failure.
iii
ACKNOWLEDGMENTS
The author wishes to extend his sincere appreciation to the
following people for their assistance during the course of the study
and in the preparation of the manuscript:
Dr. Frederick C. Dean, Head, Department of Wildlife Management,
for his suggestions, guidance, and patience throughout the study.
Dr. David R. Klein, Leader, Alaska Cooperative Wildlife Research
Unit, for his always helpful advice and for critically reading the
manuscript.
Samuel J. Harbo for his invaluable assistance with the
statistical aspect of the study and for critically reading portions
of the manuscript.
Dr. Russell D. Guthrie, Assistant Professer of Zoology, for
critically reading the manuscript.
Robert A. Gregory, Research Forester, U. S. Forest Service,
Northern Forest Experiment Station, for helpful suggestions in many
areas of the study. Dr. Leslie A. Viereck, Research Botanist,
Northern Forest Experiment Station, for identifying the lichens and
masses collected on the study area.
Mrs. Susan Savage, Institute of Arctic Biology, for patiently
instructing me in the use of the Institute's bomb calorimeter.
Timothy Howe for his assistance in surveying and marking the
grid system on the Bonanza Creek study area.
Mrs. Sylvia Kolivosky for typing the final manuscript.
iv
'•
TABLE OF CONTENTS
Page
INTRODUCTION • . . . . . 1
STUDY AREA • . . 2
METHODS . . . . 8
Field Study . . . . . . . . . 8
Colleeting and Marking • . . . . . . . . . . . 8
Stomaeh Analyses • . . . . . . . . . . . 9
~ Analysis of Old White Spruee Seed . . . . . . 9 -
Calorimetrie Determinations . . . . . . . 10
.tl. Feeding Trials . . . . . . . . . . . . 11
RESULTS AND DISCUSSION . . . . . . . 14
Field Study . . . . . . . . . . . . 14
Food Habits . . . . . . . . . . . . . 19
l·_ Spruee Seed • . . . . . . . . . . . . 19
Fungi • . . . . . . . . . . . . . 23
L'~ Spruee Buds • . . . . . . . . . . 25
Miseellaneous Foods . . . . . . . . . . 29
Territoriality . . . . . . . . . . . . . 31
t. l~ Population Density . . . . . . . . . . . . 33
Analysis of Old White Spruee Seed . . . . . . . 36
~
\ .. _ Calorimetrie Determinations . . . . . . . . 37
Feeding Trials . . . . . . . . . . . . . . 39
Cane Consumption • . . . . . . . . . . . 54
1
L APPENDIX . . . . . . . . . . . . . 60
LITERATURE CITED . . . . . . . . . . . 66
v -
.__
1.-
LIST OF TABLES
Table Page
1. Density analysis of tree species, Bonanza Creek study
area 5
2. Major plant species on the Bonanza Creek study area 6
3. Birds and mammals observed on the Bonanza Creek study
a rea 7
4. Approximate percentages by volume of red squirrel
1
L stomach contents collected near the Bonanza Creek study
area from August, 1964, to March, 1966 • 20
L 5. Old canes excavated from nine middens adjoining the grid
system in October, 1964 • 22
6. Characteristics of red squirrel-cut white spruce tips
'-collected from two squirrel territories in February,
1966 28
L 7. Number of filled seed in 10-seed samples taken from
each of lOO randomly selected old white spruce canes • 37
:
L 37 8. Calorie values of three red squirrel foods
9. Response of red squirrels to constant diet of labora-
tory chow 40
L
10. Response of red squirrels to sudden change of diet to
old white spruce seed following constant diet of
L laboratory chow 43
'-
vi
'-
~ t
L
L
l
Table
11. Response of red squirrels to gradual change of diet
to old white spruce seed following constant diet of
laboratory chow
12. Response of red squirrels to gradual change of diet to
white spruce buds following constant diet of laboratory
chow
APPENDIX
I. Numbers of old white spruce cones consumed by Group II
during feeding trials •
II. Method used in counting cones for squirrel no. 6 •
III. Climatological data recorded during feeding trials
vii
Page
46
50
62
63
64
l
111
1
Jil
' UJ
'
' '
'
1 ,1,, u.
IL .
r
l
LIST OF FIGURES
Figure Page
1. Location of grid system within the Bonanza Creek
Experimental Forest • 3
2. Grid system and midden locations on the Bonanza Creek
study area • 4
3. Response of red squirrels to constant diet of
laboratory chow 41
4. Response of red squirrels to sudden change of diet to
old white spruce seed following constant diet of
laboratory chow 44
5. Response of red squirrels to gradual change of diet to
old white spruce seed following constant diet of
laboratory chow 47
6. Response of red squirrels to gradual change of diet to
white spruce buds following constant diet of
laboratory chow 52
7. Average percent weight change for individual feeding
trials 53
viii
.....
INTRODUCTION -This thesis represents the second part of a study aimed at
determining the ecological relationships of red squirrels -(Tamiasciurus hudsonicus) to the production and supply of white
spruce (Pièea glauca) seed and the regeneration of white spruce in
lia interior Alaska. The project was financed by the U. S. Forest
Service's Northern Forest Experiment Station (Contract No. 12-11-
010-1650) and administered through the Alaska Cooperative Wildlife
ililld
Research Unit at the University of Alaska at College. The time
period included in this study was from July, 1964, to April, 1966. -The thesis presents sorne aspects of red squirrel ecology
~ during two consecutive years of cone crop failure in a mature white
~ 1 -spruce forest in interior Alaska. Since red squirrels utilize
large quantities of white spruce seed as the major constituent of
'!! i ~ti
Ill ~ their diet during years of normal cone production (Brink 1964), this
study was oriented to determine food habits in the absence of a
~
'' ' fresh supply of cones. One phase deals with observations of natural
1 ..
conditions on a study area within a mature white spruce forest. The
r,
b
second phase consists of a series of feeding trials conducted with
illl ~ captive red squirrels within an outdoor enclosure on the University
.. L
of Alaska campus. The feeding trials were set up in such a way
that the data would be directly comparable to that obtained by
'f
Brink (1964) in a similar study.
lt -'! -1
f
1
" ll.i J..o
r Li
r
~JI
STUDY AREA
The study area was situated within the Bonanza Creek
Experimental Forest approximately 25 km southwest of College
(Fig. 1). The major portion of the field work was conducted within
a large stand of mature white spruce on a south-facing slope with
an average elevation of 305 rn above sea level. A permanent grid
system 457 rn square, or 21 ha (52 acres), was established within
this stand to facilitate the mapping of midden locations and terri-
tories (Fig. 2). The largest spruce are over 125 years old,
standing greater than 34 rn in height with a diameter at breast
height (d.b.h.) of 45 cm. Quaking aspen (Populus tremuloides) is
common around the small and scattered blowdown-caused openings in
the canopy. Alder (Alnus crispa) and paper birch (Betula papyrifera)
are the only other large vascular plants on the area. Two 30 X 30 rn
sample plots were randomly selected, and density and basal area
measurements were made on these tree species (Table 1). The forest
floor is a continuous carpet of moss (primarily Hylocomium splendens
and Dicranum fuscescens). Mushrooms are common and bracket fungi
are found on most paper birch throughout the area. Beard lichens
(Parmelia sulcata and f. physodes) are abundant on the lower dead
branches of the spruce. Table 2 contains a listing of the major
plant species on the study area.
All birds and mammals, or their sign, observed on the study
area are listed in Table 3.
2
if J
,,., ..
J
,,,
"
ê''"lil
:1, •
l •
J '
'
'~ ~i ..
J ' ,
l ' .ti
lill
"'~ J
'l JI
ii
J
'1
l' u
j
"""
1
1
1 0.62 1gn t
1.0 miles1
Fig. 1. Location of grid system within the Bonanza Creek Experimental Forest.
w
C7 • C1 •
C3 •
C2 •
•
•
C4 •
ce •
1
6
•
11
16
21
2 3
7 8
.
.12 13.
•
17 1 18
22 23
C5
•
4
9
14
19
24
5
10
15
• 20
• 25
SC ALE
91m
1 300ft
C9 •
Fig. 2. Grid system and midden locations on the Bonanza Creek
study area.
4
-
ca
•
~
111111 "
~. --
1 •
5
-Table 1. Density analysis of tree species, Bonanza Creek study area.
Density Average Basal Area'~
per Diameter'~ OM2 per (Ft2 per
Species Hectare Acre (cm) Hectare) Acre)
-Plot No. 1 (30 X 30 m)
Picea glauca 1042 422 21 43 132
R2Eulus tremuloides 183 74 34 17 52 -Alnus crispa** 64 26 -0.2 0.6
~ J!.SE.Y:rifera 11 4 10 0.1 0.3 ....
Plot No. 2 (30 X 30 m)
Picea glauca 849 344 27 56 169 -Populus .tremuloides 32 13 31 2.5 8
Alnus crispa 43 17 -0.2 0.6 -Average of Plots No. 1 and 2
~ glauca 946 383 24 49 150 -Populus tremuloides 108 44 34 10 30
~crispa 54 22 -0.2 0.6 -~ papxrifera 6 2 5 -0.2
*Measured 1.4 m (4.6 ft) above the ground. .. -**Since most alders divide into several main branches less than 1.5 m
above the ground, such plants were counted as one when calculating
density, but the diameter of each branch was used in calculating basal
area.
"" -
,..,
-
..
-
Table 2. Major plant species on the Bonanza Creek study area.
Masses
Hylocomium splendens
Dicranum fuscescens
Lichens
Peltigera aphthosa
Parmelia physodes
f. sulcata
Alectoria jubata
Usnea~
Vascular Plants
Equisetum pratense Horsetail
Lycopodium complanatum Ground Cedar
Picea glauca White Spruce
f. mariana Black Spruce
Platanthera obtusata Small Northern Bog Orchid
Populus tremuloides Quaking Aspen
~ papyrifera Paper Birch
Alnus crispa Green Alder
Geocaulon lividum Northern Comandra
Delphinium glaucum Glaucus Larkspur
Rubus idaeus Red Raspberry
Rosa acicularis Wild Rose (Rose hi.ps)
Epilobium angustifolium Fireweed
Cornus canadensis Bunchberry
Pyrola secunda One-Sided Pyrola
Vaccinium Vitis-~ Lowbush Cranberry
Galium boreale Northern Bedstraw
Viburnum ~ Highbush Cranberry
The scientific names of mosses follow Conard (1956). Those
of lichens follow Mason and Culberson (1960), and those of
vas~ular plants follow Hultén (1941-1950).
6
_.
.....
-
-
._...
:.......
......
,_
-
-
Table 3. Birds and mammals observed by the author on the Bonanza
Creek study area.
BIRDS
Accipiter gentilis* Goshawk
Canachites canadensis Spruce Grouse
Dendrocopos villosus Hairy Woodpecker
Q. pubescens Dotmy Woodpecker
Picoides tridactylus Northern Three-toed Woodpecker
Empidonax hammondii Hammond's Flycatcher
Perisoreus canadensis Gray Jay
Corvus ~ Common Raven
Parus atricapillus Black-capped Chickadee
f. hudsonicus Boreal Chickadee
Certhia familiaris Brown Creeper
Ixoreus naevius Varied Thrush
Regulus calendula Ruby-crowned Kinglet
Dendroica coronata Myrtle Warbler
Q. townsendi Townsend's Warbler
Pinicola enucleator Pine Grosbeak
Acanthis hornemanni Hoary Redpoll
A· flammea Common Redpoll
Junco hyemalis Slate-colored Junco
MAMMALS
Lepus americanus Snowshoe Rare
Tamiasciurus hudsonicus Red Squirrel
Glaucomys sabrinus Northern Flying Squirrel
Clethrionomys rutilus Northern Red-backed Vole
Erethizon dorsatum Porcupine
Vulpes fulva Red Fox
~ americanus Black Bear
Martes americana Marten
Lynx canadensis Lynx
Alces alces Moose
7
*The scientific names of the birds follow the 1957 check-list of
the American Ornithologists' Union, and those of mammals follow
Hall and Kelson (1959).
