.
v
..
n s. . _
. ., . _ _
. ...... _
.. . .
. a 1 _
III _ . ._._.
w . o a .
u K . _ . I
§ _ O . . c
.. C . .
v m . .. .
. o
.. ..
w . . ... . . , Inflow.“
. . . 'l.. t: ...! I...
.O N . _ . . ..dqfunIfl 7'. \...._.I..o..( ... s.
. . , r. . .J
.1 0 . . . . I. h4/’gf.01’.”¢cr/'91W.IIWJ .r.“
. _ . . u , . .
.0 s . _ . . o Dfllur/’.. -. v1" crvrlf'. I“v.’.'£u..41¢ 0 Co. . c. ,‘
. .. _ .. a 7.. ... I.I. . 1...... nor...-
. , .. . I .. . .H../¢.r1...J.l.I.f¢/:Do.aao.o. walusuA...wn .3.-
. I .... . .. ... .v . . . .,r.Pv ,‘rr .1W54?U.:. Ito.
. _ . a . .. . ..J.....v.w.o.. 3d. ’3 v. . or” oatI
. . . _ O . . . u . .
M. S . I. .. ,.?.:.._.?.a. .
. . _ . . . ...... .v../L_‘ol il. l’...’
. o . ... . _ fa. .r.-a.u0(s/~.c.v tor” fl
. .. . a ., . ....-o.’;|-.4 .0 .9
_ v . . . . . . . . . ”I. ..a o 1' p .frl. I..
f .. . _. I. . 4 .rr 0.. JI¢"‘O
.. o o t I... . ...-.va I 1..
c o . . . . .; ... .. ....xf.
. ... .-.—o.....u.~... If..-
. ... . \.JI.I..ro. - .ttdu-;
o . . u. a.’. .a....A..o..r...‘t..
, . . . .o- l .u:o.v;l.4.¢.Id-.
e . . o . . 5 .0 I It .uAII‘.
. . _ r . ._ . . . ..(u. ..I..oo.!...v
n v. e.. . . . ,. . . . . ..d.r_rol. .. I...
I . ,
r U . z . .
. . ..
R g 1 . . .. . . . I.
. . .. at
0 Tl . l - .
D ;
V. Eu 1 . . . . ..
I .~ I ‘41..
an .hlbl m . . . . ..,. 2?... 4... My f.:...,.um.......
. . . . . .v .u c‘ {I In .. I.
. .. . .. . ... .... HIM. ..IIFW201H1 JIM
...... B I _ . .
. . o .. I an. fv
r N . . I
B - . I . '. .. .
.A .
o I . . c . I. . a no... ll..o} r.'.'I'.«oov.
poll U . . . - I . 34::
., . . a-.. .a ply.....J..Il,Iv.fi....o....o.\‘
. . . . . . o 4. 9 .
.. . I. . ,. I.. ......
G 'I A ...... . r
N ....I H D .
.. . r I..4..c...
. I K.. ....vo... .avoln.
I . . .. . I. Ior.’
. 1 .
. o of
, _ . . 4.7.
. . . , .
NI . .. ,,
A . . . L. .. ..
. .I. 200....
. o . . _
. I. ...on. .2...
.z .. ... ...p. ,. a..::t
. r _ r
, . .. . .. ...r
. W . . f. . . o c I
v... .1: . c. I! . .
~ R I . I 0. fl .0 . I . .
. o. . x . . .r
. . v . .
. O / I I a . r.
, I . 1 . . . . .. I: .
. ‘ . . .. .
... . I it .I
o . .. Z. I .... I .
- I ., o ..
. . . . . at . ‘
a . .
, . . .. .. r .q .... .
v . I . . ... . orot.
N I . ‘V- J
.. ..I.... u .,. ..
. _ . . .. . ..
z .. 5...». .. u
. O ..‘l- . \ I < .r r. n-
. . I 5. l . f . f A. t...
. .. I. . .
. .v . . _ ... f
H . I. . . . II . I
4 .f . I . yo. I
D I . . o .4 III. . ad a I
a . . .. . ... _v.
9. . _ . o .. . . .. c
. ., . . u . . 1.. I
. . a .
Ito . u I .... . a.
o. .. .. 1
fl . .o A. .9 u ..1. .lo . ..; I . _
. ’I ....o _ I
” . I. 9-. .
. . . I
g o .I . .
fl . a. 0.. 1. .... 4 o. IA
. u .. .1. .
‘0 I. '10. .v. I .40
o .. .1 . I; .
o . . .A. ..t . 0...:
0 o 14 I A ‘l. . o . O
I I ... ID .. . . 5'21 v
.m . I... . ......In..;...l1.
r. i. o. . V 0'. to... . c: a.... _'O.ld. ......34'
u . 0 .C l . I. l I I
l

        

  
 

.. IO .‘
. .. . I u ... . mm... ..... ..I....gp.......7.......0.xacvt.na«:3$ 5.4.5,.v13r. otflfiflmofiirynwrw... .
o. . .. . v‘ s. .. . . a n”. v ”D. 1.. 5.9 I o
. . V I .. . p. o .1 _¢.co. ..... _ _. .. ... . ...f. :1. £11.. . l . .. ".... . -.. ...... ...» .... ..‘Ioss.~..n . .. . . . . ..I-Ioto..rJ . .. . . .10\.o.\o...c\.-QVN~I.¢0 3!.“ . i’Ivowrxfi‘o‘h.“
. . 1-. . W a. o . . u .A I .... .u‘ :.o\v ‘s. 30% It‘w‘awm Lug” h-I— \
. _ . . .I . .. . ,._ ..... ..~ ... .. ...._.....o...o ...}... «......In. .0 ...Oo..w.\con HIV—‘70 .IVITE I..- J .le.
nub: .. . _ ...I o .8. .. I I . I1... ...I. .v ~.I Kola. . .. .AL..\.P.LJ....3.l."..<..'\...Ax... .....v..5\ to . r 8 o
..00. I
o . I

 

ABSTRACT
COLONIAL NESTING BEHAVIOR OF COMMON GBACKLES
By
Paul Bernard Hamel

Colonial nesting is a common feature of the breeding
biology of blackbirds, Icteridae. While the Common Crackle
(3mgégglgs guiscgla) often nests in groups it may be nest-
site limited rather than actively colonial. This investigation
addresses the questions: Is this bird nest-site limited? Are
any of its display behaviors contagious? Are any of its nes—
ting activities synchronized? The questions were examined
in three locations in central Michigan. One of these, the
Michigan State University campus, was the site of an intense
study of grackle biology in the 1950's and census data from
the two studies are compared. In addition, I undertook a
study to determine techniques for sexing nestlings based upon
external measurements.

