BREEDING CHARACTERISTICS
OF A

CAPTIVE FLOCK OF CANADA GEESE

by

John P. Weigand

AN ABSTRACT OF
A thesis submitted to the College of Science and Arts
Michigan State University of Agriculture and
Applied Science in partial fulfillment of
the requirements for the degree of

MASTER OF SCIENCE

Department of Fisheries and Wildlife

1960

Approved 6/25“ .9 .Q 494

Title: Breeding characteristics of a captive flock of Canada geese.

ABSTRACT

An attempt was made to establish basic information on the nesting
density, productivity, and behavior of a captive Canada goose flock at
the Mason Game Farm near Mason, Michigan. A brief history of the ac-
quisition and management of the flock is given.

Territories, and behavior patterns displayed by the geese in
territorial defense, are described. Discussion of copulatory behavior
of a mated pair and the probability of promiscuous breeding by unmated
geese are given. Several instances of remating after death of a mate
are recorded. Reactions of nesting Canada geese to other waterfowl and
whitetail deer are discussed.

An average nesting density of 2.9 pairs per acre prevailed during
the 1959 nesting season and an average clutch size of 5.44 eggs per nest
was computed for the period 1953 through 1959. Islands, floating plat-
forms and ditchbanks were the most preferred areas for nesting. Incu-
bator-hatched first-clutch eggs had an average hatching success of 34.7
per cent over a seven-year period; incubator-hatched second-clutch eggs
averaged a 37.1 per cent success over a four-year period. In contrast,
second-clutch eggs hatched by the geese had a 60.1 per cent hatching
success over a three-year period.

Desertion of nest sites was believed due primarily to overcrowding
of the geese. It was found that, over the seven-year period, an average
of 65 percent of the geese renested after an average renesting interval

of 17.5 days.

It was suggested that further detailed studies of behavior patterns
and causes of low hatching success in incubators be undertaken. Further-
more, it was recommended to increase the extent of ditchbanks and the
number of floating platforms on the pond, and thereby increase egg and

gosling production of the captive flock.

BREEDING CHARACTERISTICS
OF A

CAPTIVE FLOCK OF CANADA 63858

by

John P. Heigand

A THESIS
Submitted to the College of Agriculture
of Michigan State University in partial
fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Fisheries and Wildlife

1960

ACKNOWLEDGEMENTS

This was a cooperative study between the Game Division, Michigan
Department of Conservation, and the Department of Fisheries and‘wild-
life, Michigan State University.

I wish to acknowledge the efforts of Dr. George A. Petrides,
Professor, Department of Fisheries and Wildlife, Michigan State Univer-
sity, who initiated and directed the project and who offered valuable
editorial suggestions. I also wish to express my gratitude to the
Game Division of the Michigan Department of Conservation for making
the study area available, and especially to Martin J. Pollok, manager
of the Mason Game Farm, for his complete c00peration in the use of
game farm records and facilities.

Gratitude is also expressed to my wife, Judith, for her encourage-

ment and help in typing the manuscript.

INTRODUCTION

Canada geese (Branta cansdensis) are raised at the Mason Game
Farm, Mason, Michigan, for the purpose of starting new resident flocks
in suitable breeding areas in Michigan. The breeding geese are perman-
ent captives; their offspring and geese failing to breed are released
for restocking.

PrOpagation is both from artificially-incubated and naturally-
hatched eggs. There is evidence that at least some behavior character-
istics-related to breeding habits are inherited in this species (for
example, see.Balham, 1954). In view of the widespread distribution of
the captive birds into new breeding areas and the possible introduction
of their behavior patterns into the wild population, it seemed desirable
to document the breeding history and behavior patterns of this density-
tolerant flock. The study was conducted from late March until early

June, 1959.

DESCRIPTION OF THE STUDY AREA

The Mason Game Farm is in gently rolling farming country and includes
an impounded creek. The farm is divided into five fenced units (Map 1),
four of which in 1959 contained a nesting population of 110 pairs of
geese.

At the time of the study, Unit A (Map 2) consisted of two areas:
a wooded parkland of 6.5 acres and about nine acres of cropland. Vege-
tation in the parkland was primarily American elm (9M smericanus)
and willow (S9135 sp.) with some white ash (Prsxinus americana) and
white oak (mrcus _a_lb_a_). There was no understory of shrubs; only
annual and perennial weeds and grasses were present. Bare earth was
exposed on much of the surface. The cropland was usually planted in
corn (Egg lag) for winter forage for the geese.

Unit B (Map 3) contained the upper 5.5 acres of the 8.3 acre pond.
The upper border of the pond was an earthen dike with two water-control
devices. Maximum pond depth was about six feet. Several mud bars and
partly sunken logs, which served as resting and preening areas for the
geese, were present. The major land portion of six acres was planted
with annual brome (Bromus arvensis) and oats (Av—ma; 89.). American
elms and willows were the main trees along the western shore. Rocks
lined the northern two-thirds of this shoreline. Stands of willow,
white pins @333 strobus), red pine (Lin—u! resinosus) and red maple
(Leg rubrum) grew along the eastern shoreline.

Unit C (Map 4), 11.6 acres, contained the lower 2.8 acres of the
pond. A field planted in oats occupied one-half of the eastern portion

of the land; wooded parkland and a beech-sugar maple (Fagus grandifolia-
3.

 

.Map 1 -— General Layout of
the Mason.Gans Farm
near Mason, Michigan

Unit E 1959

 

N

/

C ounty
Hi ghvay

Sgalg
1 inch : 300 feet

 

 

 

 

fond

Unit B

 

 

 

Unit G

 

 

5

Map 2 -- Unit A

 

 

Corn

    

Mature
‘\ Tree
E; Parkland

i?