METHODS
Field Study
Twenty-four of the middens on the study area, 15 tvithin the
grid system and niue located outside but within 140 rn of the grid,
were each marked with fluorescent orange flagging, plotted as to
size and location, and repeatedly checked for activity throughout
the study (Fig. 2). No attempt was made to accurately determine the
exact size of the individual territories.
In October, 1964, the nine middens outside the grid system
were searched, and all canes which could be found were dug out and
removed. In November, 1965, four of the remaining six active
squirrels on the study area were live-trapped and marked to determine
if any change in midden ownership occurred during the winter. Aside
from this live-trapping and marking, the squirrels and middens
within the grid system were in no way disturbed. During the study,
over 300 hours were spent in observing red squirrels. Since the
squirrels quickly adjusted to being observed from a distance, it is
believed that these observations did not significantly affect normal
red squirrel behavior.
Collecting and Marking
No squirrels were collected on the study area as its population
was being constantly observed. However, squirrels were regularly
collected from a similar white spruce stand approximate1y 1 km from
the study area. Stomachs were immediately fixed in 10% forma1in
8
Wj
-
._
-
.._
......
9
for future analysis.
"Nyanzol A" dye was used in marking squirrels, and was
purchased from the Nyanza Co1or and Chemica1 Co., 109 Worth Street,
New York, N. Y., 10013, for $5.50 a pound (.45 kg) delivered. Two
grams of the dye were dissolved in 100 ml of distilled water. Just
before application, this solution was mixed in a ratio of 2:1 with
a 10 volume (3%) solution of hydrogen peroxide (Fitzwater 1941).
Cotton swabs were used for application, and the fur was allowed to
dry before the squirrel was released. Since just four squirrels
were marked, only one shoulder or hip was dyed on each squirrel.
Stomach Analyses
A total, of 29 red squirrel stomachs was analysed during the
study. The contents of each stomach were observed under a dissecting
scope, and approximate percentages by volume of the contents were
estimated. While it is often difficult to determine accurately red
squirrel stomach contents collected in areas with a greater variety
of foods, the usually monotonous diet of the red squirrels near the
study area permitted reasonable accuracy.
Analysis of Old White Spruce Seed
During February, 1965, a germination test to determine the
percentage of viable seed was conducted with the "old" canes used
during the feeding trials. A sample of 1,000 seeds from the
productive zone of lOO randomly selected old cones, collected on the
study area in the fall of 1964, was put on moist cotton gauze in
petri dishes and placed in a controlled temperature seed germinator
10
for 28 days. A self-timing deviee maintained the temperature between
28° and 32° C for 12 hrs and at approximately 22° C for the remainder
of the day. Brink (1964) described the productive zone of a white
spruce corre as lying between the upper six and lower two whorls of
bracts.
A second random sample of 1,000 old seeds, selected in the
same manner as described above, was analysed to determine the per-
centage of filled seed in the productive zone. Each seed was
opened and the endosperm examined. Seed containing even shriveled
endosperm was considered filled. Hollow or resin filled seed was
recorded as unfilled.
Calorimetrie Determinations
Approximate calorie values were determined for three of the
foods utilized during the study: the old white spruce seed, white
spruce buds, and mushrooms. The seed was taken from con:es collected
from middens on the study area in October, 1964. Only endosperm and
embryo, the part of the seed the squirrels actually eat, were used.
The bud samples were eut from the crowns of four white spruce near
the study area in March of 1966. Since squirrels discard the outer
sheath of bud scales and eat only the inner portion of undifferenti-
ated meristematic tissue, only the latter was used. Both the white
spruce seed and buds were kept frozen until tested. The mushrooms
were picked in the summer of 1965, and were~preserved by drying.
The mushroom values represent a ground up mixture of pilei from six
species of mushrooms known to be eaten by the squirrels on the study
lê
.J
-
.....
-
~
-
i.-
--
_,_
11
area. The measurements were made in a Parr adiabatic oxygen bomb
calorimeter, model 1221, after all food samples had been dried for
10 days in an oven at 750 C.
Feeding Trials
The feeding trials were conducted within a 30.5 X 30.5 m wire
enclosure in a black spruce (Picea mariana) stand about 1.6 km
northwest of the main campus. The squirrels were kept individually
in 61 X 61 X 122 cm galvanized wire cages each furnished with a
30 X 30 cm wooden nest box. See Brink (1964) for construction
details.
An Ohaus triple bearn balance reading to the nearest 0.1 g was
used for weighing, and the weights are believed accurate to within
1 g. Squirrels were weighed in live-traps of known weight. The
weights were recorded at two-day intervals for Groups I, II, and
III, and daily for Group IV. Weighing was done close to sunset when
the squirrels had finished feeding and were in their nest boxes.
After covering the small opening in the nest box, the box was removed
from the cage to the weighing table where the squirrel could be
easily frightened into a live-trap.
The squirrels used for the feeding trials were captured in
National Live Traps within 3 km of the University campus. All
squirrels were taken from white spruce middens and were captured
from 5 to 36 days before the trials begàn. During this period and
when not participating in a feeding trial, the squirrels were main-
tained on Purina Laboratory Chow which was determined to have a
12
calorie value of approximately 4,390 cal/g (Brink 1964). Snow was
provided as a water source.
Feeding trials with Groups I, II, and III were conducted from
11 to 31 March, 1965, utilizing only "old" white spruce canes. In
the absence of a cone crop on the Bonanza Creek study area in the
fall of 1964, the only supply of spruce canes available to red
squirrels was that which they had cached in previous years and could
excavate from their middens. Approximately 27,000 of these old white
spruce canes, collected ~inly from eight,middens adjoining the grid
system, were used for the trials. The ability of a squirrel to main-
tain body weight was the criterion used in determining success or
failure of a trial. Three groups of squirrels, with five in each,
were fed in the following manner. Group I acted as a control and
received a constant diet of laboratory chow. Group. II was fed only
old white spruce canes and made an abrupt change in diet from the
laboratory chow to spruce canes. The canes were pre-counted, and
250 were fed daily to each squirrel. At the end of the trials, the
remaining canes were removed from each cage to determine the total
number of canes stripped by each squirrel. Group III made a gradual
. change from the laboratory chow diet to old spruce canes during a
seven-day period. As the original daily ration of 50 g of laboratory
chow was reduced by 7 g each day, the ration of canes was increased
by 35 each day until the squirrel was entirely dependent upon the
250 cone daily allotment. The cages and nest boxes were checked
each day to remove any laboratory chow or spruce canes remaining
from the previous feeding.
-
-
1......
-
-
-
-
1-
The feeding trial with Group IV was conducted from 29 March
to 18 April, 1966, to determine if red squirrels could maintain
body weight on a diet of only white spruce buds. The branch tips
fed during the trials were eut between 2 and 24 March, 1966, from
the crowns of white spruce growing near the study area or in the
vicinity of the University. Tips averaged approximately 5 cm in
length with six buds per tip. The lateral and terminal buds, in-
cluding vegetative and reproductive buds, usually remained intact
13
on the tips. The branch tips were kept frozen in plastic bags until
fed to the squirrels. The five squirrels used for this test were
gradually changed from a laboratory chow diet to one of white spruce
buds during a five-day period. As the initial daily ration of 35 g
of laboratory chow was reduced by 7 g each day, the ration of white
spruce tips was increased by 500 each day until the squirrel was
entirely dependent upon the 2,500 branch tip daily allotment. The
initial daily ration of 35 g of laboratory chow was fed to Group IV
when results from Group III showed that the 50 g daily allotment was
higher than necessary for normal consumption. The cages and nest
boxes were checked daily to remove any laboratory chow or branch
tips remaining from the previous feeding. No attempt was made to
count the number of tips or buds utilized.
RESULTS AND DISCUSSION
Field Study
This section is a synopsis of the chronological sequence of
events relating to red squirrel ecology on the Bonanza Creek study
area during the two consecutive years of cane crop failure (July,
1964, to March, 1966).
In July of 1964, it became evident that no white spruce cane
crop of any consequence would be produced on the study area. Using
the 10-category crop rating system applied by Werner (1964), the
crop would have received a rating of 11 211 (few canes on occasional
trees).
In a year of "nomal" cane production by white spruce in
interior Alaska, the cutting and storage of canes by red squirrels
will usually begin sometime between 1 and 15 August, depending upon
the individual squirrel and the cane crop. The seed usually matures
and begins ta fall about 20 August with at !east 50% falling by
15 September (Robert Gregory, persona! communication). Within this
six week period, the major portion of red squirrel food gathering
and storage normally takes place. While sorne squirrel-cached canes
have undoubtedly shed a portion of their seed, the large majority of
canes are probably eut before they open.
In the absence of a cane crop, the·squirrels during August
remained active primarily on the ground or in the lower branches of
the spruce, very seldom venturing as high as the cane bearing portion
14
~ J lili
-
-
1....
-
......
......
-
......
"'-
15
of the crown. On three occasions squirrels were seen moving about
in the tops of large white spruce. They vmuld run out on a limb,
check the branch tips, then climb to another limb. They were not
seen cutting or eating anything during this time. At no time during
this period were any squirrels observed cutting or caching new
spruce cones •
During the first two weeks of September the squirrels began
digging within their middens and retrieving old, scattered white
spruce cones which had been stored in previous years. Many of these
cones were rotten, mildewed, or insect infested, but were neverthe-
less re-cached in large numbers in active middens. By early October
most middens had large quantities of these cones cached throughout
the tunnel systems, often with many cones easily visible on the
surface. Mushrooms were abundant on the area, and the squirrels ~
were often observed eating and storing them among the lower dead
branches of the white spruce in the vicinity of their middens.
The typical red squirrel social structure of the northern
coniferous forests (Smith 1965) was constantly in evidence during
this period and continued throughout the winter. Established terri-
tories were aggressively defended. Territorial calling was frequent
and usually elicited similar responses from neighboring squirre1s.
In ear1y October, 18 of the 24 marked middens on the study
area were active (Cl, C2, C3, C4, CS, C6, C7, C8, 1-SW, 3-S, 6-NW,
7-SE, 9-SE, 10-S, 12-C, 13-SE, 17-E, 25-W) (Fig. 2). These inc1uded
eight of the niue "cone1ess" middens (those middens adjoining the
16
grid system which had been searched for canes). By mid-November two
of the coneless middens, C3 and C7, were inactive. These had
yielded only 472 and 0 canes respectively when searched, and their
desertion was probably due to lack of a cone supply. It is believed
that both squirrels were young-of-the-year.
Determining whether a midden was active was an easy task
since active middens contained from three to seven fresh bract piles,
often up to 30 cm in height and 45 cm in diameter. New snow soon
covered the bract piles on inactive middens.
From late November until mid-March the squirrels continued to
strip large numbers of cones,and with the exception of midden 1-SW,
there was no change in the active status of the other 15 middens.
About mid-January midden 1-SW became inactive. On 9 February the
squirrel from midden 6-NW, approximately 55 m from 1-SW, was seen
digging for canes on 1-sW. For the next six weeks this squirrel
regularly visited 1-SW and utilized sorne of its cached canes, though
this represented only a small number compared to the number of canes
being stripped on its own midden. In late March a new squirrel
occupied 1-SW and successfully defended the midden as its territory.
While canes were still available to sorne degree on this midden, the
large bract piles characteristic of the other active middens were
not evident. Perhaps the absence of a large cache of canes was the
cause for abandonment in mid-winter.
In early March midden C-1, inactive since November, was again
occupied, but like 1-SW the characteristic large bract piles were
.·· ~~ ;.,j
-
......
1-
-
.._
.......
'-
loo..
.._
......
._
17
not in evidence.
During late February and March, squirrels were occasionally
observed cutting branch tips high in the white spruce and feeding
on the spruce buds. Scattered branch tips were also frequently
observed on the snow under these trees. While buds may have consti-
tuted a part of the late winter diet, the major portion still con-
sisted of spruce seed.
By mid-April spring had arrived. The 1964-65 winter was one
of the coldest on record with prolonged periods of temperature
below -36° C (-32.8° F). Despite the severe winter and absence of
a 1964 cone crop, 15 of the original 18 middens active in October
remained active throughout the winter, and two, C7 and 1-SW, were
active for parts of this period •
Between mid-April and late May the spruce cone supply on all
middens became depleted. This was a graduai process with squirrels
apparently finding small, previously undiscovered, caches of cones
from time to time. By June little sign of freshly stripped canes
could be found anywhere,and for all practical purposes white spruce
seed ceased to constitute any significant part of the diet.