Colony groups were observed during the pair formation
and nest-building periods. Data were taken concerning the
relationship of group size and song-display behavior, nest-
site choice, and date of beginning nest building. Seventy
nestlings were collected at known ages and wing, tail, tarsus,
culmen, innermost primary, and weight measurements were
taken. Sex was determined by internal examination of the

gonads.

Paul Bernard Hamel

These studies reveal that the Common Crackle is actively
colonial. Song-display behavior in groups of grackles is
contagious. It stimulates a synchronization of beginning
nest building by the females. Nest-site choice is also
affected, the modal site within colonies being the site first
chosen, independent of relative availability of the sites.
The census data showed that nest sites were not in limiting
supply at the time of either the 1952 or 1971 census, but
that the population has increased six-fold in the interim.
Partial reasons for the increase are an increased availability
of nest sites with new ornamental plantings of conifers and
a decrease in competition for sites with Robins (Tgrdgs
migratorigs).

Tentative criteria for sexing based upon Tarsus and
Culmen measurements of 11- and 12-day old young are presented.
The possible significance of a previously unrecorded series

of three color phases is discussed.

COLONIAL NESTING BEHAVIOR OF COMMON GRACKLES

BY

Paul Bernard Hamel

A THESIS

Submitted to
Michigan State university
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Zoology

1971

ACKNOWLEDGMENTS

This study was supported in part by a grant from the
Josselyn Van Tyne Memorial Fund of the American Ornithologists'
Union. I am deeply indebted to G. J. Wallace for criticism
and advice throughout the study, and to the other members of
my Guidance Committee, R. H. Baker, J. H. Beaman, and M. F.
Denton. B. G. Murray, Jr., and G. Divoky were helpful in
formulating the study at its inception. J. Hamel, A. Petkus,
F. Heineman, P. Dalby, S. Blaisdell, and J. Euker gave
material help in the execution of the work. The presentation
of the data has been greatly aided by discussions with B. L.
Monroe, Jr., R. Stuart, F. Zak, and G. Starks. C. E. Cress
and R. Hodgson gave valuable statistical advice. The kind
assistance of S. Keesler for permission to trap and band on
the University property and that of G. Zorb for permission
to conduct studies on the Rose Lake Wildlife Research Center
are acknowledged. My wife, Mary Kay, has not only borne the
brunt of ornithological widowhood, but also critically

evaluated and typed the manuscript.

ii

List of Tables .
List of Figures
Introduction . .

Study Areas . .

Methods and Materials
Results and Discussion
Changes in the East Lansing Population .

Social Stimulation of Breeding and Nest-Site

Selection

Sex of Nestlings

Summary

Literature Cited .

TABLE OF CONTENTS

iii

Table
Table

Table

Table

Table

Table

Table

Table

Time Table of Common Grackle Reproduction

Common Crackle Breeding Population
in East Lansing, Michigan

Frequency of Group Size Xé Stage

LIST OF TABLES

of Nesting Cycle .

Colony Development in Common Grackles
in Central Michigan, 1971

Nest- Site Choice in Central Michigan

Grackles .

Placement of First Nest in Relation to

Centroid of Colony .

Criteria for Sexing Nestling Grackles
11 and 12 Days Old .

Color Phase Distribution in Nestling

Grackles .

0

iv

0

Page
. 7

.14

.21

.25

.28

.30

~37

.hO

Figure

Figure

Figure

Figure

LIST OF FIGURES

.Page

Locations of Nesting Colonies of
Common Grackles, East Lansing, Michigan . . . 5

Regression of the Square Root of Number of
Vocalization Pairs on Group Size . . . . . . .2h

Internest Distances in Grackle Colonies . . .32

Tarsus and Culmen Measurements and Color
Phase of Nestling Grackles . . . . . . . . . .36

INTRODUCTION

Coloniality holds a prominent place as a breeding
regime among the New World blackbirds, Icteridae. The prac-
tice is well documented for the oropendolas (Fsargcgligs)
and caciques [Cacicgs (Skutch, 195“; Chapman, 1928; Schafer,
1957; Dunham, 1971; Meyerreicks, 1958)], southern grackles
[suiggalgg (Mcllhenny, 1937; Tutor, 1962; Selander, 1965;
Selander and Ciller, 1961)], Brewer's Blackbird [Egpnaggg
czanocephalgg (Williams, 1952; Horn, 1968, 1970)], marsh
blackbirds [Agelaigs (e. g. Orians, 1961a, 1961b)], and some-
times even certain orioles [Ictgrgs (Thomas, 19h6; Dennis,
l9h8)]. The Common Crackle (Qgiscalgs guiscgla) has repeated-
ly been called colonial (Gibbs, 1903; Peterson and YOung,
1950). However, nest-site limitation is a viable alternative
explanation for a clumped pattern of nesting in this species.
Shrubby marshes and swamps, the grackle's original habitat,
have been drastically reduced, forcing adaptation to other
nesting sites; clumped plantings of ornamental conifers have
thus become favorite alternative sites (Gibbs, 1903; Bent,
1958).

This study is an attempt to ascertain whether breeding
aggregations of Common Grackles are colonies in the sense

developed by Horn (1970) or are functions of a patchy

2

distribution of nest sites. Factors that must be accounted
for in the evaluation of colonial status are the general
social structure of groups of birds which involves possibly
contagious behavior, and nest dispersion in time as well as
in space. Consideration must be given to the background
noise variables, age of nesting birds in the group, and the
effect of nest-site availability on nest dispersion.

A census was taken of the grackle population in a 3.5
square mile area around the Michigan State University campus
to identify changes that may have occurred in the population
since the 1952 census by Eyer (l95h). ‘A study was also made
of nestlings to determine criteria for sexing birds of known
age, to complement and evaluate an earlier scheme by Willson

6t 31. (1971) °
STUDY AREAS

This study was carried out in the months of March to
July in the years 1968, 1970, and 1971 in three locations in
Ingham, Kent, and Shiawassee counties, Michigan.