41>

{I
1:

 

Lasers

a: First clutch nest
CD Second clutch nest
---) Indicates moving to new location
for second clutch
-—--—-— Fence
Buildings

Scale

1 inch : 150 feet

Map 3 -- Unit B

  
   

: Apple
: Orchard
'--—....__
Scale
Woodland

1 inch = 150 feet

 

 

 

   

 

 

Q\ Ir
3
O ,r
8
Annual
Brome
U
@
’@ ..
O
r O ' .
4. “Q/ ’//: I.
B .’ ‘
._ __ I ’ Woods
2 o o
Oats P Fianna
a G ‘l L9 0
I © 0
ngend

First clutch nest
Second clutch nest
B Nest on elevated structure
0 Nest on floating platform on pond
*‘r" Indicates moving to new location
for second cluxch
H—I—l- Fence
“-r- Cover type boundary
Water-control device
Building

Acer saccharum) forest occupied the remaining half. The western land
portion was sparsely covered by American elm, oak and basswood (Eilifi
americana). An island, densely vegetated with rose (32;; sp.), was in
the lower end of the pond.

Unit E (Map 5) contained 6.9 acres. The area east of the creek
was planted with bluegrass (Egg pratensis), annual brome, and rye
(Secale sp.). Vegetation along the creek banks was nettle (Urtica sp.)
and.mullein (Verbascum thapsis). Red pine and white spruce (giggg’glauca)
were the main tree species on the eastern border of the unit. The area
west of the creek contained a meadow of rye, bluegrass and perennial
weeds. The two islands at the southern end of the creek.were densely
covered with young willows. These islands were one to three feet above
the surrounding water level.

Since Unit D was net used for nesting in 1959, description of the
unit has been omitted.

Ten elevated structures had been placed in the units for nesting
use. These were on land and away from the water. They were wooden
frames, four feet by six feet, covered with chicken.wire and mounted
horizontally on legs about a foot above the ground. Each contained
several bales of straw. Sometimes one or two bales were broken up by
the manager and the loose straw spread out on the frame. In other
instances, loose straw was placed on top of several solid bales. In
Units B and C, the lower halves of barrels, and in one a shallow wooden
box, were placed, open end up, on floating platforms anchored in the

pond. These were filled with loose straw.

8

Map 4 - Unit C

 

 

I 9 °
l
; s
U i s O
| 9 O O
l
‘\ Pond °
\ 0 O
\Vooded o
) Parklan
Oats “ e
i e ‘P
| I
t o 5
l
‘ -
l
1
: 0 W odlanc.
\ O
....... s 9'“:G
t "’
M 1 \ 0 ® \‘

A}: e- \ I ‘ ‘\
Beech. \ 1° fl. \
Forest \\ ‘\ o... \

Kg Lowland \
Legend Y o

0 First clutch nest
Second. clutch nest
8 Nest on elevated structure
0 @ Nest on floating platform in pond
- — -—-+ Indicates movins' to new location

for second clutch
arr-PH- Fence

————— Cover type boundary
2mm
1 inch 2' 150 feet s

 

9

Mop 5 -- Unit E

 

   
  

Unit D
(Not used by nesting geese)

 

 

 

 

 

‘3 First clutch nest
Setond clutch nest
B Nest an elevated structure

- ---> Indicates moving to new location

for second clutch
*H Fence Scale
--- Cover type boundary
32:1 Water-control device .1 inch 1' 150 feet

 

METHODS OF STUDY

Trips to all nest sites were made once or twice a week, accompany-
ing the farm manager on his feeding and egg-collecting routes. Nests
were mapped and markedlwith,white wooden stakes, each exhibiting a code
letter. Band numbers of mated individuals were listed in conjunction
with stake locations in recording observations.

Periodically during the nesting and laying season behavior data
were collected by field observations from several lookout points.

Clutch size, hatching success, and renesting data were obtained

from the game farm records which dated back to 1953.

-10.

FLOOR HISTORY AND OPERATIONAL MANAGEMENT

The initial flock comprised three wild pairs of geese which stopped
at the farm during the spring of 1920. They brought off four young
during their first months of captivity and all were retained at the
farm. This provided the nucleus for the present flock.

In 1951, several unrelated geese were acquired from the Seney
National‘Wildlife Refuge in the Upper Peninsula of Michigan. An addi-
tional introduction into the breeding stock occurred in 1957, when a
private individual donated 15 geese. The major portion of the breeding
stock therefore, 110 pairs as it existed in 1959, was composed of birds
which had been hatched and retained at the farm. The breeders acquired
since 1953 have been wing-clipped annually; geese acquired earlier were
pinioned. AAll geese were leg-banded at the time of acquisition.

All of the five farm units, except Unit D, contained nesting geese.
Yearling and two-year-old geese were held there during the nesting
season. Unit A was used in winter to hold the entire flock.

Placing of the paired geese in a given nesting unit (Map l) was
arbitrary on the part of the farm manager except that older established
breeding pairs were returned to the unit in.which they had nested in
previous years. Pairs newly-mated since the last nesting season were
then placed in the units until the farm manager, on the basis of previous
experience with nesting geese, judged that the limit of toleration of
crowding had been reached. Selection of a nest site, nest-building, and
laying of at least one clutch of eggs by a pair indicated that the pair

was successfully established.

11

Pairs having no success in nesting in one unit were moved to
another. This continued until a site was selected and nest-building
began, or until the laying of eggs by other pairs began. In the latter
case, unsuccessful pairs were left in that unit for the remainder of the
laying season, whether they nested or not. It is the policy of the
game farm to hold geese until they are two-years old before introducing
them into a stocking area. Pairs which failed to nest after three
years were shipped to a stocking area along with some of the two-year-
old geese.

Nesting geese were provided.with food once daily. A shallow feed-
ing pan placed near each nest site was filled with a mixture of one-
third each of shelled-corn, wheat and commercial turkey breeder-pellets.
Additional food was obtainable by grazing on oats, rye, annual brome,
and junegrass planted for that purpose. Unmated geese were also pro-
vided.with food.

During the laying of the first clutch, the eggs were collected
daily at the time of feeding. When the first egg appeared in the nest,
it was taken for artificial incubation and replaced by an artificial
egg. This artificial egg, developed by Game Farm Manager Pollok, and
still under study by him, was left in the nest throughout the laying of
the first clutch. When a period of two to three days elapsed without
an egg being laid, the first clutch was judged to have been completed
for that goose and the artificial egg was removed. Pairs which layed
a second clutch or laid a first clutch late in the season (in May), were

allowed to retain their eggs for parental incubation.