The normally rigid territorial structure which had persisted
throughout the winter seemed to loosen a bit during May and June.
While many squirrels remained on and actively defended their normal
territory, others moved a great deal through the study area, and
many territories changed ownership during June and July. Territorial
calling, while still evident, was not as prevalent as during the
preceding summer nor were these calls answered as regularly by
18
neighboring squirrels.
By July, 1965, it was obvious that for the second consecutive
year there would be no significant white spruce cone crop on the
study area. Using Werner's (1964) rating system, this crop received
a rating of "l" (no cones on any trees).
In early September the normal bustling activity of the red
squirrels was conspicuous by its absence. As in the preceding
autuntn, squirrels were seldom observed high in the spruce. The
squirrel population on the study area had dropped considerably and
with territorial calling at its nadir, a person could have walked
through the area without realizing red squirrels were present.
By early October only six middens were still active of 18
active at the same time the preceding year (Cl, CS, 6-NW, 10-S, 17-E,
25-W). This represented a 67% drop in the number of squirrels.
Since the squirrels had exhausted their cone supply the previous
winter, no cones were available as they began the 1965-66 winter.
While they had relied upon mushrooms to a large extent for food from
July to October, the squirrels did not seem to store any more mush-
rooms thau in the preceding autumn despite the absence of a cone
crop. Tips of v1hite spruce branches were found in sorne quantity on
the snow in late October, indicating that utilization of buds hàd
begun a full four months earlier thau during the winter of 1964~65.
The six middens remained active throughout the winter with
large quantities of spruce buds being eaten by the squirrels as their
primary food. The four squirrels marked in November remained on the
iliÎi
-
-
i.-
1,..,.
1 ......
1-
-
L
-
-
......
'-
-
',
19
same middens during the winter with no change in midden ownership
observed. Home ranges were considerably larger than in the preceding
year as squirrels were lower in numbers and traveled greater distances
in search of food. The lower frequency of territorial calling also
continued until the conclusion of field observations in March.
Food Habits
The great variety of foods utilized by the red squirrel at
lower latitudes has been amply discussed by Klugh (1927), Hatt (1929),
Seton (1929), Hamilton (1939), Layne (1954), Hazard (1960), Smith
(1965) and others. The red squirrel in interior Alaska, however, is
limited in its diet by the homogeneous character of the white and
black spruce forests which it inhabits. In the interior, spruce seed
constitutes the major portion of the year-round diet during a normal
cone crop year (Dice 1921, Murie 1927, Brink 1964). Therefore, in a
homogeneous stand of spruce such as exists on the Bonanza Creek
study area, a cone crop failure leaves no alternative source of
conifer seed.
Table 4 lists the approximate percentages by volume of 29 red
squirrel stomachs collected from August, 1964, to March, 1965.
Spruce seed.--During the summer of 1964, red squirrel middens
on the study area contained large numbers of white spruce cones
cached there in previous years. These cones constituted the major
single portion of the summer diet, supplemented with mushrooms,
fruits, and other available foods. With the failure of the cone
crop, the squirrelsoxiddled their middens with tunnels and exhumed
Table 4. Approximate percentage by volume of red squirrel stomach contents collected near the
Bonanza Creek study area from August, 1964, to March, 1966,f
Spruce Spruce Green Plant Fleshy
Date Seed Fungi Buds Lichen Matter Insects Fruits Bone Mise."'
Aug,, 1964 80% 20% T
Il 70 30 T
Il 20 60 10% 10%
Il 50 50 T T
Il 40 60
Oct, 90 T T T
Il 60 40 T
Dec. 80 20% T
Il 100 T
Jan,, 1965 lOO T
Feb, 20 70 10 T
Il lOO T T
Apr. lOO%
Il ·k·k 90 10% T
Il *";'( 80 10 T T
May 90 10 T
Il 100 T
Il 60 20 10 10
July T 50 50
Il 70 30% T
Il 30 60 10
Aug. 20 80 T
Il 90 10 T
Sept, 100
Oct, T 60 40 T
lt 70 20 10
Dec, 90 10%
Mar., 1966 100 T T
Il 10 90
f Each line represents one stomach. N
,., Wood, bracts, hair, dirt, moss, or gravel. 0
'>'"'<Collected in white spruce stand near the University.
1.
! -
~
1
1
!.-
.....
.....
1
i,_
......
'-"'
.......
21
large numbers of these cones in preparation for the coming winter.
Between 30 to 50% of these cones were rotten, mildewed, or insect
infested. While a small amount of insect damage may occur after
the spruce canes are eut and stored, the major portion occurs while
the canes are still on the tree (Roy Beckwith, U. S. Forest Service,
personal communication). Even though red squirrels may selectively
distinguish between healthy and infested canes when harvesting, in
years of normal cane production squirrels probably eut and store a
certain small percentage of these damaged cones. The_ high percentage
of infested canes exhumed from the middens in 1964 could have been
the result of selection of uninfested cached canes from the middens
by the squirrels for eating during years of normal cane production.
With the cane crop failure, this previous selection could have
resulted in the high proportion of infested canes remaining in the
middens. However, Halvorson (1963) found that even a cane crop
failure in Montana did not induce red squirrels to use insect in-
fested canes of Douglas-fir (Pseudotsuga taxifolia).
In October, 1964, excavation of all the old canes which could
be found in the nine middens adjoining the grid system produced a
total of 24,457 canes (Table 5). Three of the nine middens (C3,
CS, C9) produced less than 500 canes apiece while a fourth (C7)
failed to yield any. The remaining five middens produced an average
of 4,627 canes each, with one (Cl) yielding 8,518. While an attempt
was made to remove all the canes from these nine middens, it became
obvious during the winter that many thousand canes were still
available to the squirrels from areas within the middens not uncovered
22
during the search. All 15 squirrels that remained active during
the winter had cone supplies which lasted until mid-April or May.
Table 5. Old cones excavated from nine middens adjoining the grid
system in October, 1964. '
Midden Designation No. of Cones
Cl 8,518
C2 6,020
C3 472
C4 3,963
cs 1,750
C6 2,866
C7 none
C8 461
C9 407
Total 24,457
Spruce seed remained the primary food in th~ squirrels' diet
throughout the winter and was found in 11 of 12 stomachs analysed
between December and May (Table 4). After the supply of canes on
the middens became depleted in April and May, spruce seed ceased to
constitute any significant portion of the diet. Only three of 11
stomachs analysed between July, 1965 and March, 1966, showed sorne
-
i.-
._
:.....
......
"'-
-
.....
-
trace of spruce seed. Any seed utilized by squirrels after May,
1965, probably resulted from chance discovery of an occasional
cached cane.
23
Fungi.--Fungi, primarily mushrooms, constituted an important
part of the red squirrels' diet throughout the study. While spruce
seed was the major food during the summer and fall of 1964, fresh
mushrooms were eaten constantly before they rotted or were covered
by snow. These mushrooms probably supplied a major portion of the
moisture requirement as no open water supply was available. Five
stomachs collected in August, 1964, contained approximately 20 to
60% mushrooms by volume, indicating more or less alternate ingestion
of mushrooms and spruce seed (Table 4).
Mushrooms were stored singly or in small groups among the
lower branches of many of the white spruce surrounding the active
middens. Within a 7 m2 area on one midden near the study area,
three white spruce contained a total of at least 24 mushrooms stored
at heights of 2 to 9 m. Most of these were well preserved, but a
few were soft or rotten. Buller (1920) mentions a squirrel's store
containing two to three hundred mushrooms in an unused house.
There was no evidence of such large numbers stored in any single
location either within the middens or in trees or stumps. However,
a squirrel could have possibly stored an equivalent total number of
mushrooms in varions smaller caches within its territory. During
the winter partially eaten mushrooms were found on stumps or small
dead branches almost in the center of active middens, but the
squirrels would often leave these untouched for two or three weeks
'··----------------------'
24
time. One large mushroom remained in such a conspicuous position
for over three months on midden 17-E with the squirrel only
occasionally eating a fevJ mouthfuls.
During the winter, squirrels utilized their stores of mushrooms
sparingly. Seldom was a squirrel seen eating a mushroom, and only
one of eight stomachs collected between December and April contained
mushrooms. However, that one stomach contained approximately 70%
mushrooms by volume.
With the depletion of the spruce cone supply in late spring
of 1965, mushrooms became the most important food in the diet.
Species of Clavaria, Hydnum, Boletus, Amanita, several of Russula
and other genera were commonly eaten. Squirrels were twice observed
feeding on the pilei of Amanita muscaria. On both occasions the
mushrooms were collected for identification. Seven of eight stomachs
collected between July and October, 1965, contained mushrooms. These
constituted over 50% by volume in all instances (Table 4).
While mushroom production on the study area in 1965 was
similar to that of the preceding summer, the squirrels did not seem
to store them in greater quantities despite the absence of a cone
supply. Utilization of mushrooms was again spread out over the
entire winter, and they were again found in conspicuous places on
the middens, not being disturbed for two or three weeks at a time.
Although stored mushrooms were generally scarcer in the spring of
1966 than in the preceding year, thawing snow brought dmvn many un-
eaten mushrooms originally stored well up in the spruce. The
squirrels had therefore failed to utilize all the mushrooms available
'' ~ ~
-
.....
~
-
'-
'-
-
.......
.._
-
-
1 -
""""
25
to them during the winter.
In March and April, 1965, portions of che\ved bracket fungi
were found on five middens. Bracket fungi were common in the area,
but no other evidence of such utilization was found. Layne (1954)
mentions that bracket fungi were occasionally eaten by red squirrels
in central New York. It is doubtful that these fungi were an
important source of food for the squirrels.
Smith (1965) lists over 40 species of fungi eaten by red
squirrels in the primarily coniferous forests of southern British
Columbia. Interpolation of his feeding observations of six adult
red squirrels (four females, two males), conducted during various
periods in July and August, showed that the squirrels spent an
average of approximately 23% of their daily feeding time eating
fungi. This occurred when bath old and new conifer canes were
available to the squirrels. Although I do not have similar data
from the study area, 23% is a reasonably close approximation of the
amount of time spent eating fungi, primarily mushrooms, when old
canes were still available in the summer of 1964. However, during
the complete absence of canes in 1965, a figure approaching 60%
would not be unreasonable. In the absence of a plentiful cane
supply, mushrooms seem to constitute a major portion of the red
squirrels' diet from late June until early October in interior
Alaska.
Spruce ~.--The first evidence of white spruce bud utiliza-
tion-was found in late February, 1965. Since all the active middens
still retained fair quantities of old canes, buds were apparently
26
eaten more out of a desire for variety of diet than from necessity.
The squirrels continued to occasionally eat buds during March,
gradually tapering off in April as the buds opened and growth began.
In late October, 1965, eut spruce branch tips were again found
in sorne quantity on the study area. The complete absence of a cone
crop for the second consecutive year had apparently forced squirrels
to utilize spruce buds four months earlier than during the preceding
winter. For the remainder of the 1965-66 winter spruce buds consti-
tuted the major portion of the diet. Large numbers of tips were
regularly found at the base of the white spruce surrounding the
six active middens. One squirrel stomach collected in December and
two in March contained almost 100% spruce buds by volume (Table 4).
Two o.f these squirrels were shot while feeding on buds near the
study area.
Utilization of white spruce buds by squirrels has been
reported by Klugh (1927), Hatt (1929), Hart (1936), Rowe (1952),
and Wagg (1963, 1964). Hosley (1928) and others seem to indicate
that such utilization occurs primarily during periods of deep snow
when the normal food supply is eut off. This is not the case in
interior Alaska. Despite the snow cover, red squirrels maintain an
open tunnel system within their middens and thus have constant
access to their corre supplies during the wiriter. The heavy use of
spruce buds during the winter of 1965-66 can be attributed to the
lack of any suitable supply of a more preferred food. Very few buds
by comparison were utilized the preceding winter when a good supply
of old spruce cones was available.
j
-
-
.._
~~
.._
......