In 1968, I censused the breeding birds in a 1.3 acre
buttonbush swamp called Manhattan Marsh. (T.6N., R.llW.,
l/h mile from Reed's Lake, #2050'N, 85°30'W, East Grand Rapids,
Kent County, Michigan) The swamp was bordered by landfill
areas to the north and east, an old field to the south, and
a road to the west. Water depth in the swamp varied but was
never more than 1.5 meters. Two pools in the northwest and

southwest sections were deeper than the eastern edge and an

3

east-west strip between the pools. The dominant plants are
buttonbush (Cephalagtggs gggiggntglis) which occur chiefly
in areas where the water is 0.5 meters or more deep, cattails
(Typhg latifolia) in areas where the water depth is 0.5 meters
or less, and several large black willows (Sail; nigra) which
are located along the southern and northwest edges of the
swamp. All 30 active grackle nests were placed in buttonbushes.
In 1970, a breeding bird census was conducted in a wet
second-growth forest at the Rose Lake Wildlife Research Sta—
tion. (T.5N., R.1E., n2°ho'N, 8u°2o'w, Section 21, Shiawassee
County, Michigan) The dominant plants in the 20-acre area
are quaking aspen (Popglgs treggloiggs) which covers approxi-
mately the southwest quarter with a very dense 15 to 30 foot
tall forest, willows (Selig spp.) which form a dense, shrubby
cover in the southeast and northeast sections and alonthhe
banks of a stream on the northern edge of the area, and
spruce (Eigga spp.) and red pine (Pings resigosa) planted in
rows in the center. The area is bordered to the east and
north by roads, to the south by a controlled wetland, and
to the west by an old field. The 30 grackle nests in this
area were all located in conifers. A similar breeding census
was conducted in 1971 in the same Rose Lake area but more em-
phasis was placed on a detailed study of the grackles and less
on other species. The 16 grackle nests were placed in the
conifers.
Also in 1971, a count was made of the breeding grackles
on the Michigan State University campus. This is part of

u
the area previously censused by Eyer (195%). Due to limita-
tions in time and the abundance of the grackles, the area
censused in 1971 was decreased from 12 to 3.5 square miles
which mainly includes the central university campus and some
surrounding farmland. (T.nN., R.2W., h2°ho'N, 8h°2o'w, East
Lansing, Ingham County, Michigan) The campus now is pri-
marily landscaped and extensively planted with ornamental
shrub and tree species from temperate areas throughout the
world. Shrubby Swampland no longer exists in the area al-
though a small cattail swamp still remains north of the rail-
road tracks. Figure 1 shows the area and both 1952 and 1971
colony sites. Mbst of the 151 grackle nests in the censused
area were built in conifers; the rest were located in man-

made structures.
METHODS AND MATERIALS

From the middle of March until the second week in
April, 1971, during the pair formation and early nest-building
period, data were collected concerning the relationship of
group size and the intensity of breeding activities. Group
size and the number and type of vocalizations occurring with-
in five minute observation periods were recorded. All obser-
vations were made before 10:00 EST. Vocalizations were broken
down into two categories based upon the presence or absence
of responses from other birds. The display vocalizations
used as major components in this breakdown were the Ruff-out

Squeak (ROS) of the male, and the Ruff—out Chuga (ROC) of the

 

 

 

 

 

 

 

A N I. A_N

 

 

 

 

 

 

 

 

 

 

'11 15] 5
Gracklo Celonlos 3
3 21
1952-0 0 0
1971-0 ° .13 43 ....... -
3300 9:.

.14

 

 

 

Figure 1- Locations of nesting colonies of Common Grackles,

East Lansing, Michigan. None of the 12 solitary nests is
shown.

female (Ficken, 1963). The categories paired the ROS or ROC
with two possible response alternatives, a situation in
which a response was given by another bird and a situation
where no bird responded to a particular vocalization. The
recording technique necessitated that many ROS's or ROC's be
recorded twice, once as a stimulus and once as a response,if
the vocalization was part of more than one sequence. Regres-
sion analysis (Sokal and Rohlf, 1969, pp. #20 ff.) is the
appropriate technique for testing the relationship between
these variables. A colony was defined as more than one pair
of grackles. Distance between nests was usually less than
50 feet, but at times was as great as 220 feet. Nests in

the Manhattan Marsh and Rose Lake areas were visited daily
in 1971, and two or three times weekly in 1968 and 1970.
Nests in the censused area in East Lansing were visited at
most two or three times during the 1971 nesting season and
often only once. As is the case with Eyer's (195%) census,

I attempted to census only breeding birds, not the entire
population, so any non-breeding surplus escaped notice.

Nests located during the building stages were followed
especially closely. Table 1 gives the length of time required
by a female to build a nest, lay her eggs, incubate them, and
rear the young to fledging. Data from Eyer (195”), Peterson
and YOung (1950), Bent (1958), and Schaller and Emlen (1961)
were used in addition to my own in compiling this table. From
available information about a particular nest, I used this

table to estimate the date when that nest was started. Names

, .mnonudm maoaam> no zaamo coaumooaz

on» no mama on» ma =mdax3wsum no unfinooonomr .onoa no noon om 0» 0m «0 woodenmao

you manaoda meadow haoo ocnaonfi mmcfiaumo: on» an coma moSBOm on moononoMon omega n m
.maomav uoaam one

noaamaom : o «madam man» u o “nommHv undo» cam nowaoumm a n .AJmmHV nohm s a “Sony mama

 

 

0:.m: nun: mum nun: mega o>woq op wawmcoam
m:.m:-mm ---- ---- ma.ma-m omeeazaeem oo maeaeoeaem
m:.~:-mm ---- ---- ma.ema-m Haeaaom Hezea> on maaeoeoeo
em-am ---- ---- e.: eeeaaeea
mm ---- ---- m-o meaaeea oe aooem
om m.NH nus- n.ma coapmnaooH
ua-oa mu: m.: m-: «canon mum
NH H.N o.m w.m . pamz
¢Im H.N :.m m mdfinaq
o N m N Q50 6::
J m.m Q50 onwwpao
H omw ea.mw ea aaoeoeaa
mama o>Humass=o ouwum no summon macho mo madam

 

:oHuoaoOHQom manomno coaaoo mo manna mafia
a magma

8

of nest-building stages are those of Eyer (195%), but not the
parallel ones of Peterson and YOung (1950), or Holcomb and
Twiest (1968), which have been adjusted for presentation in
the table. In many cases a nest was determined to have been
started between two dates, as much as two weeks apart in
cases where only one observation was made during the incuba-
tion period. In all cases the nest was assumed to have

equal probability of being started on any day during the
interval. Mean starting dates for each colony and the popu-
lation as a whole were computed from these data with days
during the nesting season numbered consecutively (April 1 = 1,
May 1 = 31, etc.). A variance of nesting dates was computed
for the entire population and for each colony based only upon
those nests for which the starting date was known with‘: 1 day.
Mean colony nesting dates and variances were tested against
the total population estimates to determine the extent of any
significant deviations, using Student's t-test and the F-test
respectively.