12.

TERRITORIAL BEHAVIOR

Territorial pairs are defined by Hanson and Browning (1959) as
"pairs of geese that were observed to be closely associated.with a
small area and that retained their identity with relation to other
geese." An established nesting pair (Balham, 1954) is dominant over
all other geese within its territory. Balham stated two functions of
the territory: (1) to prevent disturbance of the incubating female,
and (2) to provide social stimulation between the members of a pair.

A territory, in the present study, consisted of the area defended
by a pair of geese. It contained the nest and its surroundings, and
adjoining areas for grazing and preening. In the case of pairs nesting
along a shoreline, it also included a portion of water.

On the game farm mated geese began selecting nest sites and estab-
lishing territories in early March. Pollok, (pers. comm.) indicated
that older nesting pairs returned to their old nest sites and territor-
ies while pairs mated since the previous nesting season tended to select
sites in remaining vacant areas for establishment of territories. Upon
selection of a nest site by the female, the male established and de-
fended the territory containing the site.

The hours from 5 a.m. to 7 a.m. and from 4 p.m. to 7 p.m. were
the periods of most active territorial defense during the study. This
was also the period for general bathing, drinking and grazing. Geese
traveling to and from water from sites on the mainland necessarily often
crossed several neighboring territories. Since one pair was seen return-
ing to its territory just after dawn, it was suspected that some of the

less aggressive geese went down to water before daylight.

13.

Intrusion of territories by other geese and by humans brought about
slightly different types of defense. Although the geese were accustomed
to the daily intrusion of the farm manager, who could walk within the
boundaries of a territory with little interference, all but one pair
would attack as he approached the nest site. Encroachment within
territorial boundaries by strange geese would cause immediate alarm
and attack by the defending gander. Balham (1954) also found that
"geese which were continually exposed to the activities of man reacted
to intrusion at the nest with defense and flight attack."

In defending his territory from other geese, the gander first
bobbed his head and neck up and down, occasionally turning the side of
his head toward his aggressors, revealing the white cheek patch. He
then approached the intruders while honking loudly. If the intruding
geese did not then retreat, the defending gander would run at it,
thrusting his neck and head forward and hissing. Retreat by the intruder
occurred in most instances. When an aggressive pair did succeed in
driving a neighboring pair from the latter's nest, the former would
usually increase their territory toward the abandoned nest. They were
not observed to include this second nest within the new boundary. Simi-
lar findings were reported by Collies and Jahn (1959).

In one case an intruder met the attack of the defending gander in
the water. For about a minute they grasped one another's necks with
their bills and pounded each other with their wings. Nearby geese
began to honk loudly. Finally the invader began to retreat toward its
own territory. The defending gander then mounted the back of the intruder,

and holding the defeated.male‘s head under water with his bill, rode

14.

along as the latter swam across the pond. At one point, escape attempts
of the invader became feeble and seemed to cease. Then, with a sudden
forward thrust, it upset its rider and swam back to its mate.

The victor returned to his territory, honking loudly and period-
ically rising up from the water to beat the air with outstretched wings.
Geese adjacent to the Victor's territory continued to honk until the
latter reached its mate at the nest. The winner then preened itself
and activity in the area returned to normal.

The intensity of territorial defense as evidenced by the number
of times pairs defended their territories, was observed to increase
during nest-building and egg-laying, and reached its peak during
incubation. Territorial defense continued until the first clutch had
been completed and all of the eggs were taken by the manager. During
the renesting interval (see beyond) territorial defense slackened notice-
ably. Upon preparation of the nest for the second clutch, defense of the
territory would resume. Since about 70 per cent of the renesting geese
(see Maps 2 - 5) occupied the same nest for the laying of both clutches,
territorial boundaries remained about the same.

Copulatory Behavior. Copulation was observed only once. This was
by a pair which had completed its first clutch and had temporarily left
the nesting site. About 2 p.m. on April 22, the pair was observed in
the pond going through a long series of head and neck bobbings (see
Cellias and Jahn, 1959). Each dipped its head and neck under the water
and then threw the head upward and toward the back. After several
minutes of this behavior, the male swam to the female and mounted her

while holding the back of her neck with his bill. He then reared back

15.

for several seconds with his wings outstretched and it was suspected
by the author that coition occurred at this time. The two birds sep-
arated immediately afterward and began another shorter series of head
bobbings. On completion of this display, they swam to a log together
and preened themselves.

Pollok, who works with the geese throughout the year, advised that
he had Observed similar copulatory behavior, but only on a few occasions
during the daytime. Balham (1954) did not observe copulation on the
nesting grounds in Manitoba and suspected that it occurred during
spring migration. The Michigan date does not completely clarify this
point but indicates that copulation.may occur on the nesting grounds.

Collins and Jahn (1959) reported observing frequent copulations
in captive geese during the day, with the most active time for this
behavior between 7 a.m. and 9 a.m. and between 1 p.m. and 3 p.m. It is
believed, without actual evidence, that at least some copulation occurred
between sunset and the following sunrise in the present study. This
seemed especially likely for mainland pairs (see beyond) which would
have had to cross other territories to reach water. No copulations were
observed to occur on land.

gghggior of Uhmated Geese During;the Nesting Season. A flock of
unmated mature geese was placed with the mated geese in Unit C with the
thought that they might mate and nest, but they did not. Several un-
tended eggs were laid at random throughout the unit which could have
been laid by these geese or by mated birds.

The unmated geese were not observed to disturb nesting pairs in the

vicinity and, for the most part, nesting geese displayed no defense

16.

behavior in their presence. This type of discrimination somewhat res-
embles Hochbaum's (1955) observations that mallard (Anas platyrhynchos)
drakes on their breeding territories give no response to migrant mallard
pairs passing overhead while resident pairs that threaten to alight in
his domain are driven away.

Re-matigg After the Death of a Mate. Several incidences of re-
mating after the death of one member of a pair have been recorded at
the farm (Pollok, unpublished). In the cases recorded it was the male
which died and the female then attempted to mate again within a year or
two. No mating was observed during the same season that the mate died.