.-
27
Lampio (1952) states that squirrels (Sciurus vulgaris) in
Finland rely upon the seed of spruce as their number one food, but
if spruce and pine seed, an alternate food, are not available, the
main winter food consists of the buds of the spruce. Vartio (1946),
also studying ~· vulgaris in Finland, found that in areas with a
successful cone crop, the diet consisted of 58% conifer seed and 19%
buds of the conifers. Squirrels in areas with a cane crop failure,
however, utilized conifer buds as 49% of their diet. Rowe (1952)
reported that the damage to leaders and upper branches of white
spruce, which occurred in Manitoba during the spring of 1950, was
apparently due to heavy utilization of the buds by red squirrels
following the cone crop failure in 1949. This same situation existed
on the Bonanza Creek study area during the 1965-66 winter. Observa-
tions made in the wild and during the feeding trials in this study
agree with Wagg's (1963) report that red squirrels eut lateral and
terminal twigs of white spruce and ate both the vegetative and
flower buds. No evidence was found showing that the inner bark of
the spruce tips was eaten as reported by sorne authors for other
species of squirrels and conifers. All squirrels observed eating
spruce buds were in the top third of the tree. To what extent such
heavy utilization of spruce buds will affect future cone crops is
not definitely known •
Individual squirrels apparently vary in their methods of eating
spruce buds. MOst squirrels seem to eut the tips of a branch immedi-
ately behind the first whorl of lateral buds and then hold them with
their front feet while eating the buds. An average of 85% of the
28
tips from two, 50-tip samples collected under white spruce trees on
two squirrel territories showed that they had been eut immediately
behind the first lateral whorl of buds. These tips averaged
approximately 3 cm in length. A great variation in bud clipping
existed, both with respect to the proportion of the terminal buds
eaten and in regard to the total number of buds removed from the
branch tips (Table 6). Whether this variation reflects differences
in palatability between trees or merely the feeding habits of the
squirrels is not known.
Table 6. Characteristics of red squirrel-cut white spruce tips
collected from two squirrel territories in February, 1966.
Sample Sample
1 2
Length of tip 2.5 cm 3.3 cm
Tips eut at lateral whorl 92% 78 %
Number of buds in lateral whorl 3.9 4.0
Buds eaten in lateral whorl 82 % 82 %
Terminal buds eaten 90 % 29 %
Average no. buds per tip 7.2 8.4
Total buds on tip eaten 84 % 67 %
Other squirrels were observed feeding upon spruce buds, but
were not seen to eut the twigs in the process. The tips were bent
J
I''i~
;:;
..J
-
-
' -
-
,_
~
._
i.....
\-
;.....,
......
'-
1 --
-.
29
up>vard with the front feet, and the buds pulled off with the te eth.
Thus, some squirrels do not drop the tips which usually connote bud
feeding. Observations also showed that up to two tips discarded by
the squirrels will be caught in the lower branches of the spruce for
every one tip which reaches to ground. The number of tips found
under a tree is therefore at best only an index of the destruction
which has occurred.
Miscellaneous foods.--While spruce canes, spruce buds, and
mushrooms probably supplied over 90% of the red squirrels' diet
during this study, a number of miscellaneous foods were also utilized.
Squirrels were seen at varions times of the year eating lichen
(Parmelia sp.) which grows on the lower dead branches of the white
spruce. They would run their mouth along the top of the branch and
scrape the lichen off, usually taking a portion of the dead wood
with it. Most of.the spruce surrounding a midden showed many of
these scars, seldom more than 1.5 cm in length, on their lower
branches. Eight stomachs collected during the study contained small
amounts of lichen (Table 4)" It is probably eaten incidentally as
the squirrel moves about its territory.
Although two stomachs collected in July and August, 1965,
contained raspberries (Rubus ~), no other evidence indicating
fleshy fruits as a food was found. Highbush cranberry (Viburnum
edule) and rose hips (Rosa acicularis) were the only fleshy fruits
found over a good portion of the study area. Raspberries were
available only to those squirrels near the raad. Klugh (1927) and
Hatt (1929) mention rose hips as a food eaten by red squirrels, and
30
Murie (1927) mentions two bushels of highbush cranberries stored by
a squirrel in southcentral Alaska. Neither species was known to
have been eaten by the squirrels during this study. One rose bush,
within 30 m of an active midden, retained its fruit throughout the
winter of 1965-66 when the squirrel was eating large numbers of
spruce buds.
In December a squirrel near the University was observed tear-
ing open a piece of wasps' nest and eating the larvae and dead
adults. During the study portions of wasps' nests up to 15 cm in
diameter were found on six middens, and two red squirrel stomachs
contained the remains of wasp larvae and adults (Table 4).
Red squirrels seem to have a strong penchant for collecting
the boues of various dead animals. Hatt (1943) found that boues of
various animais were frequently encountered on the middens of the
pine squirrel (Tamiasciurus fremonti) in Colorado. Carlson (1940),
Coventry (1940), Keith (1965), and Smith (1965) all report squirrels
eating bone. The feet of snowshoe hares (Lepus americanus) were
found on 11 middens during the study. Three of these, observed
under a dissecting scope, showed squirrel teeth marks. The feet
are probably picked up by the squirrels and brought to the middens
to be chewed. Traces of bone were found in three stomachs.
The stomach of a squirrel shot while feeding in a quaking aspen
tree in April, 1965, contained 100% aspen buds. Klugh (1927) also
observed red squirrels feeding upon these buds.
The remains of an egg were found on one midden in October of
1964. The size and shape suggest that of a spruce grouse (Canachites
31
canadensis).
Although the following red squirrel foods, listed by other
authors, were known to be available on the study area, no evidence
of their utilization was found~ the green outer bark of quaking
aspen (Seton 1909), bark of paper birch (Lutz 1956), catkins of
paper birch (Hatt 1929), alder cones {Murie 1927), bunchberry
(Cornus canadensis) {Klugh 1929), and the young of various bird
species reported by many authors. Since the normally preferred red
squirrel foods were very scarce during the second year of this
study, a number of these foods were possibly utilized to sorne extent.
Territoriality
The various "degrees" of red squirrel territoriality mentioned
by Seton (1909), Klugh (1929), Hatt (1929, 1943), Gordon (1936),
Clarke (1939), Hamilton (1939), Layne (1954), Kilham (1954), Hazard
(1960), Smith (1963, 1965), and others seem to depend upon the type
of habitat in which the squirrels were studied. They range from
Layne's (1954) conclusion that red squirrels in central New York,
" ••• do not normally establish exclusive property rights over
specifie areas within the home range, and that territorial behavior
is usually restricted to a small area immediately surrounding a
feeding station or particular den site •••• " to the conclusion by
Smith (1965) that territories from one-half to three acres (.2 to
1.2 ha) are defended equally well by either a male or female during
all seasons of the year in southern British Columbia. The situation
in interior Alaska parallels closely the latter. For the purposes
,.
32
of this discussion, normal territorial behavior will be interpreted
as defense of an area; calling, the harvesting and caching of a food
supply, and certain other activities within that area are included
as manifestations of the defense pattern.
Smith (1965) presents an excellent argument for the evolution
of red squirrel territoriality built around food as a resource in
limited supply. Since conifer seed is often the only major food
source present in large quantities in the northern portion of the
squirrel's distributional range, he concludes that the function of
territoriality, " ••• is to allow each individual squirrel the
optimum conditions for harvesting, storing, and defending a food
supply that is naturally available in a usable form only part of the
year, so that it will be available all year." This may explain the
variation in degrees of territoriality as the red squirrel in the
more eastern and southern portions of its range is not as dependent
upon a single source of food which can be easily stored and defended.
If the normally rigid territorial structure exhibited by red
squirrels in conifer forests is dependent upon a defendable stored
food supply, then the great singular dependence upon the white
spruce cane crop by squirrels on the Bonanza Creek study area may
explain their departure from normal territorial behavior following
exhaustion of the old spruce cone supply in the late spring of 1965.
Despite the failure of the 1964 cane crop, the availability of the
previously cached old spruce cones permitted territorial behavior
and structure to remain intact throughout the 1964-65 winter.
Smith (1965) found the frequency of territorial calling by
~~~~~~~~~---~-.. ~~~~
33
Douglas squirrels (Tamiasciurus douglasii) much higher in the spring
of 1963 following a fair cone crop in 1962 than it was during the
spring of 1964 following the corre failure of 1963. He attributed
this drop in the frequency of calling to fewer squirrels calling
because of a population decrease and fewer calls per squirrel because
of the decreased threat of territorial invasion. Halvorson (1963),
studying red squirrels in Montana, found that after a corre crop
failure of the two available conifer species, ponderosa pine (Pinus
ponderosa) and Douglas-fir, the squirrels in late summer and early
fall, " were rather inconspicuous because of subdued calling
rates and general activity." Smith (1963) mentioned that, "In late
August no squirrels were seen or heard in a half hour's walk and
Douglas fir trees were almost bare of new cones." The decrease in
territorial calling which became evident on the study area in May
and June of 1965, and the absence of the normal bustling activity
during late August and September can probably be attributed to the
lack of a defendable food supply, the cone crop failure, and the
decrease in population density as squirrels emmigrated into black
spruce stands, thus lessening the threat of territorial invasion.
Population Density
Occupancy of one midden by a single red squirrel has been
reported by Gordon (1936), Clarke (1939), Kilham (1954), and Smith
(1963, 1965). The same situation exists in interior Alaska. Both
males and females are equally aggressive in defending their terri-
tories, and the sex of a squirrel has no significant bearing upon the
outcome of a territorial dispute. At no time during the study,
34
except during mating and territorial disputes, was more than one
adult observed on a territory.
While it is possible for red squirrels to occupy more than a
single midden, only one such occurrence was observed during this
st~dy. In that instance, the second midden showed only minor evi-
denee of use and was abandoned by the squirrel when another squirrel
defended the secondary site as its territory. It is believed that
the continuous observations made on the study area would have
detected any other similar occurrences.
Assuming one squirrel per active midden, the red squirrel
density on the study area during the winter following the cane crop
failure of 1964 was approximately one squirrel per 1.6 ha (4 acres).
The following winter, after a second cane crop failure, the density
had decreased to one squirrel per 4.8 ha (12 acres). These figures
represent only the winter populations which did not change signifi-
cantly during either winter. A number of smaller, inactive middens
were interspersed among the larger active ones. These middens
probably represent territories normally occupied only during years
of good cane production and when active might raise the density to
one squirrel per 1.2 hti (3 acres). Seton (1909) fe1t that one
squirrel per 3 acres (1.2 ha) indicated abundance. Klugh (1927)
estimated approximate1y two squirre1s per 100 yard square area, or
one per acre (.4 ha), in the spruce woods of New Brunswick, and
Fitzwater (1941) estimated one squirrel per 2.2 acres (.9 ha) on the
spruce flats in the Huntington Forest in New York. Smith (1965)
plotted territories in two different forests and estimated the amount
....!
1
.J
...!
J
-
...t
-
.....
.....
-
._
i.-
1.-.
-
......
~
-
-
1......
-
._
',._
35
of food energy available on territories in southern British Columbia.
He found that stable adult territories in one forest, one squirrel
per 1.27 acres (.5 ha), were significantly smaller (P .005) than
those in the other forest, one squirrel per 2.24 acres (.9 ha).
Since territories in bath forests contained about the same amount of
food, he concluded that territory size appeared to be adjusted to
the quantity of food contained on the territory. The densities
found by Smith would not be expected to fluctuate as greatly between
years as those in interior Alaska since there were at least two
major alternate conifer species in southern British Columbia for the
squirrels to rely upon during a cane crop failure in the third and
principal species. The failure of the white spruce cane crop on the
Bonanza Creek study area, however, left the squirrels with no
alternate supply of conifer seed. The sharp decrease in squirrel
numbers during the intervening summer probably resulted from emmi-
gration into surrounding black spruce stands when the second cone
crop failure became evident and no old vn<ite spruce cones remained
in the middens. There is no reason to believe that the drop in
numbers was caused by predation or disease. Although three goshawks
(Accipiter gentilis) were active on the study area during the summer
of 1964 and accounted for at least one known red squirrel death, no
evidenc2 of predation was found during the summer of 1965. No
squirrels collected near the study area showed any incidence of
disease or abnormal parasitism •
The black spruce stands, 300 rn removed from the study area at
their nearest point, had two normal corre crop years during the study
36
and could have supplied enough food to maintain most of the immi-
grating squirrels. Brink (1964) found that red squirrels presented
with bath white and black spruce canes showed a definite preference
for the former. Although Dice (1921) found red squirrels only
rarely in black spruce, Brink concluded from observations in
interior Alaska that where stands of white and black spruce meet,
red squirrel densities are about equal in bath types. He felt that
competition was for the preferred white spruce habitat with the less
successful squirrels being forced into the marginal black spruce.