This procedure differs from that of Horn (1970) in that
the statistical variance is used as a measure of the temporal
closeness of nestings in a colony, and all nests in a colony
area are included as part of that colony. By excluding
late nesting individuals a large number of the younger birds
would be eliminated from consideration as members of a colony
to which they really belong. The difference in the methodology
is based primarily on the use of dates of beginning nest-build-
ing for the data set in this study as opposed to the dates of

9
beginning egg-laying in Horn's study (1970). Horn believes

that invalid results may be obtained by using the variance of
nesting date within a colony as the measure of synchrony un-
less a population variance canbe used for comparison. Part of
the purpose of conducting the 1971 census was to provide a
population against which the synchrony of a certain colony
could be tested. If the variance within a colony is signifi-
cantly less than the variance of the population as a whole,
the evidence favors social interactions synchronizing nest-
building behavior. If the colony variance is not significantly
different from that of the population, it is an accurate pre-
dictor of population variance. Thus colony members are not
having a synchronizing effect on each other because they are
no more synchronized than the entire population.

Following the same reasoning, there will be occasions
when the mean nesting date for a colony is significantly
‘earlier or later than that of the population of which the
colony is a part. In these situations, a cautious state-
ment can be made about the age structure of the colony. If
the colony nests earlier than the population, it is composed
primarily of older birds, and if it nests later, it is com-
posed of younger birds. Selander and Hauser (1965) found that
in the related Great-tailed Crackle (W W)
first year females breed later than older adults. Conadal
maturation of first year males also occurred later than that
of the adults, although they do not breed. Common Grackles

of both sexes breed in their first year. Assuming the same

lO

situation is true for Q;_ggi§ggla, an earlier nesting colony
will be composed primarily of older birds, and a later nesting
colony will consist mainly of first year birds.

The nest site itself was catalogued for each nest as to
location and type of site, i.e. mainly genus, but in some
cases species of plant, or man-made structure. For each
colony, an estimate of the number of potentially available
nest sites was also determined, based on a one tree : one
nest theorem. Since more than one nest may be placed in a
single tree, these estimates are conservative ones. unfor-
tunately, a light meter was not available to measure the
threshold density of vegetation necessary before a tree might
be used as a site.

The relationship between available sites and sites
selected, noted by tree genus or species, was tested using
Fisher's Exact Test or a C-test for independence in R x C
contingency tables (Sokal and Rohlf, 1969, pp. 59 ff.).

These very easily performed tests involving the functions

Y = logf! and Y = f lnf respectively are superior toixg tests
especially for my purposes. Only colonies where more than
one site was available were tested.

Eggs were marked with nail polish in the Rose Lake
colony in 1971. Ybung birds were individually marked on
the foot or tarsus, using a black felt-tip marker during
their early nest life. The felt-tip marker is preferable
to nail polish for marking the young because it is easier to
apply, lasts Just as long, and dries much more quickly,

11

which prevents the young bird from rubbing the marker all
over itself. Birds were ringed with U. S. Fish and Wildlife
Service aluminum bands after the eighth or ninth day in the
nest, and color-banded with plastic leg bands on the 11th or
12th day of nest life. On each visit to the nest the young
were weighed in grams, and on most visits the wing, tail,
tarsus, culmen, and the innermost primary (No. l) on the
right wing were measured in millimeters. The wing measure-
ment was taken from the bend of the wrist to the tip of the
'manus in young birds, and to the tip of the longest primary
in older nestlings, the tail from its base on the pygostyle
to the tip of the longest central rectrix, the tarsus from
the center of the tibiotarsal Joint to the last undivided
tarsal scute, the culmen from the tip to the posterior end
of the cornified area, the innermost primary from its pro-
jection through the skin to the tip. In addition, the per
cent of this primary length taken up by the erupted feather
was recorded. weights were taken with a Pesola 100 gm. x

1 gm. division scale obtained from Bleitz Wildlife Foundation,
Hellywood, California.

Seventy nestlings collected for sex determination con-
tributed significantly to a near 100% mortality at the Rose
Lake colony in 1970. The birds were taken in either the 9th,
10th, 11th, or 12th day of nest life. Sex was determined by
internal examination of the gonads and the same measurements
were taken on these specimens as on the living nestlings.

Mbst of the specimens were deposited in the Michigan State

12
University Museum as spirit or skeletal material. The re-
mainder were given to the Grand Rapids, Michigan Public

Museum as skeletal material.
RESULTS AND DISCUSSION

Changes in the East Lansing Population

Eyer's (195M) census of the East Lansing population was
taken in May, 1952. He found 7% nesting paris, 72 in nine
colonies and two solitary, in a 12 square mile area. (Because
his paper will be referred to often for comparison with the
present census, it will not cited every time.) My 1971
census was limited to a 3.5 square mile area in the center
of Eyer's census tract (Figure l). I found 151 nesting pairs,
139 in 17 colonies and 12 solitary pairs. The 1952 grackle
population in the 1971 census area was 2% pairs in three col-
onies and no solitary pairs. Both of these censuses are
conservative estimates of the actual population, since non-
breeding pairs may have been missed in the 1952 census,
and some early nesters had already finished before the 1971
census was complete. Most nests, both colonial and solitary,
were located in coniferous trees of nine or more different
species. These included several different species of spruce
(Blggg), red or Norway pine (Rings resinosa), white pine
(P, strgggs), Austrian pine (£,‘gig;§), several species of
fir (Aging), Scot's pine (Rings sylyestris), juniper
(Junipgggg spp.), Arborvitae (Thug; occidentalig and some
ornamental Junipgggg), Japanese red pine (Pinus dengiflggg),