Collies and Jahn (1959) recorded a female goose that continued to
incubate her eggs for 10 days after the death of her mate. Then she
paired with another male and deserted her eggs.

In the present study, one female lost her mate shortly after com-
pletion of the first clutch and almost immediately an unmated.male began
to court her. Although the new male continued to escort the female on
her wanderings about the unit, she gave no indication of completely
accepting him and no copulation was observed.

Interspecific Relationships. Five blue geese (£352 caerulescens)
and one male lesser snow goose (Chen hyperborea hyperborea) were present
in unit C and the lower part of Unit B. Four of the blue geese tended
to remain in a single flock and did not seem to disturb the nesting
Canadas. Like the unmated Canada geese, they had access to all parts
of the unit including the pond. Pollok reported (in conversation) that
the remaining blue goose, a male, was mated to a Canada female during
the 1958 nesting season. .Although he successfully defended the nest

site and she layed a clutch of six eggs, the eggs were found to be in-

fertile. , l7.

In 1959, this female mated with a male Canada goose. The pair
was placed in a pen.(Map 3) and the female built her nest against the
fence. Continued attempts were made through the fence by the male blue
goose to drive off the Canada male, but with no success. The blue also
tried unsuccessfully to court the female and.win her from the new mate.
Several times the Canada male stood on the back of the female as he
rebuffed the blue's courtship.

The single male snow goose courted a female Canada in 1959 and
although he kept all other geese from her, copulation was not observed
and no nest was made.

Two pairs of wood ducks (gig spouse) nested in two of the elevated
wood duck houses on the pond. They were observed swimming near goose
nests several times but no attempt was made by the geese to drive them
away.

Two great blue herons Q9595; herodias) which reportedly had nested
near Unit C in previous years, returned in late April. Shortly after
dawn one morning, they flew to the shallow water near the lower end of
the pond. One pair of geese pursued the herons to the edge of the pond
and then returned to their nest. At a later date, three pairs of geese
made an attempt to drive a heron from the pond. While the attempt
eventually succeeded, the herons continued feeding in the vicinity. On
another occasion, a gander rushed a heron from behind but stopped short
when the latter threatened to strike at the goose.

Once several whitetail deer (Qdocoileus virginianus) were seen
going down to water in Unit B. Upon their approach, widespread honking

occurred and geese in the path of the deer moved to one side. The geese

18.

seemed confused while the deer watered. After the deer left the unit,

the geese returned to their nests and became quiet.

NESTS AND NESTING DENSITY

Maximum nesting density is desirable at the game farm for high egg
and gosling production. High densities would also be desirable in the
wild where suitable breeding areas are limited.

From observations in ten years of management at the Seney National
wildlife Refuge in.Michigan, Johnson (1947) concluded that on refuges
nesting geese cannot be crowded if high gosling production is expected.
He also concluded that not more than one nesting pair to each half acre
or acre of nesting territory was desirable. Unit nest densities at the
game farm ranged between 2.2 and 4.9 pairs per acre (Table 1). Average
nesting density on the mainland alone (see beyond) was about 2.8 pairs

per acre.

19.

Table 1.

Unit Nesting Density

 

 

 

 

No. No. No.
Unit Acres Nesting Pairs
Prs. per Acre
A 6.5 14 2.2
B 13.6 38 2.8
C 10.8 24 2.2
B 6.9 34 4.9
Totals 37.8 110 3.0

 

20.

Williams and Marshall's (1937) study revealed that 77 per cent of
the nests of wild geese at the Bear River Refuge, Utah, were within 30
feet of water. Atwater (1958) noted that the majority of his study nests
in Montana.were in open, short-grass areas within 25 yards of water.
Hanson and Browning (1959) located an average of 7.6 territorial pairs
per river-mile based on four years of observations on the Hanford Reser-
vation in.Washington. In the present study of captive birds, 54 per
cent of the 110 nests studied were within 30 feet of water. Within 30
feet of the pond and stream, there was one nest per 166 feet of shore-
line, or about 32 nests per mile of shoreline.

Nests at the game farm'were composed mostly of the loose straw
distributed to the geese, with some down feathers and grass. The
shortest and longest distances between two adjacent nests were 10 feet
and 300 feet. Kossack's (1950) figures on these distances for wild
geese were 40 feet and 90 feet.

Craighead and Craighead (1949) noted that 95 per cent of wild geese
nesting in the Snake River in Idaho were on islands. Geis (1956)
observed island-nesting in over 90 per cent of 479 pairs of geese in
the Flathead Valley of Montana. Naylor (1953) recorded an island of
30 yards by 75 yards, which contained 31 wild goose nests. Sixteen
of the nests were deserted, resulting in a final nesting density of
about 32 pairs per acre. Steel, Dalke, and Bizeau (1957) indicated
that 16 per cent of 380 nests at Gray's Lake in Idaho were on islands
and that 77 per cent were on muskrat houses. Miller and Collins (1953)
also found a distinct preference for islands and muskrat houses as

nesting sites.

21.

At the game farm, in terms of density, shorelines were the most
preferred areas for nesting (Table 2). The wooded shoreline was found
to have the highest nesting density with 6.4 pairs per acre. It incl-
uded the greater portion of the pond shoreline in Units B and C, and

the entire creek shoreline in Unit A.

22.

Table 2.

Nest Site Preferences

 

 

 

No. Nests
Cover Type Nests Acreages per
Acre
Islands and floating
platforms 29 8.3* 3.2*
Wooded shoreline** 21 3.3 6.4
Meadows 15 3.2 4.6
Barren streambanks 9 0.2 4.2
Annual brome fields 9 3.3 2.4
Wooded parklands 21 11.6 1.6
Oat and rye fields 7 10.0 0.7

 

* The acreage given is that of the pond. The nest density using this

acreage is lower than some of those following.

Highest preference is

assigned this category, however, because of the high degree of use of

these sites (see tex

t).

** Within 30 feet of shoreline.

23.

Kossack (1950) observed 10 and 12 wild goose nests per acre in
1945 and 1946 respectively on a 2.3 acre island in Illinois. Hammond
and Mann (1956) found that Canada geese nesting on the Lower Souris
Refuge in North Dakota had increased to 16 nests per acre on islands by
1953. Jensen and Nelsen (1948) determined nesting densities of wild
geese on small islands in an irrigation reservoir to be 54 to 66 nests
per acre.