My observations agree with those of Brink. Large active middens may
be found in black spruce stands even in years of good cane production
in white spruce. Numerous reports of movements in red squirrel
populations when food shortages occur have been cited in the litera-
ture. The interspersion of white and black spruce stands in
interior Alaska may therefore provide an ideal buffer habitat during
food shortages in the preferred white spruce stands.
Analysis of Old White Spruce Seed
Although approximately 31% of the old white spruce seed sample
consisted of potentially viable (filled) seed (Table 7), only 14 of
1,000 seeds (1.4%) germinated. Brink (1964) found approximately
46% filled seed in a sample taken from squirrel-cached canes of the
current year. MacGillivary (1955) found very reduced rates of
germination for spruce seed after stratification in moist sand at
0° to 2° C for 27 months. Considering that these old canes may
have been cached for at least two or three years in middens, it is
not surprising that viability was so low.
~'
J
-
...... Calorimetrie Determinations
Table 8 contains the calorie values obtained from the three
squirrel foods tested. While only two samples each of white spruce
.......
seed and buds were tested, they do provide some indication of
approximate calorie values. -
Table 8. Calorie values of three red squirrel foods.
Average
Food n cal/g S.D. -
Mushrooms (mixture of pilei
from six species) 4 4,552 36
..... Old White Spruce Seed
(minus seed coat) 2 5,976 54
i White Spruce Buds (undifferen-
tiated meristematic tissue) 2 4,986 46 ' ......
38
The average value for the old white spruce seed (minus seed
coat) was 5,976 cal/g. Brink (1964) reported a value of 6,615 cal/g
for white spruce seed (with coat) taken from squirrel middens near
Fairbanks in 1962. Smith (1965) found values ranging from 6,738 to
7,558 cal/g for 10 species of conifer seed (minus coat), with a
value for Engelmann spruce (Picea engelmanii) seed of 7,107 cal/g
(minus coat). Kendeigh and West (1965) found higher values for
hulled seed in the four instances where samples were tested both
with and without the seed coats. The consistently high calorie
values obtained by Smith for conifer seed minus the coat, and the
higher values for hulled seed found by Kendeigh and West, lead me
to conclude that the value for white spruce seed (minus coat) should
fall between 6,800 and 7,200 cal/g. Since many of the old spruce
cones had been cached for from one to three or four years in middens
before the seed samples were taken, the low calorie values obtained
for the white spruce seed in this study were probably due to energy
loss through respiration. Temperatures recorded at a depth of 50 cm
in a riverbottom stand of white spruce in interior Alaska show that
between early July and early November, the temperatures are above
0° C (32° F), reaching a maximum of 8° C (46.4° F) in August
(Leslie Viereck, personal communication). Temperatures in a spruce
stand on a south-facing slope, such as the Bonanza Creek study area,
may be even higher. The majority of cached cones exhumed during
the study were buried within the upper 50 cm of the middens and
would therefore be subjected to temperatures above freezing long
enough to lose energy by respiration. Over a three-or four-year
;1
lililil
-
-
.__
~
-
1-1
L
L
'
' -
39
period, this loss might account for the low calorie value found for
the old spruce seed,
The calorie value of 4,986 cal/g found for the white spruce
buds shows that they can provide a relatively high-energy food
source in the absence of a cane supply. However, large numbers of
buds would have to be consumed to meet the daily energy requirements.
The four samples of a mixture of pilei from six species of
mushrooms averaged 4,552 cal/g. Smith (1965) found values for fungi
ranging from 3,980 to 5,200 ca1/g with the average for four samples
of a mixture of Ascomycetes and Basidiomycetes surface fungi being
4,320 cal/g. Since mushrooms may contain well over 85% water by
weight, a relatively large quantity of fresh mushrooms must be
eaten to ingest the equivalent of 1 g of dry weight. While these
values range from 2,000 to 3,000 cal/g below those of various species
of conifer seed commonly eaten by red squirrels, mushrooms neverthe-
less provide a relatively dependable alternate food source in
interior Alaska.
Feeding Trials
Group l· Control--The five squirrels (three males, two
females) ":hich remained on a laboratory chow dict and ""teL; as
controls had an average initial weight of 273 g (standard error
18.8 g). The average maximum individual weight change recorded
during the trial was 6.3% of initial weight (s.e. = 2.1%). The max-
imum weight change recorded for any individual was 11.5% (Table 9;
Figs. 3, 7). These fluctuations are within the limits of normal
Table 9. Response of red squirrels to constant diet of 1aboratory chow.
Weight in Grams and
Percent Change from Initial Weight
Squirrel Days into Experiment
No, Se x 0 2 4 6 8 10 12 14 16
2 M 333 333 337 334 328 335 340 344 330
0 0 +1.2 +0,3 -1.5 +0.6 +2.1 + 3.3 -0.9
7 M 287 285 291 286 290 283 283 283 289
0 -0.7 +1.4 -0.4 +1.0 -1.4 -1.4 -1.4 +O. 7
9 M 270 274 282 292 291 295 295 301 278
0 +1.5 +4.4 +8.2 +7.8 +9.3 +9.3 +11.5 +3.0
12 F 258 258 263 261 261 253 269 268 265
0 0 +1.9 +1.2 +1.2 -1.9 +4.3 + 3.9 +2.7
u-F 218 227 233 228 224 226 231 235 240
0 +4.1 +6.9 +4.6 +2.8 +3.7 +6.0 + 7.8 +10.9
Average Percent
Change from 0 +1.0 +3.2 +2.8 +2.2 +2.0 +2.3 + s.o +3.3
Initial Weight
Standard Error
of Average Percent 0 1.0 1.1 1.5 1.3 1.6 1.4 1.8 1.9
Weight Change +--0
( '-"-=_j
.... ...
.... •
...,
0
PERCENT CHANGE FROM INITIAL BODY WEIGHT .... ..., 0 ., \. 1 • ' . ' 1 \ ., \\ \ \ . ·. ' '(,\ \ \ \. . . ,'. \ \ ,, ., /
r.; /\
. ' 1 . 1 • • 1 1 :
1 !
i \
\ ':. . \ \ .
\ \ \ . ti . , ...
1
1
1
1
1
1
\
'. !
. i
: ;
\ ! \ .
\ \,
\ \
\ \ \ ..
\ . . i >.··,. . ; ,. ' . ..,·· .,
. \ l."'· \ ~ • 1 -• Zc ..., ' wo ::;· ~.. \. . ;
1 \
-+ .... Ut
-
42
daily variation and agree closely with Brink (1964).
Group II. Sudden change !Q old ~--The five squirrels
(three males, two females) which made the sudden change in diet from
laboratory chmv to old white spruce cones had an average initial
weight of 262 g (s.e. = 3.0 g). All five squirrels showed an
immediate and sharp loss in weight, reaching an average of 11.5%
below initial weight after the first day and 15.0% after the third
day (Table 10; Figs. 4, 7). The weights then increased steadily,
reaching an average of 5.8% below initial weight on the llth day
when the squirrels were returned to a laboratory chow diet. All
five squirrels demonstrated an ability to increase body weight after
the initial drop. One female, no. 17, had regained her original
body weight while still on the spruce cane diet.
In attempting to make these results comparable with those of
Brink (1964), certain assumptions were made. Since the canes were
stored in the middens for from one to possibly three or four years,
a large number of these old canes were rotten, mildewed, or insect-
infested, with many canes containing no filled seed at all. Since
Brink (1964) encountered few rotten canes and counted all partially
stripped canes as half-stripped, it was necessary in this case to
determine whether a partially stripped cane had been dropped because
the squirrel was receiving no filled seed, or whether it had simply
eaten only part of the available sound seed (see appendix for
detailed discussion).
The male squirrels in this group consumed an average of 196
canes per squirrel per day (s.e. of the mean of individual daily
! ;
~
',-1,""1
.J
.l illiill
-
( r r ( [ r { r r ~ ~ f r r r r
Table 10. Response of red squirrels to sudden change of diet to old white spruce seed. fo11owing
constant diet of laboratory chow.
Weight in Grams and
Percent Change from Initial Weight*
White Spruce Seed Diet Laboratory Chow Diet
Squirrel Days into Experiment
No. Sex 0 1 3 5 7 9 11 13 15 17
6 M 274 250 242 243 248 254 251 267 271 276
0 8.4 11.7 11.3 9.5 7.3 8.4 2.6 1.1 + 0.7
15 M 292 251 245 253 253 262 269 288 285 292
0 14.0 16.1 13.4 13.4 10.3 7.9 1.4 2.4 0
17 F 248 225 213 224 227 241 249 256 268 264
0 9.3 14.1 9.7 8.5 2.9 +0.4 +3.2 +8.1 + 6.4
21 M 238 206 198 202 197 208 219 248 255 269
0 13.4 16.8 15.1 17.2 12.6 8.0 +4.2 +7.1 +13.0
22 F 260 228 217 228 233 239 249 273 277 266
0 12.3 16.5 12.3 10.4 8.1 4.2 +5.0 +6,5 + 2.3
Average Percent
Change from 0 11.5 15.0 12.4 11.8 8.2 5.8 +1.7 +3.6 + 4.5
Initial Weight
Standard Error of
Average Percent 0 1.1 LO 0.9 1.6 1.6 1.6 0.6 1.4 2.8 -1'-
Weight Change w
i<Percent values negative unless otherwise indicated.
1
+12
.... 9
:t:
0
"' ~ 6
>-
Q
0 3
ra
~
~ 0 ....
z
~ 3
0
~ ... 6
"' 0 z
C( 9 :t: v
.... z 12
"' v
~
"' 15 a.
-18
0
WHITE SPRUCE: LABORATORY CHOW
SEED DIET ; DIET
1 Squirrel No. 21
/
1
1
.-.L, .J .. -~ ..... __ ..,. '· "'17
IAr '
1 r..'7 '· ... , ' ,.·'11 '22
c• 1 1 ..• 6
• • • • r-
• . . 1 / ..... -· ~ /.-: 1 1 • ,. ' 1 1 .-. • • • ..... 15
\t
. :, • . ·-•' / . 1 • • .,
\
• • !# ,_. -~ /. ,.: 1 ••• \ .· .,:1/. \ / • . L"
\
l· ~·· . J/ y= .· .. . . -··' ... •'. . . . d ' . .~··-·· ... .,. ., .~ .. . . ·' . . . . _,. ·' / : 1 ' . . . -~·. . . . _,-•-' ,.-' / : ,, ·. .. .,-·' ,. / : ,, ' / ·' ~· / : .._ .••• / ~· .1 • --~-' •/ /.,·-·-· -·' ~' : ,. : '~ &" , . '· _,__ , :
..... ~ .... ---, : ..__, Ë
~
2 4 6 8 10 12
DAYS INTO EXPERIMENT
14 16 18 20
Fig. 4. Response of red squirrels to sudden change of diet to old white spruce seed following
constant diet of laboratory chow,
-1'-
-1'-
-
....
-
....
_,
.....
-
:_
....
-
,__
45
averages = 1.3) while the females averaged 192 cones per day (s.e.