13
and Ponderosa pine (P. pgpgerosa). Of these potential sites

the\Austrian pine appears to be the most favored kind of
tree, with the others selected about equally as nesting places
if P. giggg was absent. Certain ornamental Junipgrgs species
are very similar in color and texture to the IQ!1§.Q££1§§£§211.0
In the analysis that follows these are included as "Arborvitae"
rather than as "Juniper". Numbers in the following discussion
indicate the number of the colony in Figure l and Tables 2, h,
5, and 6. Table 2 presents a comparison of the two censuses.
Three of the 17 colonies discovered in 1971 were used
in 1952 as well. Two of them, colonies No. 2 with 23 nests
and No. 9 with 11 nests, are essentially the same now as
they were before except that the trees used for nesting are
larger. Each contained nearly twice as many nests as in
the previous study. The third colony, No. l, where Eyer
carried out much of his detailed work, was a marsh in 1952,
but has since been filled along the eastern edge and a
superhighway occupies the central and western portions of it.
This colony, which has switched from a marsh to a suburban
nesting habitat, is the same size as during the earlier
study.
Colony No. 1% was composed of ten nests, all in the
same Austrian pine. It is the only potential grackle nesting
site in a quarter-square-mile area. The closest potential
site was an occupied pine, the only nest in a pine at Colony
No. 13. This coincidence may indicate that the tree was

saturated, as discussed in greater detail later.

it

Table 2
Common Crackle Breeding Population in East Lansing, Michigan

 

Colony Breeding Pairs

‘ 19528 19?1

l 6 6

2 12 23

9 6 11

10 NEb 1h

11 NE 3

12 NE 10

13 NE 10

1% NE 10

15 NE 6

16 NE 11

17 NE 11

18 NE 3

19 NE E

20 NE 2

21 NE 7

22 NE 5

23 NE 3

Loners - no. 0 12
n - %° 0 8.1

Total - colonies 3 l7

" - pairsd 2H+5 l5l+lO

Mean colony size - pairsc 8.0 8.5

Density - Birds/m2 l3 .7 87

 

data from Eyer (195%)

Nonexistent

differences not significant (p>0.05); other pairs of
items not tested

includes estimate of suspected error

D: 00‘”
l

15

The nine pairs at Colony No. 16 and three pairs at
Colony No. 23 were the only grackles in the census area not
nesting in coniferous trees. The nests were placed in open-
air concrete parking ramps. These colonies provided the only
examples of the minimum internest distance tolerated by fe-
male grackles. As with many other colonial Icteridae (Chapman,
1928; Skutch, 195%; Schafer, 1957; Orians, 1961b; Selander,
1965) the female defends an area around the nest. Once, at
each ramp, a female started building a nest too close to an
already active nest and was unable to finish building in one
.case (internest distance 10 feet, Colony No. 23), or to lay
any eggs in the other case (internest distance three feet,
Colony No. 16). These colonies also provide the only cases
I know about where an entire nestful of young starved in
spite of both parents being present. This was true of both
active nests at Colony No. 23 where no young fledged from
the colony. All the young of one nest at Colony No. 16
also starved, perhaps because of predation on the parents al-
though that seems unlikely. The young at Colony No. 23 were
emaciated, weak, and small for their age. A possible cause
of death in a high-use parking ramp is carbon monoxide poi-
soning, but as many nests in Colony No. 16 fledged young
successfully, I discount this as a mortality factor.

Twelve pairs of grackles (8.1%) nested solitarily in
1971, an increase of 5.%% over Eyer's figure. This increase
is not significant and thus gives little insight into the

mechanism maintaining or retarding the occurrence of solitary

l6

nesting. This feature of grackle biology deserves further
study. All isolated nests were located in coniferous trees,
about equally divided between sites potentially available in
1952 and those that were unavailable at that time. Not all
sites were isolated either, as nests were located in exten-
sive borders of pines planted around parking lots as well as
in single trees.

The supporting vegetation for five of the 17 colonies
occupied in 1971 was present in 1952. These areas account
for 85 of the 151 nesting pairs (56% of the current popula-
tion) which is an increase of 61 pairs (253%) over the earli-
er figure for the same area. An increase in the number of
available nest sites has probably been a factor in the increase
of the population. This involves more sites within single
trees as they grow, and new sites made available with the con-
struction of suitable buildings and new plantings of trees.
A further factor in the increase of the grackle population may
be the severe reduction in the Robin (Tgrggs migratorius) pOpu-
lation on the Michigan State University campus following pesti-
cide applications for Dutch elm disease control in the mid-1950's
and later (C. J. wallace, pers. comm.). As Robins and Common
Grackles compete for nest sites at times (Hamel, MS) a reduc-
tion in the Robin population would make additional nest sites
available to the grackles. Conifer plantings made between
the two censuses allowed 66 pairs, 269% of the original popula-
tion, to breed in 1971. An increase in the nest sites avail—

able for a population creates a potential for increased

17

production of young (Haartman, 1956) which must be taken into
consideration when trying to pinpoint regulating mechanisms
for a bird population. The evidence presented here indicates
that nest sites were in excess of number needed at the time
both censuses were made of the grackle population in East
Lansing. The site is thus normally a minor factor in con-
trolling population size in grackles in suburban areas,
although the case in swamps is not yet settled. The existence
of an excess of nest sites must be verified before other poten-
tial regulating mechanisms can be tested. When sites are
limiting, as probably occurred in 1971 at Colony No. 1%,

the effects of other factors will be obscured.

Social Stimulation of Breeding and Nest-Site Selection

The group has been considered as having a stimulatory
effect on individual members of bird aggregations, both be-
haviorally and physiologically (Lehrman, 1959; Crook, 1961;
Orians, 1961b; Selander, 1965; Selander and Hauser, 1965).
Austin (1951) has documented group tenacity in breeding Com-
mon Terns (Sterne hirgngo) and Horn (1970) has diagramed
social interactions among Brewer's Blackbirds. Darling (1938)
claimed that group interactions produced early mean nesting
dates and nesting synchrony in gull colonies. Coulson and
White (1956, 1960) have since discounted Darling's findings
by showing that in the kittiwake (fli§§§_trigactyla) age of
breeding birds is the variable responsible for the observed

effect. As is the case with the kittiwakes, young Great-

18

tailed Grackles breed later than the adults (Selander and
Hauser, 1965). Circumstantial evidence indicates that this
is the case in the Common Crackle as well. In testing
whether the behavior of a group of birds is stimulating in-
dividuals to synchronize some aspect of their breeding bio—
logy, it is necessary to examine the null hypothesis that
there is no synchronization at the point in the breeding
cycle when the contagious behaviors are most apt to be
effective. That part of the cycle most appropriate for this
test will vary from one species to another. In the Brewer's
Blackbird, it is contagious precOpulatory display by the
females (Horn, 1970), and in the Common Crackle group song-
display behavior is the synchronizing feature, as a short re-
view of the phenology of the species' reproductive activities
shows.