On the present study site, the six pairs of geese nesting on the
three islands present, totaling 0.021 acres, represented an island
nesting density of 286 pairs per acre. Doubtless, however, some water
areas (here of unknown size) should be included as portions of the
geese's total occupied area, and such additions would be relatively
much more important for small islands and would reduce this apparent
concentration somewhat.

Half of the ten nest platforms described earlier were used during
the 1959 nesting season. Twenty-three of the 29 floating structures
were used during the first laying period and nine during the second.
One pair of geese in Unit C controlled two floating platforms but used
only one during its two nestings.

Three pairs nested in the rocks along the pond shore in Unit B.
These nests were constructed of twigs and down; no straw was used
although it was readily available.

The least preferred cover type was fields containing oats and rye.

24.

CLUTCH SIZE

Records kept during the period 1953 through 1959, showed that the
average size for the first clutch of eggs ranged from 5.0 in 43 nests
in 1955 to 6.0 in 29 nests in 1953 (Table 3) with a seven-year average
of 5.5. Second clutch sizes during the years 1953 through 1956, when
the second clutch was also removed for incubator hatching, ranged from
5.0 to 5.6 with a mean of 5.3. Those second clutches retained and
incubated by the geese, 1957 through 1959, ranged from 5.2 to 5.9 with
a mean clutch size of 5.6 (from data of Table 4). The mean overall
clutch size of 5.44, for the seven-year period, was similar to average

clutches found in wild geese (Table 5).

25.

Table 3. Clutch Size -- First Clutch

 

 

 

 

No. ‘ Clutch Size Frequency Total Ave.
Year Nesting Number Clutch

Prs. l 2 3 4 5 6 7 8 9 Eggs Size
1953 29 - - 2 - 7 ll 7 1 l 173 5.97
1954 38 1 2 - 5 9 11 6 1 2 214* 5.63
1955 43 3 1 4 5 10 15 4 - l 214 4.98
1956 56 2 3 3 7 13 14 11 l 2 297 5.30
1957 96 2 4 6 15 24 28 16 - 1 497 5.18
1958 104 l 2 3 10 19 42 19 7 1 599 5.76
1959 110 1 2 6 13 19 32 33 4 - 626 5.69

 

Totals 476 10 14 24 55 101 153 96 14 8 2620 5.50

 

*Includes one clutch of ten eggs.

Q

26.

Table 4. Clutch Size -- Second Clutch

 

 

 

 

 

No. Clutch Size Frequency Total Ave.
Year Nesting No. Clutch

Prs. '1 2 3 4 5 6 7 8 Eggs Size
1953 23 1 - l 2 8 6 4 l 124 5.39
1954 20 - - 2 - 5 10 3 - 112 5.60
1955 17 l - - 2 7 7 - - 86 5.06
1956 39 3 2 1 3 9 14 6 - 194 4.97
1957** 45 - l 3 7 12 19 3 - 234 5.20
1958** 80 - 2 - 6 18 43 8 2 460* 5.90
1959** 85 2 1 - 8 17 37 19 1 484 5.69
Totals 309 8 6 7 28 76 136 43 4 1706 5.40

 

* Includes single clutch of twelve eggs.
** Eggs during these years were incubated naturally by the parent birds;

in earlier years, all eggs were incubated artificially.

27.

Table 5.

Average Clutch Sizes
of Canada Geese

 

 

 

Status of No. Ave.
Authority Location Population Nests Clutch
Size
Collies and Wisconsin Captive 17 5.2
Jahn (1959)
Dow (1943) California Wild 140 5.09
215 5.10
Geis (1956) Montana Wild 169 5.55
189 5.15
Hanson and Washington Wild 1032 5.3
Browning (1953,'55)
(1959) 5.5
(1954,'56)
Miller and California Wild 201 5.13
Collins
(1953)
Naylor California Wild 360 5.53
(1953)
Steel, Dalke Idaho Wild 361 5.2
and Bizeau
(1957)
Present Michigan Captive 476 5.44
Study

 

28.

HATCHING DATES AND INCUBATION PERIOD

The only hatching dates recorded at the game farm for first clutch
eggs were for 1953 through 1956. The earliest hatching dates varied
from May 7 in 1953 to May 16 in 1956.

Game farm records for these years revealed that the annual average
lengths of incubation periods for 2620 incubator-hatched first-clutch
eggs were 28.6 to 29.2 days. For second-clutch eggs incubator-hatched
prior to the 1957 laying season, the incubation period ranged from 28.1
to 28.6 days. These findings are similar to those of Atwater (1958) who

observed that incubator-hatched eggs required about 28 days of incubation.

29.

HATCHING SUCCESS

Game farm records disclosed hatching successes for first clutches
from 28.0 per cent in 1956 to 47.4 per cent in 1953 (Table 6) or 1.5
to 2.8 goslings hatching per nesting pair (Table 3). That portion of
the second clutch which was incubator-hatched during the 1953-56 period
had 27.3 to 47.5 per cent success in hatching (Table 7) or about 1.4 to
2.5 hatched goslings per nesting pair (Table 4). However, in 1957-59
when nesting geese were allowed to retain their second clutches for
natural incubation, hatching success increased to limits of 55.6 and
63.9 per cent (Table 8), or 2.9 to 3.7 goslings per pair (Table 4).

In contrast, studies involving wild geese show much higher rates
of hatching success. Hanson and Browning (1959) found hatching on the
Hanford Reservation in Washington over a four-year period (1952-56) to
be 92 per cent successful. Miller and Collins' (1953) survey of 201
nests in California indicated an 87 per cent hatching success with an
average hatch per nest of 4.43 goslings. Steel gtflgl. (1957) in Idaho
computed a hatching success of 91 per cent in 1949, 83 per cent in 1950
and 88 per cent in 1951 with an average hatch per nest ranging from

4.2 in 1949 to 4.6 in 1951.

30.