0.7). A "t" test indicated the mean cone consumptions of the two
sexes were different at the 84% level. Brink's (1964) data, however,
indicates a highly significant (P<O.Ol) difference on the basis of
sex with respect to daily consumption of cones. The reason for the
failure to obtain such a high significance level in this study is
not known •
During the experiment the squirrels as a group consumed an
average of 194 old cones per squirrel per day (s.e.= 1.2). Brink
(1964) found that red squirrels consumed an average of 144 new cones
per squirrel per day (s.e. = 4.4) under similar conditions. The
mean for male daily consumption of old cones found in this study,
and the mean for male daily consumption of new cones found by
Brink, showed a significant difference at the 0.02 level. A similar
test made between the female means also showed a significant
difference at the 0.02 level. The very poor condition of a large
portion of the old cones and the lower calorie value of the seed
presumably necessitated this 35% increase in cone consumption to
allow the squirrels to receive sufficient nourishment.
Group ,llb.. Gradual change ~ old ~--The five squirrels
(two males, three females) which made the gradual change from labora-
tory chow to a white spruce cone diet bad an average initial weight
of 259 g (s.e. = 13.1 g). When first presented with both foods, all
five squirrels showed an immediate average weight loss of 5.9% of
initial weight after the first day (Table 11; Figs. 5, 7). The
average weight then increased to 2.7% below initial weight on the
Table 11. Response of red squirrels to gradual change of diet to old white spruce seed following
constant diet of laboratory chow.
Weight in Grams and
Percent Change from Initial Weight>'<
White Spruce Seed Diet Laboratory Chow Diet
Squirrel Days into Experiment
No. Se x 0 1 3 5 7 9 11 13 15 17
3 F 269 263 266 269 265 261 259 263 269 276
0 2.2 1.1 0 1.5 3.0 3.7 2.2 0 +2.6
5 M 261 246 250 256 244 241 229 237 259 254
0 5.8 4.2 1.9 6,5 7,7 12.3 9,2 0;8 2.9
10 M 280 256 267 262 263 250 255 262 271 275
0 8.6 4.6 6.4 6.1 10.7 8.9 6.4 3.2 1.8
11 F 207 196 205 202 201 195 193 198 205 210
0 5.3 1.0 2.4 2.9 5.8 6.7 4.4 1.0 +1.4
18 F 276 255 262 268 251 235 244 259 268 276
0 7.6 5.1 2.9 9.1 14.9 11.6 6.2 2.9 0
Average Percent
Change from 0 5.9 3.2 2.7 5.2 8.4 8.6 5.7 1.6 0.1
Initial Weight
Standard Error of
Average Percent 0 1.1 0,9 0.9 1.3 2.0 1.3 1.2 0.6 0.4 +'
Weight Change "'
*Percent values negative unless otherwise indicated.
1 l_:~
r r r
+12
.... 9 :z:
" "' ~ 6 ,..
Q
0 3 Ill
. _,
< .... 0
z
~ 3
0
Go=
~ 6
"' " z < 9 :::t v .... z 12
"' v
Go=
"' A. 15
-18
r -r~-··· r r ( r
WHITE SPRUCE
SEED DIET
r r r
LABORATORY CHOW
DIET
Squirref
No.
·' 3 __ .,.,,, "" _ _, lt _,. .. ,.,.~~ .. ~--
·-··-·. .... .... ~ = ··r"" . \'· ·--'"'-.. ...... . _, . / :,.......; ....... : ··' ..... ,/
\
,/ ,., ,~ ... --...... ...... ··=" ·"' _..,.., .. 10 ~ 1 ,. • ' ' ~ -··-··"' • i 1 ~ • . ...... .. • , , , • .... ··-..... • ,, ...c·l' • iJ ) / ,, ... ,..· '\ ' .... . )· l' .. .,. )'.,, . . . ·, ' ..... ·" : ,..· ~ ,.,. • ••• , ·'le· --/ ~-., . .,.. ' .... ___ ............. , ... 1 . . . ........ ·/: ' ·. ' .... :/ . • ' • , t'
'
. . . ·. _,.. , /• ' . . ' ,· / ·, '.// . .:.(.. ' ,/ . / '·/·
0 2 4 6 8 10 12 14 16 18
OAYS INTO EXPERIMENT
Fig. S. Response of red squirrels to gradual change of diet to old white spruce seed :Eollowing
constant diet of laboratory chow.
r-~
20
+:-......
(
48
fifth day. From this level the average weight decreased to a point
that was 8.6% below initial weight after 11 days. On the !3th day,
when the squirrels were returned to a laboratory chow diet, their av-
erage weight had increased to a point 5.7% below initial weight.
The graduai change of diet was used to determine if the
initial sharp drop in weight recorded by Brink (1964) could be
eliminated during future trials, thus causing less strain upon the
animais. Figure 7 shows the average percentage weight change for
each group of squirrels participating in the trials. After two days,
the squirrels Brink suddenly changed to a spruce cone diet had an
average weight of 6.5% below initial weight. The average weight
then increased almost steadily to 4.4% below initial weight on the
lOth day when the females were returned to a laboratory chow diet.
The squirrels making the sudden change to old spruce cones in the
present study showed a rouch sharper and more drastic average drop of
11.5% after the first day, reaching a maximum average loss of 15.0%
of initial weight on the third day. The average weight then increased
steadily, reaching 5.6% below initial weight on the llth day when the
squirrels were returned to a laboratory chow diet. The maximum
average weight loss recorded by the latter group was 131% greater
than recorded by Brink. This difference may be attributed to the
very poor condition of the old cones as most other factors were
equal.
The initial weight loss and subsequent increase recorded for
the squirrels making a gradua! change in diet is probably attributable
to the availability of both foods simultaneously and possibly to sorne
j
-
.....
i
i-
-
r
l
L
-
.....
'
1-
-
!
L
.....
i-
L
49
physiological reaction related to digestion. Although more than
enough laboratory chow was available to maintain body weight during
the first four days, all squirrels stripped sorne portion of the
cones each day. The decrease in average body weight after the fifth
day was undoubtedly due to the progressively smaller amounts of
laboratory chow being fed and the greater dependence upon the spruce
cones. The maximum average weight loss of 8.6% of initial weight
occurred on the llth day and was only 57% as large as the maximum
recorded for squirrels in Group II. Since even the new cones fed by
Brink initially caused a sharp decrease in body weights, it seems
that sorne physiological digestive acclimation must occur when a
change in diet is made. A gradual transition in diet allows this
change to occur more slowly, resulting in less strain upon the
animals. Squirrels and other smaller mammals are undoubtedly
capable of maintaining body weight on a number of foods usually
available, but not ordinarily preferred. Feeding trials using such
foods should employ a gradual change in diet lest the physiological
strain of a sudden switch to such unpreferred foods result in
drastic weight loss, possibly death, and spurious conclusions.
Group IV. Gradual change !Q spruce È.!!!!.§_--The five squirrels
(three males, two females) which made the gradual change from
laboratory chow to white spruce buds had an average initial weight
of 240 g (s.e. = 12.6 g). All five squirrels showed a very sharp
drop in body weight beginning the second day, and reached a maximum
average weight loss on the sixth day of 23.6% of initial weight
(Table 12; Figs. 6, 7). Squirrel no. 26, a male, had lost 29.4% of
Table 12. Response of red squirrels to gradual change of diet to white spruce buds fol1owing
constant diet of laboratory chow.
Weight in Grams and
Percent Change from Initial Weight*
White Spruce Bud Diet
Squirrel Days into Experiment
No. Sex 0 1 2 3 4 5 6 7 8 9
26 M 255 253 248 252 222 195 180 Squirre1
0 0.8 2.8 1.2 13.0 23.5 29.4 Died
27 F 207 219 214 191 176 163 158 162 159 Squirre1
0 +5.8 +3.4 7.7 15.0 21.3 23.7 21.7 23.2 Died
28 F 217 221 214 197 188 174 176 183 177 180
0 +1.8 1.4 9.2 13.4 19.8 18.9 15.7 18.4 17.1
29 M 276 277 268 251 243 222 214 210 217 221
0 +0.4 2.9 9.1 12.0 19.6 22.5 24.0 21.4 19.9
30 M 247 244 237 226 207 194 187 188 192 186
0 1.2 4.0 8.5 11.2 21.5 23.5 23.9 22.3 24.7
Average Percent
Change from 0 +1.2 1.5 7.1 13.9 21.1 23.6 21.3 21.3 20.6
Initial Weight
Standard Error of
Average Percent 0 1.0 0.4 1.3 0.6 0.7 1.7 2.0 1.0 2.2
Weight Change \J1
0
*Perc'ent values negative unless otherwise indicated.
i
f r---r r--r ( r r [ r r ( [
Table 12 (cont'd), Response of red squirrels to gradual change of diet to white spruce buds
following constant diet of laboratory chow.
Weight in Grams and
Percent Change from Initial Weight*
White Spruce Bud Diet Laboratory Chow Diet
Squirrel Days into Experiment
No. Sex 10 11 12 13 14 15 16 17 18
26 M De ad
27 F De ad
28 F 188 191 190 194 201 209 218 229 233
13.4 12.0 12.5 10.6 7.4 3.7 +0.5 +5.5 +7 .4
29 M 229 234 230 227 238 249 262 276 274
17.0 15.2 16.7 17.8 13.8 9.8 5.1 0 0.8
30 M 184 188 191 187 198 216 230 243 251
25.5 23.9 22.7 24.3 19.9 12.6 6.9 1.6 +1.6
Average Percent
Change from 18.6 17.0 17.3 17.6 13.7 8.7 3.8 +1.3 +2.8
Initial Weight
Standard Error of
Average Percent 3.6 3,6 2.9 3.2 3.6 2.6 1.9 2.8 2.1
Weight Change
V> ,__.
*Percent values negative unless otherwise indicated,
f
+ 6
WHITE SPRUCE ~ LABORATORY CHOW / Squirrel
BUD DIET ~ DIET ./ No •
l-x
0
141
~
>-
Q
0
IID
.... c
E z
~
0
3
0
3
6
9
::: 12
141
0 z 15 c x u
1-18 z
141 u
: 21
A.
-24
. . . . 1 28 1
:,/\ . ..,;... ..
·.
1
1
1 30
'~ '·' . . , : r-\·~ ·\-;,./ \ 1 ! 1
29
,.,. ·, . i' ,. • • 1 . 1 ~: \ / il
'
• ·, 1 :J • 1 ,,
t· ' ; /1
\
\ 1 .. 1
• :/ // ~"1 ~ .. ,
' . . 1 \~ -·-"' E : 1 "., .,/' : 1 1 t~ 1 : : 1
":\\ • : 1 1 ~~ ~ 1 /·, : .. 'i:\·t· 1 ·, 1 ·· •• E 1 1 \.! • .i. / .... • • 1 ~· 1 ' , ... . .•.
1\ ·" 1 "': 1
't.-,.-' •· : 1
'! .. , ··' : 1
'(1 •• .tl' : 1 ·"" ' . . :
\ lill. o • • ,/A : / '\.. • • -/1' ' ,., • . ~ ..:."'· ~ . ,. ' :, \ • ...,, • ' ./ -i
"26 27 ~ , • .... ., :
0 2 4 6 8 10 12 14 16 18
DAYS INTO EXPERIMENT
Fig. 6. Response of red squirrels to gradual change of diet to white spruce buds following
constant diet of laboratory chow.
20 Vl
N
r· f r
+ 3
"'" 0 J:
~
"" ~ 3 >-Q
0
1111 6 ....
~
"'" 9 z
~ 12 ...
"" ~ 15 z • J:
v 18 ... z
"" ~ 21 "" a.