Grackles arrive on the breeding grounds in the central
Michigan area in early or mid-March but do not begin nest
building until the first or second week in April (Eyer, 195%;
Bent, 1958). In the intervening weeks, members of a group
or colony spend much of the day as a single large group or as
several smaller ones in the area where the colony will be
located. Pair formation occurs mainly during the earlier
part of this period. Females are chased by males in the
leader flights described by Ficken (1963), and the level of
aggression, especially on the part of the males, is very
high. Supplanting often occurs during these flights, one
male aggressively taking the position directly behind the

19

female. Head-Held-Ups (Ficken, 1963), or bill pointing

(Eyer, 195%), as the display has been variously called, are
very common when birds meet each other on perches, especially
among the males. This display of sleeking the feathers and
pointing the bill vertically upward is the generalized aggres-
sive display of this species and perhaps serves that function
among other icterids as well. As the pair formation period
progresses and more birds become paired, the group begins
examining various trees and shrubs in the area for their
acceptability as nest sites. Commonly, the ma1e(s) will sta-
tion himself at the top of a tree and the fema1e(s) will dis-
appear into the foliage of the conifer or branches of the
shrub where a nest might be placed. After a period of a few
seconds or minutes, the pair or pairs will move to another
tree and repeat the process. This activity serves to acquaint
the members of a colony with the suitable sites where the fe-
males will build and the suitable points available to the
males as lookout posts (Eyer, 195%). Similar kinds of ex-
ploratory activities have been noted in female American Red-
starts (Setophaga rgticilla) as they seek out nest sites (Fic-
ken, 196%). At times the males grackles themselves become
involved in nest site evaluation, building "dummy" nests in
various places (this study; Eyer, 195%; Peterson and Young,
1950). The dummy nest usually consists of a handful of dead
blades of grass entwined around the twigs at the site, and is
easily differentiated from actual nests in the "platform"

stage. Females do not use the flimsy materials for the base

20
of their nests that the males use in constructing the dummy
nests. Data concerning the influence of a male's dummy nest
on his mate's choice of nesting site are nonexistent. The
nest is not always built atop a dummy nest because dummy
nests are found that have not been used as bases for actual
nests. The extent to which males participate in dummy nest
building is not known. If many such nests are used as bases
for actual nests, an observer would be unable to gauge the
occurrence of the practice by simply counting dummy nests
after nest-building. This study does not demonstrate a direct
role for the male grackle in nest-site choice.

During the pair formation and nest-building periods,
the birds are very often singing. The male's song-display is
the Ruff-out Squeak (ROS) and the female's is the Ruff-out
Chuga (ROC). The members of a pair keep in contact with
each other by duetting, the male giving an ROS and the female
responding with an ROC (Ficken, 1963). Group size tends to
be larger in the pair formation period than during the nest-
building period. Table 3 demonstrates this. The difference
in frequency of occurrence of the group size N = % is enough
to explain the significant interaction between group size and
period of the nesting season. This is understandable since
the female spends more of her time building the nest then and
has less time available for the group. Because the females
spend less time with the group after nest-building has begun,
they have less contact with the potentially contagious be-

havior of the group. Thus, the place to look for any

21

.coHpmH>oo one no nofiuaoa gonna map mopnnanunoo nadaoo man» a n
.moo.ov menace an mcaaasooo nodusnaapmao man» mo moaaanwnoaa on» n m

 

 

 

nude mm ma om :o mam and mom mom eeaeeoa
mm: a a ma am or no Hua man meaoaaem-omez
are mm m ma m: «me man man owe aoaeeaeoa teem
Heooa aA a o m o: m m a

ome macaw macho ho ommpm

macho mnfiuwmz mo mumpm mm owfim nsoao mo honmduonm

m m3 we

22

synchronizing stimulation is in the group before nest—build-
ing begins, and the effects of stimulation, if it occurs,
should be seen in the phenomena associated with nest-building.
Further synchronization may occur at later times in the cycle,
but in order to demonstrate that the colony is a dynamic
grouping, a synchronization need only be shown at one point
in the cycle.

The intensity of singing behavior, as measured by the
number of ROS's or ROC's eliciting responses in a group of
known size during a five-minute period, is the potentially
contagious behavior most easily quantified in groups of the
Common Crackle. As both the ROS and ROC consist of a dis-
play posture as well as a vocalization (Ficken, 1963), they
can be stimulating both visually and auditorily. A bird
that cannot see the other displaying in the group can none-
theless hear it and respond on the basis of the auditory
stimulus. Grackles may respond to an auditory stimulus only
as well as to the audio-visual stimulus of a display. A fe-
male building her nest in a tree apparently will respond just
as quickly to her mate's song when he is on the ground and
she cannot see him as she does when both are sitting close
together in a tree. The display behavior may be contagious
as indicated by the observation that when one of four or
five birds sitting together in a tree displays, the others
follow suit, one after the other. The appropriate test of
the hypothesis that display behavior is contagious is a

regression of vocalization pairs against group size. If the

23

relationship is linear, the inference is that each bird con-
tributes a certain amount of display irrespective of group
size. In such a case, the behavior is not contagious. If
the relationship between group size and vocalization pairs is
non-linear, i.e. quadratic or higher order, then the inference
is that the individual bird is stimulated to contribute a
greater amount to the interactions of the group as the group
increases in size. Figure 2 shows the regression of the
square root of the vocalization pairs against group size.
This regression explains more than 99.9% of the variation in
the data, indicating that the relationship between vocaliza-
tions and group size is non-linear, and that the display be-
havior within colony groups of grackles is definitely con-
tagious.