Table 6. Fate of Eggs by Percentages,
First Clutches1

 

 

 

 

 

Year Total Infertilez Dead Goslings Misc.
' Eggs Embryos Hatched Mortality
1953 173 34.7 13.9 47.4 7 Destroyed
by vandals
1954 214 23.8 35.1 38.8 1 Frozen
4 Stolen
1955 214 30.4 28.5 33.6 15 Stolen
1 Broken3
1956 297 22.9 48.8 27.9 1 Broken3
1957 497 23.8 35.9 38.1 4 Broken3
3 Goslings
dead in nest
4 Missing4
1958 599 24.4 32.6 42.6 2 Broken3
1 88331884
1959 626 28.9 36.9 31.6 16 Broken or
taken by
predators
Totals 2620 27.0 33.1 37.1 59
1 Incubator hatched.

2 Appeared infertile when candled; may have contained some embryos

which died within 24 hours after the eggs were laid.
3 Broken by parent geese walking on nest.

4 May have rolled into heavy brush or water, or may have been stolen.

31.

Table 7. Fate of Second-Clutch
Incubator-Hatched Eggs

 

 

 

 

Year Total Infertile Dead Goslings Misc.
Eggs (2) Embryos Hatched Mortality
(7.) (7.)

1953 139 10.1 15.1 47.5 38 Rotten

1954 112 46.4 10.7 42.9

1955 86 37.2 23.3 34.9 3 Broken*
1 Missing**

1956 194 21.7 49.0 27.3 3 Broken*
1 Stolen

Totals

and 531 28.9 24.5 38.2 46
Averages

 

* Caused by parents walking on nest.
**’Mby have rolled into heavy brush or water, or may have been stolen.

32.

Table 8. Fate of Second-Clutch
Naturally-Hatched Eggs

 

 

Dead Goslings
Year Total Infertile Embryos Hatched Misc.
Eggs (1) (1) (1) Mortality

 

1957 234 6.0 23.5 55.6 16 Broken*

1 Stolen

6 Goslings dead
in nest

3 Nests flooded

9

Missing**

1958 457 4.6 17.4 63.9 12 Broken*
17 Taken by
predators
ll Goslings dead
in nest
l Nest flooded
21 Missing**

1959 484*** 5.6 30.4 60.7 16 Broken*

 

 

Totals

and 1175 5.4 23.8 60.1 113
Averages
* Caused by parents walking on nest.

** May have rolled into heavy brush or water, or may have been stolen.
*** Plus twelve goslings found dead in nest.

33.

Salter (1958) in Idaho found an average of 4.6 to 5.2 goslings per pair
in 273 to 345 pairs, respectively. Williams and.Marshall (1938) stated
that in Utah, of 84 nests found, there was an average of 3.9 goslings
per nest.

Infertile eggs ranged from 10.1 to 46.4 per cent (Tables 6 and 7)
among those incubator-hatched, and from 4.6 to 6.0 for naturally-
hatched eggs (Table 8). In wild geese, Steel gt 21.(l957) attributed
only seven per cent loss in successful nests to infertility. Kossack
(1950) found only 4.6 per cent of 325 eggs to be infertile in 1945 and
1.7 per cent of 404 eggs in 1946. Two per cent of the 350 eggs in
Naylor's (1953) study proved to be infertile. Miller and Collins (1953)
found 1.9 per cent of 810 eggs to be infertile. Hanson and Browning's
(1959) study revealed one to two per cent infertile eggs.

In the present study, losses assigned to infertility are not only
markedly higher than those reported by other workers, but are also
higher for artificially-incubated eggs (Table 6) than for those incubated
naturally (Table 8). Apparently either (a) more infertile eggs were
among those collected than the geese normally retained in naturally-
incubated clutches, or (b) some of the losses listed as due to infer-
tility actually were of fertile eggs which died in handling before con-
spicuous embryonic development occurred. Since there is no appreciable
difference in clutch size between eggs collected and those hatched

naturally (Tables 3 and 4), the first factor seems to be unlikely.

34.

NEST DESERTION

Miller and Collins (1953) in northeastern California found that
there was a high rate of nest desertion due to crowding and intra-
specific strife. Naylor's (1953) study included an island (30 yards
by 75 yards) on which crowded conditions resulted in the desertion of
16 of the 31 goose nests.

On the game farm, desertion of nesting sites also was believed to
be due primarily to crowding. In one case, a single pair succeeded in
driving two adjacent pairs from their nests. Since maximum egg pro-
duction was desired at the farm, this aggressive pair and a small
portion of its territory was enclosed with chickendwire fencing. This
action resulted in the less-aggressive pairs returning to their nest
sites and to the continued egg-laying of all three pairs.

Flooding was not a major factor in nest desertion in the present
study. Of 45 second-nests incubated by geese in 1957, only three
were destroyed by flooding. Of 80 second-nests in 1958, only one was
flooded.

Salter (1958) attributed a decline in goose egg production in
Idaho to raccoons (Procyon lotor) and badgers (Taxidea taxus) during
1952-1955. In a later study (1959), he listed inclement weather and
predation as the chief factors in nest desertion. Geis (1956) indi-
cated that 14 per cent of 141 unhatched eggs were destroyed by crows
(Corvus brachyrhynchos) in the Flathead Valley of Montana.

Predation on eggs at the game farm occurred only on a.minor scale.

.A total of 4326 eggs was layed from 1953 to 1959 and only 39 (less than
35.

one per cent) were destroyed or stolen by predators. Principal mammal-
ian predators in the area.were the raccoon, striped skunk (Mephitis
mephitis) and red fox (Vulpes £2115). Although crows were common in
the game farm area, no evidence of egg destruction by them was found
by the farm manager. Predation did result in some observed nest des-
ertion, but in all instances the geese selected a new nest site and
continued egg laying.

Geis (1956) attributed increased nest desertion to disturbance by
humans over extended periods. Hochbaum (1944) found that frequent
disturbance of nesting mallards and canvasbacks (Aythya valisineria)
by humans resulted in several desertions.

Disturbance of geese by humans at the game farm did occur during
daily feeding and egg collecting by the manager or his side. The
geese became accustomed to this intrusion, however, and although
defense behavior was displayed daily, no nest desertion due to dis-

turbance was observed.