-24
r r 1 r r r-·-· [ r r
.,··-··-· ............ . ,· ·-·· CONTROL ,· '"'!!o.or"'-·-..,_~.,. ..... ·· -··-··-··-··-··-·· ... ·· ........ ~ , . ., ~
/
/ l 1 , . .,
1 ·' 1 r ' /-" 1
• ~ 1
.' .~ 1 ' ADUAL : • ~ \
GR·--·, BRINK! :J t,• 1 \ -E ... • • '• 7 •• • \ •êHANG '•.._ •' · : j .... 1 \\ . .... .... . . ) . . . . . • • • ..-, •• • • • • •.,. • • M964 .-: "' : 1 "'< \ ••••• ........ · ...... 1 .. ., . . ..... . . ,'lot, ·-"" 1 ' . " ,-·--
0
• \ • 1 . ,
' \ ·' 1 ~ \ SUDDEN ..... 1
\ \ -·-·-1 • • . 1
'-•.._, \ ·'éHANGE •l .... .....'{ il
,. \ , ___ -c
2 4
\ .... :
~ . \ ~
\ /
\ SPRUCE ., \ _,
, 'aÜos ' / y
6 8 10 12
DAYS INTO EXPERIMENT
14
SPRUCE
SEED
OR BUD
DIET
16
LABORATORY
CHOW
DIET
18 20
Fig. 7. Average percent weight change for individual feeding trials.
r
ln
w
54
initial weight and was found dead in its nest box on that day. On
the eighth day squirrel no. 27, a female, was also found dead after
losing 23.2% of its initial body weight. The remaining three
squirrels showed an increase in average weight for the next five
days, reaching an average of 17.6% below initial weight on the 13th
day when they were returned to a laboratory chow diet. Only two of
the squirrels, no. 26 and 28, began eating buds as saon as they were
available. The remaining three squirrels did not begin to eat buds
until after the third day when only 14 g of laboratory chow were
given. After the second day the five squirrels consumed all avail-
able laboratory chow whether or not they ate any buds.
The death of two of the squirrels during the trial does not
necessarily mean they could not have survived on a spruce bud diet
in the wild. Squirrel no. 27 managed to subsist on only buds for
almost three days with no great additional loss of weight before it
succumbed. It is doubtful that any of the squirrels would have
survived a sudden change to the spruce bud diet, as discussed earlier.
Even this gradual change in diet may have caused too great a strain
upon nos. 26 and 27. The fact that three of the squirrels were able
to survive for eight days on only white spruce buds shows that buds
may serve as a food supply in the absence of preferred foods such as
spruce cones or mushrooms.
Cone Consumption
Wagg (1964), using Yeager's (1937) assumption that red
squirrels eat 2 g of Norway spruce (Picea aibes) seed per day, cal-
iJ
-
--------------~----------------~--------------~
-
.._
......
l..
-
:....
.....
.......
l.
.....
55
culated that one bushel (35.2 liters) of white spruce canes could
feed a squirrel for six months from November to April inclusive •
Assuming 240,000 seed per 454 g (Anon. 1948), then 529 seeds equal
1 g. Wagg's estimate of 1,300 seeds consumed daily would equal
approximately 2.5 g of seed per day. Assuming, as discussed earlier,
an average value of 6,300 calories actually available to the squirrel
for each gram of white spruce seed (weight of coat included) eaten,
this would total approximately 16,000 cal/day. This estimate is far
below the amount of spruce seed eaten by red squirrels in interior
Alaska.
Smith (1965) did extensive calculations to determine the
energy requirements of red and Douglas squirrels in the wild in
southern British Columbia. By interpolating his data, values of
approximately 78,000 and 74,000 cal/day are needed by male and
female red squirrels respectively during the six colder months of
the year when stored foods are used and a good deal of time is spent
in the nest. Brink (1963) found that from 24 September to 12 Oeta-
ber, 1963, 17 adult red squirrels (11 males, 6 females) in individual
cages within an outdoor enclosure consumed an average of 35.1 g of
laboratory chow (4,390 cal/g) daily, or 154,000 calories per squirrel
per day. Data collected during the current study showed that from
11 to 21 March, 1965, five squirrels (three males, two females)
consumed an average of 32.7 g of laboratory chow daily, or 143,000
calories per squirrel per day. The caged red squirrels in these
studies consumed considerably more laboratory chow than dictated by
their energy requirements, resulting in substantial weight increases •
56
The 154,000 and 143,000 cal/day values are therefore much higher
than actually needed to maintain a squirrel at constant body weight.
These figures show, however, that the average value of 76,000 cal/
day calculated from Smith's datais not unreasonably high.
Brink's (1964) results from feeding trials conducted with
eight male and eight female squirrels showed an average of 144 new
cones consumed per squirrel per day. His analysis of these cones
showed an average of 86.9 (s.e. = 2.0) seed in the productive zone
of each cone with 46.1% (s.e. = 2.1%) of these being filled, giving
a total of 40 sound seed per cone. Assuming 529 seed per gram by
interpolation (Anon. 1948), approximately 13 cones were needed to
supply 1 g of white spruce seed. Using the estimate of approximately
76,000 cal/day needed by an adult red squirrel and an average of
6,300 calories per gram of spruce seed (weight of coat included)
consumed, a squirrel would have required approximately 12 g of seed
per day. If 13 cones provided 1 g of seed, then 156 cones would
have supplied the daily requirement. This figure is not unreasonably
higher than the total of 144 found by Brink. Thus, it would appear
that red squirrels require approximately 11 to 12 g of sound white
spruce seed per day during the calder months of the year if spruce
cones are the only source of food.
The number of white spruce cones a red squirrel is capable of
cutting and caching in one year is not known. Tripp and Hedlin
(1956) found that 25 white spruce in a heavy cone year had an
average of 8,000 cones per tree and Roe (1952) estimated a total of
11,874 cones on a mature white spruce during a heavy crop year in
;-:c~
• <•;} -
··:q
j
liillii
~
/) ~
-
-
-
......
......
i.-
._
......
......
57
northern Manitoba. Even in years of light to medium cone production,
the majority of trees would be expected to bear an average of at
least 500 to 1,000 canes per tree. Two 30 X 30 m sample plots on
the study area had an average of 333 white spruce per hectare (135/
acre) with a d.b.h. of over 30.5 cm (12 inches). Even with a
squirrel density of one squirrel per 1.2 hectares (3 acres), it
would appear that more than enough canes to maintain a squirrel for
one year are available for the squirrels to eut and store during
years of normal cone production •
Clarke (1939) reports a red squirrel picking and storing at
least 1,000 red pine (Pinus resinosa) canes in one day, each cane
representing a trip up and down the tree, or about one trip every
45 seconds during a 12 hour day. Red squirrels in interior Alaska
eut and drop large numbers of canes at one time from the tops of
white spruce and therefore save considerable amounts of time. Dice
(1921), watching a red squirrel cutting cones near Tanana, Alaska,
stated, "During fifteen minutes he worked continuously, dropping
cones sometimes one per second •••• " During the period of cone
harvesting by red squirrels in interior Alaska, approximately
1 August to 15 September, day length decreases from 18.5 to 12.5
hours respectively. It would seem, therefore, that if enough cones
are available, a red squirrel would have more than ample time to
eut and cache a large supply of canes.
The actual numbers of canes needed to sustain a squirrel
between cone crops will vary with the amount of sound seed per cane.
This value may fluctuate widely. Werner's (1964) data show that in
58
five annual cane samples sound seed constituted 74, 37, 53, 10, and
22% of the total seed in the productive zone of white spruce canes
in interior Alaska. Thus, depending upon the year and the individual
area or spruce stand, the number of canes needed to sustain a
particular squirrel through the winter will vary considerably. The
fact that approximately 35% more old spruce canes were needed ta
maintain the squirrels in this study than was found by Brink for new
canes is a good example.
The number of canes needed to sustain a squirrel between cene
crops, however, is not necessarily related to the number of canes it
will eut and store. Assuming that the number of canes eut and cached
by a squirrel is directly proportional to the size of the cane crop,
up to a physical limit, the two major variables which will determine
the number of canes remaining in a midden, if any, when the next
cene crop matures are the size of the previous cone crop and the
amount of sound seed per cone. In a poor crop year, a high percent-
age of sound seed might be enough to sustain a squirrel while in a
year with a low percentage of sound seed per cane, sheer numbers of
canes eaten might suffice. Therefore, the number of canes eaten
would be expected to vary considerably.
The "average" number of canes a squirrel may strip in a year
is very hard ta estimate. Despite his findings that an average of
144 canes per day were needed ta maintain squirrels in captivity,
Brink (1964) estimated that squirrels in the wild probably utilize
an average of only 40 to 50 white spruce canes per day during the
winter. This was based upon his estimate of 8,000 canes being the
:.J
-
._
59
._
maximum number of cones cached by any of the 15 squirrels whose mid-
dens he excavated in the fall of 1962, a year of normal cone produc--tion. In the present study, a search of middens Cl and C2 yielded
over 8,500 and 6,000 cones respectively in a year of almost total
cone crop failure, yet these squirrels still possessed enough
additional cones to form large bract piles throughout the winter.
-In the absence of a new crop, these cones were only part of a much
larger supply which must have been available in those middens in
-the fall of 1963. Assuming a minimum of only 40 cones per day were
eaten on these middens from October, 1963, until the cone supply
became depleted in April, 1965, approximately 30,000 cones must have -been available in each of those middens in the fall of 1963; Results
of the feeding trials, observations in the wild, and the numbers of
cones excavated from middens after the cone crop failure of 1964,
make it appear that in normal to heavy cone crop years, red squirrels
1 ._ may eut and cache at least 12,000 to 16,000 white spruce cones. It
is not hard to imagine that red squirrels in such years cache many
more cones than are necessary to sustain them until the next cone
crop. Over a period of years with normal cone crops, this excess
would accrue until enough cones remained buried in the middens to ......
successfully maintain the squirrels through the winter following a
light cone crop or one with a very low percentage of sound seed • ._
This was apparently the situation on the Bonanza Creek study area
during the winter following the cone crop failure of 1964.
-
....
-
APPENDIX
Calculation of Numbers of Old Spruce Cones Consumed During Feeding
Trials
In attempting to make the results of these trials comparable
to those obtained by Brink (1964), I had to make certain assumptions
since his feeding trials were conducted with new cones which, on the
whole, were in good condition. Because a high percentage of the old
cones were rotten, mildewed, or insect-infested, and usually
contained little filled seed, a squirrel would often reject such
cones after stripping only three to six whorls of bracts. Since
the purpose of the feeding trials was to determine the total number
of old cones which a squirrel had to handle in order to receive
enough daily nourishment, those cones which had been rejected had to
be counted as part of the total number of cones handled. I had to
decide, therefore, whether a partially stripped cone had been
rejected by the squirrel because it did not yield good seed, or
whether the squirrel had merely stopped eating a cone for sorne
unknown reason as they occasionally do. Observations in the wild
and during the feeding trials resulted in the following assumptions:
1. All cones showing no bracts removed were counted as not
stripped.
2. Cones stripped greater than halfway, but not entirely
stripped, were counted as half-stripped, whether they contained
filled or unfilled seed. Almost all cones greater than half-stripped
60
-
~
~
.....
-
~
......
i....
-
1.-
....
i
L
L
i
1-
.....
'-
61
contained filled seed. Those few cones showing some unfilled seed
were counted as half-stripped under the assumption that they had
yielded enough filled seed from the stripped portions for the
squirrel to have proceeded that far. Brink counted all cones
partially stripped as half-stripped.
3. The remaining partially stripped canes, those stripped
less than halfway, were broken dawn into two groups as follows:
A. Those which contained a majority of filled seed.
B. Those which contained a majority of unfilled seed •
The large majority of cones stripped less than halfway were
obviously in poor condition and filled with unsound seed. Those
canes had therefore been handled by the squirrel and rejected. In
calculating the total number of canes handled, these were counted
as completely stripped. When a squirrel began to strip a rotten
cane, it seldom proceeded to strip it more than halfway before
completely abandoning it and beginning a new cane.
Those canes not obviously insect-infested or in poor condition
were further examined by removing the next few whorls of bracts and
determining whether they contained empty or filled seed. If the
majority of seed was filled, the cane was counted as half-stripped,
otherwise it was recorded as completely stripped.
For example, squirrel no. 6 (Table I) was fed a total of 2,750
old canes. Of this total 547 were unstripped. Ninty-six canes were
greater than half-stripped, but less than completely stripped, and
were counted as half-stripped. A total of 264 canes were less than
Table I. Numbers of old white spruce cones consumed by Group II during feeding trials.
Co nes<!:!