The hypothesis that social stimulation will have a
synchronizing effect on some aspect of breeding biology
readily follows upon a determination of contagious behavior
patterns in the group. In the Common Crackle the nesting be-
havior is the first breeding event following the pair forma-
tion period of group activity. This is the test behavior for
examining whether or not synchronization occurs in this species.
Table % presents a breakdown by colony of means and variances
of nesting dates. Of the four colonies on campus where a
variance could be computed, three of them were significantly
more synchronized than the total population. Each of these
colonies had a mean nesting date earlier than the total popu-

lation, significantly so in the case of Colony No. 12. The

Figure 2. Regression of the square root of number of

vocalization pairs on group size.
p < 0.001.

Y

2.568 + 0.230%X,

.-——. -—- -———- ._—. *—

_*—s ——‘——_—_

2%

   

 

 

5
Song 4
Pair:
3 0 ~90" Iormcflon
o-uou building
x«-(X,7)
2 o
O
1
O 5 10 15 2O 25 30

Group Six.

Figure 2.

Table %

25

Colony Development in Common Grackles in Central Michigan, 1971

 

 

 

 

Colony n nes- df $2
ting date

1 6 26.33 -- ~-

2 23 8.k0 -- --

9 11 20.35 -- --

10 1% 9.0 6 1+1..62a
11 3 12.0 2 6.00a
12 10 10.62c A 9.188
13 10 21.1k 6 600.25
1% lO -- -- --
15 6 13.75 -— —-
16 11 23.80 -- --
17 11 22.0 -- --
18 3 10.33 -— --
l9 % 18.12 -- ~-

20 2 8.5 -- —-

21 7 5.1% -- --

22 5 20.25 -- --

23 3 23-75 -- --

2% l2 19.12 -- --
Rose Lake, 1970 30 15.57 26 551.%1b
Rose Lake, 1971 16 15.73 13 235.00b
Total, East Lansing 151 15.53 23 1568.16

 

a ~ These variance figures are significantly (P<0.01) different
than the total population variance; b - these variances are not
significantly different at p=0.05; c - this mean colony nesting
date is significantly different than the population mean,

p<0.05.

26

one colony where there was no significant synchronization
had a mean nesting date later than the population mean. The
fact that within-colony variances are less than the population
variance demonstrates that the birds are synchronizing each
other's nesting behavior. There is no reason to believe
that environmental conditions over the area of the census
were different enough to warrant consideration of this
factor as the proximate one triggering within-colony nesting.
The mean nesting date at the Rose Lake colony ten miles away
from all the campus colonies was the same for two consecutive
years even though 1970 was a normal year and 1971 was a dry
year. Both mean dates were the same as the mean date for
the East Lansing grackles. In addition, the colony means
are not distributed along a discernible geographic gradient
in the census area itself. The mean nesting date and colony
variance data suggest that the late-nesting birds which I
assume to be young birds (of. Selander and Hauser, 1965) may
be less efficient at achieving synchronized breeding than the
adults. The techniques discussed in this paper provide the
necessary tools for determining whether or not such an hypo-
thesis is accurate.

The unique parklike atmosphere of the Michigan State
University campus provides appropriate conditions to test
the effects of this contagious behavior in yet another inde-
pendent way regarding nesting. In several places where there
was a nesting colony of grackles, there were several species

of trees which offered a diversity of potential nest sites.

27

The literature concerning phenomena associated with choice

of nest sites by the Common Crackle is an abundant and
interesting one. Published records of nest sites include

a great variety of locations,such as swamp shrubs, cavities
in trees, deciduous and coniferous trees, Osprey (Pangion
haliaetus) nests, beaver lodges, bird houses, and man-made
structures (Attwater, 1892; Stockard, 1905; Cameron, 1907;
Cahn, 1920; Youngworth, 1932; Lloyd, 19%3; Axtell, 1955;
Bent, 1958). However, one important point about nest-site
choice is that the birds in a single area or colony are gen-
erally noted as all nesting in a single type of site (e.g.
Colsan and Holt, 191%; Peterson and Young, 1950). Certain
observers have noted that a colony may switch nest sites from
one year to the next without changing colony location (Tracy,
1896; Peterson and Young, 1950). This switching phenomenon
has also been noted in the Brewer's Blackbird (Dawson, 1923,
p. 86).

Table 5 shows the sites available to and used by differ-
ent colonies. In 13 of the 1% colonies recorded in the table,
the modal nest site for that colony was the same as the site
first chosen. In seven of these cases the modal nest site was
not the same as the most abundant site. Due to the small
number of nests in any given colony, it is difficult to per-
form G-tests for independence of site availability and site
choice. Because of this factor and because of elimination
from the testing of colonies containing trees with more

than one nest, only six colonies could be tested for the

28

.hnowoumo was» ouofi

Ham“ modanm ooamaucooaqd one .ocam com ommnmamh .onfim mmoaoonom .xooaamm .momnmama .3oaaa3
momam .afim n o mzcmmono mopam u mmpam maanHm>m= mm oommmamwm monsmfim a o “Genome mouam

one moufim maanHm>m nmmzpmn Aao.ovov coauomampaa uanHmwnwflm u o memosaocfi mam mannaam>m

roam mno can» once mafia mmacoaoo hano n m mooomam mm: mqoaoo Ga ammo pmnam when: open a *

 

ou:a mama mm
mama a": one am
Hum mum* on
Humm oum cummr a“: Huma mnmm ma
ouom Harm muooma ma
aura muow ma
wuoma muoa we
on name . mumaa oum m": oum muwa Boa
mum oua manna on: ouem 6
owner mum onoa-m Humam m
Hue :uomH* Hum one: H
anoom mauowa* mama
anoqu mmom
Huoom mmuowfi* came
.nmqu omom
cum omuadeaa moan .amaez
ameeendez
moeam amen omen eeon> omen menu mean
6 oawm uQOuusm amofiqsh m.uoom neonaa nmflaumsa mpflnz com museum
oeeeam Baez eaeoaoo

mmaxomau nmmHnOflz Hmapcoo :H moaono oufim pmoz
m magma

29

relationship of choice and availability of nest site. The
results were mixed. Three of the colonies, Rose Lake in
both years and Colony No. 10, show very significant inter-
action indicating that the birds are choosing the sites,