36.

RENESTING

Attempts to definitely establish that renesting activities occur
among wild birds are difficult without marked birds (see Barraclough,
1954). However, Balham (1954) noted renesting in a pair of geese at
Delta, Manitoba, whose first clutch of three eggs was frozen and
cracked early in the nesting season. Abandonment of the first nest
site occurred after the third day of incubation. He also recorded four
pairs which did not renest after their nests were lost to flooding in
late May.

On the present study area, it was found (Tables 3 and 4) that
38.5 to 79.3 per cent of the geese, with a seven-year average of 65
per cent, renested after removal of the first clutch. Two pairs in
1959, which had their first egg-laying attempts interrupted by aggressive
neighbors, laid their first clutches while surrounding geese were incu-
bating second clutches.

Balham, in the case of the pair of geese whose eggs were frozen
(see above), noted a "renesting interval" (see Sowls, 1955) of 11 days.
At the Mason Game Farm, in all but a few instances, the first clutch of
eggs was taken by the farm manager one egg at a time as they were layed.
In the few exceptions where the entire clutch was taken at once, incu-
bation had not progressed more than five days and these geese also
renested. Based on the 309 pairs of geese which renested (Table 4),
the renesting interval was found to range between eight and 27 days,

averaging 17.5 days.

.37.

Sowls (1955) found, in 21 female ducks, that there was a statis-
tically greater average number of eggs in the first clutch than in the
second but that "the difference is not great enough to distinguish
first clutches from renests." Atwater (1958), in working with wild
geese in Montana, found it difficult to classify second nests as such
on nest appearances alone.

At the game farm there was an average drop in clutch size of only
0.1 egg (Tables 3 and 4) between first and second clutches. Second
nests were not noted to be different in structure or appearance from

first clutch nests.

38.

CONCLUSIONS AND SUGGESTIONS
FOR MANAGEMENT

Description of the behavior patterns exhibited by this captive
flock of geese is still incomplete. More detailed studies of behavior
in one pen unit would be desirable. Behavior of the wintering flock,
behavior of yearling and two-year-old geese during the breeding season,
relationship of the family within the flock, and mating behavior should
be studied where the birds are conspicuously color-marked. Comparisons
should be made with similar studies of wild geese.

Several behavior patterns might be inherited by the offspring of
this density-tolerant flock and introduced into wild goose populations
by restocking suitable breeding areas with these progeny. The ability
of these captive geese to reproduce under crowded conditions would be a
valuable characteristic to be instilled in wild geese where breeding
areas are lumited in number and size. .A study involving a number of
offspring hatched in an incubator, introduced into a stocking area, and
followed through several nesting seasons would contribute additional
information concerning the inheritance of some of these factors.

Since ditchbanks were among the more preferred sites for nests,
efforts could be made at the Game Farm to increase their extent. Ex-
perimental increases in the number of islands, the most preferred sites
for nesting, would be desirable. Nesting platforms on the pond, which
showed a 79 per cent use for first clutches and 31 per cent for second
clutches, could be constructed at low cost. Such structures could be
placed along the eastern pond shoreline in Units B and C, and along the

northwestern shore in Unit B. 39

Game Farm records revealed that almost two-thirds of all arti-
ficially-incubated eggs failed to hatch, while only one-fourth to one-
third failed to hatch when they were incubated by the geese. Further-
more, one-third of the nesting pairs did not lay a second clutch after
the first was removed. The expected production of day-old goslings for
a sample of ten pairs of adults under the present system of handling thus
would be:

5.5 eggs per pair x 1/3 survival in incubator a 18

plus 5.5 x.2/3 renesting x 2/3 survival under parents24
Goslings : 42

The same number of adults hatching one clutch naturally
would produce 5.5 eggs per pair X 2/3 survival : 36 goslings.

In view of these and other data, several recommendations are made:

1. The costs of raising the additional 17 per cent of goslings
under the present system.ohould.be compared*with the costs of keeping
perhaps a slightly larger colony on a less intensive basis to determine
whether a change in production methods is desirable.

2. Since an increase in hatching success even to 80 per cent in
one naturally-incubated clutch would surpass the present total produc-
tion and since this level of success is commonly exceeded in wild flocks,
experimentation should be undertaken toward gaining increased success
in natural production alone.

3. Since the high loss among newly-collected eggs seems to occur
as a result of handling, a study of the effects of various egg-handling

techniques on hatchability would be desirable.

40.

SUMMARY

A study was conducted from late March through early June, 1959, to
contribute to the knowledge of productivity, nesting density and be-
havior of Canada geese. The geese studied were at the Michigan Depart-
ment of Conservation Game Farm near Mason, Michigan.

A brief history of the acquisition and management of the goose
flock is given. Positive identification of all nesting geese by sex
was possible from leg bands. Nest locations were marked with white
wooden stakes exhibiting coded letter combinations.

After selection of a nest site by the female, the male established
and defended the territory which included the nest site and surrounding
areas used for grazing and preening. Territorial defense increased in
intensity until the beginning of incubation; it then decreased until
the goslings hatched, when defense of the territory ended.

In defending his territory from intruding geese, the gander dis-
played a typical behavior pattern which began with a series of neck and
head bobbings and loud resonant honking; and, was climaxed by the
hissing, running attack and fighting with the intruder.

Pre-copulatory behavior consisted of the female and male swimming
close to each other, dipping their heads and necks under water, and
then throwing them upward and backward. .After several minutes of this
display, the male mounted the female's back. After copulation there
was another, shorter series of head and neck dippings followed by a

period of preening and resting.
41.

Several cases of geese remating after death of a mate were re-
corded.

Five blue geese, a lesser snow goose, and several wood ducks in
the area evoked little concern from the nesting Canada geese. Several
observations were made of attempts to drive great blue herons from the
pond. White-tail deer going down to water at the creek caused great
alarm and confusion of the nesting geese.

Overall average density of 110 nesting pairs was 3.0 pairs per
acre. Islands and floating platforms were the most preferred areas for
nesting while fields containing oats and rye were the least utilized.
Nesting density along the shoreline was computed to be 32 nests per
mile of shoreline. The shortest and longest distances between adjacent
nests were 10 feet and 300 feet respectively.