Total Con es Stri1212ed Adjusted Total Daily
Sq, Cones Cones Un-<En tire>~ Sound Empty Total Un-Co nes Average
No, Sex Fed stripped Stripped Seed Seed stripped Handled (11 Days)
6 M 2,750 547 96 17 247 2,147 195
(547)* (48) (8) (0) 603
15 M 2,750 549 133 13 184 2,128 193
(549) (66) (7) (0) 622
17 F 2,750 491 268 49 266 2,100 191
(491) (134) (25) (0) 650
21 M 2,750 463 172 46 303 2,178 198
(463) (86) (23) (0) 572
22 F 2,750 505 231 28 327 2,115 192
(505) (116) (14) (0) 635
Average Values 2,750 511 180 31 264 2,134 194
(511) (90) (15) (0) 616
*Numbers in parenthesis represent those used in calculations (see text),
cr-
N
1 L'"~;~i
/;,
i
L
L
l
L
i .....
....
1 ....
i....
.._
.._
~
L
r-
f
1...1
.._
l...o
-
63
half-stripped. Only 17 of these contained sound seed, and these were
counted as being half-stripped. The remaining 247 canes contained
rotten or unfilled seed. Since they had been handled and rejected
by the squirrel, they were counted as being completely stripped.
This assumed that the squirrel had begun to strip the cone, but
because it did not yield good seed it was then totally discarded.
As cones with rotten or unfilled seed did not constitute more than
30 to 50% of the total canes fed, the great difference in numbers
between the two groups of half-stripped canes shows that there was
a definite tendency to reject canes which contained bad seed.
Table II shows the method of counting cones for squirrel no. 6 •
Table II. Method used in counting canes for squirrel no. 6.
Actual Counted
Number as
Total canes fed 2,750 2,750
Unstripped canes 547 -547
~ntirelY>7z stripped 96 -48
<% stripped~ Filled seed 17 -8
Unfilled seed 247 -0
---
Total cones utili~ed 2,147
64 .......,
Table III. Climatological data recorded during feeding trials
(U. s. Coast and Geodetic Survey, College, Alaska). ~
Temperature Precipitation
(nnn) ""'ï
Max. Min. Ra in, Snow,
Melted Sleet,
Date 0 c o F 0 c o F Snow, Etc. Ha il l'P....,
Mar. 1965
5 5.5 42 -7.5 18
6 5.5 42 -6.0 21 r1 7 6.0 43 -4.5 24 3.6 17.8
8 4.5 40 -3.5 26 T* T
9 8.0 47 0 32 T
10 10.5 51 1.5 35 1.3 r~ ' " 11 5.0 41 1.5 35 ~ J 12 6.5 44 -1.0 30
13 1.5 35 -10.0 14 T
14 -4.5 24 -12.0 10
15 -7.5 18 -20.5 - 5 r~ 16 9.0 16 -22.0 -8 ' -'
17 -5.0 23 -16.5 2 '
18 -2.0 28 -12.5 9
19 2.0 36 -11.5 11 l ,..,
20 3.5 38 -2.0 28 1 21 9.0 48 -1.0 30 .
22 5.5 42 -2.0 28
23 7.5 46 1.5 35 2.3
24 6.0 43 0 32 .....,
25 9.5 49 0.5 33
26 2.0 36 -0.5 31 T T
27 1.5 35 0 32
28 6.5 44 1.0 34 ~ 29 1.5 35 0 32 2.3 12.7
30 2.0 36 0 32 1.3 T
31 4.0 39 -2.5 27 .8
Apr. 1965 1"" 1 4.0 39 -2.0 28
2 4.0 39 -1.0 30
3 5.0 41 2.0 28
4 10.5 51 -2.5 27 T
~ !"
>'<T = Trace
u .
1!
~ <j • 1 ,
..L
i
-Î...-1
LL
l i w
l ' '
'
'
.. i
[
' '--
11 ~~" r ;,, .. L
' 1
~
'
!
L..
L
r
L
L
L
r,
1
L.
1 ....
1
loo...
65
Table III (cont'd). Climatological data recorded during feeding
trials (U. S. Coast and Geodetic Survey, College, Alaska).
Date
Mar. 1966
20
21
22
23
24
25
26
27
28
29
30
31
Apr. 1966
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Max.
0 c
16.0
-0.5
2.5
4.5
4.0
2.5
4.0
6.0
7.0
10.0
10.0
6.0
6.0
8.0
9.5
3.5
1.5
7.0
4.5
4.0
4.0
6.0
-3.5
0
-2.5
-3.5
1.0
7.5
9.0
2.5
0
Temperature
Min.
o F o C
3 -23.0
31 -18.0
37 -6.0
40 -8.0
39 -6.5
37 -5.0
39 -6.5
43 -3.5
45 0
50 -3.5
50 -3.0
43 -7 .o
43 -4.5
47 -4.5
49 -4.5
38 -6.5
35 -6.0
45 -4.5
40 -4.0
39 -6.5
39 -4.0
43 -4.0
26 -8.0
32 -12.0
27 -15.5
26 -14.5
34 -10.0
46 -2.0
48 2.0
37 -2.5
32 -11.0
o F
-10
- 1
21
17
20
23
20
26
32
26
27
19
24
24
24
20
21
24
25
20
25
25
17
10
4
6
14
28
36
27
12
Precipitation
(mm)
Rain, Snow,
Melted Sleet,
Snow, Etc. Rail
1.0 12.7
1.3 20.3
2.3
6.1
66
LITERATURE CITED
American Ornithologists' Union. 1957. Check-list of North American
birds. 5th Ed. Lord Baltimore Press, Inc., Baltimore, Md.
691 p.
Anou. 1948. Woody-plant seed manual. U. S. Dept. Agriculture,
Mise. Pub. 654. 416 p.
Brink, C. H. 1963. Red squirrel in relation to white spruce seed
production and supply, p. 9-10. In Alaska Cooperative Wildlife
Research Unit quarterly progress reoort, vol. 15(1), July-
September, 1963. (University of Alaska, College). Mimeo.
1964. Spruce seed as a food of the squirrels Tamiasciurus
hudsonicus and Glaucomys sabrinus in interior Alaska. M.S. Thesis.
Univ. of Alaska, College. 73 p.
Buller, A. H. R.
mycophagist.
1920. The red squirrel of North America as a
Trans. Brit. Mycological Soc. 6(4):355-362.
Carlson, A. J. 1940. Eating of boues by the pregnant and lactating
gray squirrel. Sei. 91(2372):573.
Clarke, C. H. D. 1939. Sorne notes on hoarding and territorial
behaviour of the red squirrel Sciurus hudsonicus (Erxleben).
Can. Field. Nat. 53(3):.42-43.
Conard, H. S. 1956. How to know the mosses and liverworts. Wm. C.
Brown Company, Dubuque. 226 p.
Coventry, A. F. 1940. The eating of bone by squirrels. Sei.
92(2380):128.
Dice, L. R.
Mammal.
1921. Notes on the mammals of interior Alaska. J.
2(1):20-28.
Fitzwater, W. D., Jr. 1941. The red squirrel: territorialism,
activity, census methods. M.S. Thesis. N.Y. State College of
Forestry, Syracuse. 117 p.
Gordon, K. 1936. Territorial behavior and social dominance among
Sciuridae. J. Mammal. 17(2):171-172.
Hall, E. R. and K. R. Kelson. 1959. The marnmals of North America.
2 vols. The Roland Press Company, N.Y. 1,083 p.
Halvorson, C. H. 1963. Red squirrel ecology in the montane forest
of the northern Rocky Mountains. Annual progress report.
Intermountain Forest and Range Exp. Station, Missoula, Montana.
Mimeo.
'1 i
!
l
~J
n '
"
n
J
l
L
~
w
'
L
L
r Il..;
'f
i1
L
[f:.: u.J
[
r.r · ., ·· ~
\'T'
LL
r·'lf l ti ûJ
r··· . !
! l ll&.it...o
.... i.-
1 ~-~
67
Hamilton, W. J., Jr. 1939. Observations on the life history of the
red squirrel in New York. Amer. Midl. Nat. 22:732-745.
Hart, A. C.
J. For.
1936. Red squirrel damage to pine and spruce plantations.
34(7) :729-730.
Hatt, R. T. 1929. The red squirrel: its life history and habits,
with special reference to the Adirondacks of New York and the
Harvard Forest. Roosevelt Wild Life Annals, 2(1):7-146.
1943. The pine squirrel in Colorado. J. Mammal. 24(3):
311-345.
Hazard, E. B. 1960. A field study of activity among squirrels
(Sciuridae) in southern Michigan. Ph.D. Thesis. Univ. of Mich.
(Libr. Congr. Card. No. Mie. 60-2536) 281 p. Univ. Microfilms,
Ann Arbor, Mich. (Diss. Abstr. 21:379-38~)
Hosley, N. W. 1928. Red squirrel damage to coniferous plantations
and its relation to changing food habits. Ecology 9(1):43-48.
Hult{n, E. 1941-1950. Flora of Alaska and Yukon. Lunds Univ.
Arsskrift N.F. Avd. 2, in 10 pts. and supplement. 1,902 p.
Keith, J~ o. 1965.
ponderosa pine.
The Abert squirrel and its dependence on
Ecology 46(1 & 2):150-163.
Kendeigh, S. C., and G. C. West. 1965. Calorie values of plant
seeds eaten by birds. Ecology 46(4):553-555.
Kilham, L. 1954. Territorial behaviour of red squirrel. J. Mammal.
35:252-253.
Klugh, A. B. 1927. Ecology of the red squirrel. J. Mammal. 8(1):
1-32.
Lampio, T. 1952. Squirrel hunting based on the ecology of the
species. Riistatieteellisia Julkaisuja (Papers on game research)
8. Pohjoismaisten riistantutkijoiden retkeily-ja
neuvottelupaivat (The conference of the game biologists of the
northern countries) p. 44-49.
Layne, J. N. 1954. The biology of the red squirrel, Tamiasciurus
hudsonicus loguax (Bangs), in central New York. Ecol. Monog.
24(3):227-267.
Lutz, H. J •
J. For.
1956. Damage to paper birch by red squirrels in Alaska.
54(1):31-33.
Murie, O. J. 1927. The Alaska red squirrel providing for winter.
J. Mammal. 8(1):37-40 •
68
MacGillivray, H. G. 1955. Germination of spruce and fir seed
following different stratification periods. For. Chron. 31(4):
365.
Mason, E. H., Jr. and W. L. Culberson. 1960. A second checklist
of the lichens of the continental United States and Canada. The
Bryologist 63(3):137-172.
Roe, E. I. 1952. Seed production of a white spruce tree. U.S.
Forest Serv., Lake States For. Expt. Sta., Tech. Note 373. 1 p.
Rowe, J. s. 1952. Squirrel damage to white spruce. Canada Dept.
Nort~Aff. and Natl. Resources, Div. Forest Res., Silvic.
Leaflet No. 61. 2 p.
Seton, E. T. 1909. Life-histories of northern animals. vol. 1.
Charles Scribner's Sons, New York. 673 p.
1929. Lives of game animals. Doubleday, Doran, New York.
vol. 4, pt. 1. 440 p.
Smith, c. C, 1963. Territorial behavior in the genus of squirrels,
Tamiasciurus. M.S. Thesis. Univ. of Wash., Seattle. 68 p.
1965. Interspecific competition in the genus of tree
squirrels Tamiasciurus. Ph.D. Thesis, Univ. of Wash., Seattle.
269 p.
Tripp, H. A. and A. F. Hedlin. 1956. An ecological study and damage
àppraisal of white spruce cone insects. For. Chron, 32(4):400-
410.
Vartio, E. 1946. Oravan talvisesta ravinnosta kapy-ja
k~pykatovuosina. [The winter food of the squirre1 during cone and
cone fai1ure years.] Suomen Riista 1:49-74. In Finnish with
English summ.
Wagg, J. W. B. 1963. Notes on food habits of sma11 mamma1s of the
white spruce forest. For. Chron. 39(4):436-445.
1964. Viability of white spruce seed from squirrel-cut
cones. For. Chron. 40(1):98-110.
Werner, R. A. 1964.
interior Alaska.
White spruce seed loss caused by insects in
Canad. Ent. 96(11}:1462-1464.
Yeager, L. E. 1937. Cone-pi1ing by Michigan red squirre1s. J.
Mamma1. 18(2):191-194.
1
\
J
l
] '
]
n u
·u~.··.·. l ' ,L
n t j