During the 1959 nesting season half of the ten elevated structures
and 23 of the 29 floating platformswere used for the first clutch.
Several of these structures were also used for the second clutch.

The average clutch size over a seven-year period, for the first
clutch eggs was 5.5. Incubator-hatched second-clutch eggs averaged
5.3 eggs per nest while parental-hatched second-clutch eggs averaged
5.6. Clutch sizes found most frequently were five, six, and seven, and
ranged in size from one egg to twelve eggs. Incubator-hatched eggs
required an incubation period of 28.1 to 29.2 days.

The seven-year average hatching success for first clutches was
37.1 per cent. Incubator-hatched second clutch eggs had a hatching
success of 38.2 per cent while those second clutch eggs hatched by the

geese had a 60.1 per cent hatching success.

42.

Desertion of neat sites was believed to be due primarily to over-
crowding of the geese. Flooding, predation and disturbance by humans
were not believed to be major factors in nest desertion.

It was found that 39.9 to 79.3 per cent, with a seven-year average
of 65 per cent, of the geese renested upon removal of the first egg
clutch. The renesting interval ranged from eight to 27 days, with a
seven-year average period of 17.5 days.

Suggestions for further study of behavior patterns and for manage-

ment of the goose flock are given.

43.

LITERATURE CITED

American Ornithologist's Union. 1957. Check-list of North American
Birds. The Lord Baltimore Press, Inc., Baltimore, Maryland. 689 pp.

Atwater, Melvin G. 1958. Canada Goose Investigations. Job Completion
Report (W-76-R-3). M. S. Thesis, Montana State University, Missoula.

Balham, Ronald W. 1954. The behavior of the Canada goose (Branta
canadensig) in Manitoba. Ph.D. Thesis, University of Missouri,
Columbia. 229 pp.

Barraclough, Mary. 1954. The biology of Canada geese (Branta canadensis
moffitti) in the Flathead Valley of Montana. M. S. Thesis, Mon-
tana State university, Missoula. 91 pp.

Burt, William H. 1957. Mammals of the Great Lakes Region. The Univer-
sity of Michigan Press, Ann Arbor, Michigan. 246 pp.

Collies, Nicholas E. and Laurence R. Jahn. 1959. Social behavior and
breeding success of Canada geese (Branta canadensis) confined
under semi-natural conditions. Auk, 76 (4): 478-509.

Craighead, F. C. Jr. and J. J. Craighead. 1949. Nesting Canada geese
on the Upper Snake River. Jour. Wildl. Mgt., l3 (1): 51-64.

Dow, Jay S. 1943. A study of nesting Canada geese in Honey Lake
Valley, California. Calif. Fish and Game, 29 (1): 3-18.

Fernald, Merritt L. 1950. Gray's manual of botany. American Book
Company, New York. Eighth Ed. 1632 pp.

Geis, Mary Barraclough. 1956. Productivity of Canada geese in the
Flathead Valley, Montana. Jour. Wildl. Mgt., 20(4): 409-419.

Hummond, M. C. and G. E. Mann. 1956. Waterfowl nesting islands. Jour.
Wildl. Mgt., 20 (4): 348-352.

Hanson, W. C. and R. L. Browning. 1959. Nesting studies of Canada
geese on the Hanford Reservation, 1953-56. Jour. Wildl. Mgt.,
23 (2): 129-137.

Hochbaum, H. Albert. 1944. The canvasback on a prairie marsh. The
Wildl. Mgt. Inst., Washington, D. C. Second Ed., 1959. 207 pp.

44

Hochbaum, H. Albert. 1955. Travels and traditions of waterfowl. The
university of Minnesota Press, Minneapolis, Minn. Second Printing,
1956. 301 pp.

Jensen, C. H. and N. Nelsen. 1948. Goose breeding grounds - Utah
and southeastern Idaho. Waterfowl populations and breeding con-
ditions - summer (1948) - with notes on woodcock studies. Spec.
Sci. Rpt., 60. U. S. Fish and.Wi1dl. Serv., and Dominion Wildl.
Serv. pp. 152-55.

Johnson, C. S. 1947. Canada goose management, Seney National Wildlife
Refuge. Jour. Wildl. Mgt., 11: 21-24.

Kossack, C. W. 1950. Breeding habits of Canada geese under refuge
conditions. Amer. Midl. Nat., 43: 627-649.

Miller, A. W. and B. D. Collins. 1953. A nesting study of Canada
geese on Tule Lake and Lower Klamath Wildlife Refuges, Siskiyou
County, California. Calif. Fish and Game, 39: 385-396.

Naylor, A. E. 1953. Production of the Canada geese on Honey Lake
Refuge, Lassen County, California. Calif. Fish and Game, 39,
(1): 83-94.

Otis, Charles H. 1931. Michigan trees. University of Michigan Press,
Ann Arbor, Michigan. Fourteenth Printing, 1956. 362 pp.

Salter, Robert L. 1958. Canada goose nesting studies in Idaho. Ida.
Wildl. Rev., May-June, 1958.

1959. Goose nesting studies. Ida. Wildl. Rev.,
Sept.-Oct., 1959.

 

Sowls, Lyle K. 1955. Prairie ducks, a study of their behavior ecology
and management. The Wildl. Mgt. Inst., Washington, D. C. First
Ed. 193 pp.

Steel, Paul E., Paul D. Dalke, and Elwood G. Bizeau. 1957. Canada
goose production at Gray's Lake, Idaho, 1949-51. Jour. Wildl.
Mgt., 21 (1): 38-41.

Williams, Cecil S. and William H. Marshall. 1937. Goose nesting
studies on Bear River Migratory Waterfowl Refuge. Jour. Wildl.
Mgt., 1 (3-4): 77-86o

1938. Survival of Canada geese goslings, Bear River
Refuge, Utah, 1937. Jour. Wildl. Mgt., 2 (1): 17-19.

45.

. :d‘fl

..
)3
v

'I
xi:

 

s
. o’akxe

. LY"... n.

sifi .
..

"I7'7141111141111111T