i. e. that the nests are being placed in trees in some other
fashion than as a function of the abundance of that tree.
The other three colonies, Nos. 1, 12, and 13 do not show
significant interaction. These results can rightly be ques-
tioned in the cases of colonies Nos. 1 and 13 because of a
limited number of nests in Colony N0. 1, and a very limited
number of one of the nest sites at Colony No. 13, factors
which interfere with the operation of the tests. Colony No.
12 yielded a valid test and the nests were indeed placed in
that colony in proportion to their availability. There are
two possible explanations of this result. Nest sites may
not have been selected in that colony in any other manner
than at random, or, the peculiar conditions of site avail—
ability masked out any choice behavior on the part of the
birds. Despite this potentially negative result, the modal
nest site, except for the aberrant case discussed below, in-
variably occurs in the site where the first nest was placed
regardless of the relative availability of that site. This
indicates that the contagion of display behavior carries
over into a socially facilitated nest-site selection. Mea-
surements of the spread of nests within colonies were taken
for eight of the colonies (Table 6). In only Colony No. 13

was the first nest at a distance much greater than the average

3O

 

ma

mm
0..”
+2”

HN
N

Hume .erea atom
came .oaea omom

aqoaoo

 

cwoapnoo o»
mommumwn name A

vacaucoo 0p
ooompmflo came an
unmaoomam

ofloaucoo o»
oocmumfic coma v

QOHOO .HO GHOHquU Cu GOflpmem EH #0102 umHHrm .HO quEmONHm

o canoe

31

distance to the centroid, or numerical center of gravity, of
the colony. Thus, not only was the first nest the trigger for
site choice by members of the colony, but also this nest is
close to the center of the colony as well with other nests
being built around it rather than away from it.

Balancing the gregarious tendency of the females to
build near to the first begun nest is the increased aggression
built up by nesting close together, since the females defend
their nest sites. Figure 3 presents the measurements of dis-
tance to nearest neighbor for the following seven colonies:
Rose Lake, 1971, and colonies Nos. 13, 1, 1%, 23, 16, and 21.
As the internest distance is in most cases a function of the
spacing of the sites the largest internest distances do not
provide as much information as the minimum distances. It is
in the latter case where the aggressive behavior of the birds
becomes a factor in determining the spacing of nests. Eighty-
six per cent of the nests sampled lay within 60 feet of their
nearest neighbor, well within the range of potential visual
and auditory stimulation by other birds. Three nests in the
East Lansing population were unsuccessful because they were
built too close to already active nests. Both of those built
within five feet of an already active nest failed, and one of
13 built between five and ten feet of the nearest nest also
failed. From these data a rough estimate of five to ten feet
can be made as the minimum distance defended around a nest

site.

32

CNN CON

.mmfinoaoo oaxomnm aw moonmpmfic amonnounH

one. .... ...... acute...

om. on. oe. om. con .6»

a. Cue-non:—

00

0%

ON

 

 

EL

C E C

now. ....) nouoc-U

«one: tosnmc..c:u.

 

 

 

 

 

 

 

 

 

.m mnSMfim

In
Aauonboa‘

or

me

33

The existence of many potential but unused nest sites
in the colonies in Table 5 indicates that nest sites are not
normally a limiting factor for the grackle population. This
is not always the case. For example, Colony No. 1% consisted
of ten nests all within the same medium-sized Austrian pine,
the only potential nesting location within a circle of a
quarter—mile radius. As it was discovered too late to observe
the number of pairs that may have tried to nest there, it is
uncertain whether ten was the maximum number in that colony
site or not. One nest in Colony No. 13 was perhaps the oddest
nest of all studied. While located in one colony, it was be-
gun a week earlier than the other nests in that colony and
was placed in a different site than all the others. The adults
of this nest foraged toward the south while the other birds
were doing their primary foraging to the north. This circum-
stantial evidence suggests that that nest may not have be-
longed to the colony. Since it was placed in the site nearest
to Colony No. 1% approximately two-tenths of a mile to the
south, it may have been a member of the latter colony, indica-
ting that the Austrian pine occupied by that colony was satu-
rated. Had Colony No. 13 not developed the potential rela-
tionship of the aberrant nest there to Colony No. 1% would
not have been recognized. It would have been considered
simply as a solitary nest. The function of solitary nests
in the population is still unexplained, but a promising lead
is that some of the nests are the efforts of pairs unable to

find sites in established colonies. In such a situation,

3%
solitary nests may serve as guides for the subsequent ad0p-

tion of new colony sites.

Sex of Nestlings

In the absence of cloacal protuberance or incubation
patch information adult grackles may be sexed by wing and
tail measurements (Wood, 1969). There is some evidence that
culmen, tarsus and weight measurements can also be used to
sex the birds (Snyder, 1937). In all of these measurements
the males are somewhat larger than the females although there
is some overlap in values. A scheme for sexing nestling Red-
winged Blackbirds has been deve10ped by Nero (1960) based upon
the weight of known-age birds. A preliminary study of young
grackles carried out at Manhattan Marsh in 1968 indicated that
marked size differences existed among nestlings late in nest
life, and further that a color phase phenomenon might be associ~
ated with this size difference. In 1970, 70 nestlings were
collected from the Rose Lake colony. The sex ratio of this
sample was not significantly different than 1:1. After all
the young were sexed and measured, I tried to relate differ-
ences in measurements and color phase with sex, and since the
birds were collected at known ages, the relationship with age
was also tested. No significant difference between the res—
pective male and female means was noted for any measurement
at any age. The variation of the wing length, first primary
length, tail length, and weight of the specimens was so high
that these variables were eliminated from any other considera-

tion, leaving culmen and tarsus measurements as the most

35

reliable characters to use for sexing the young. The varia-
tions in even these two characters were very large among
nine- and ten-day 01d birds so only culmen and tarsus data
for 11- and 12-day old nestlings were used for the following
analysis, a sample of 28 nestlings.~ Figure % presents the
data on these 28 nestlings. The ages are pooled because
essentially no difference existed between the two subsamples
for these criteria. Because of the small sample size no
attempts were made to test this distribution. In dealing with
criteria for determining sex of animals externally, measures
of central tendency of the distribution of a criterion are
not as important as the overlap between the two groups. Each
individual, representing one point in a distribution, cannot
be assigned to either class which overlaps its measurement,
even though the classes have significantly different means.
Criteria were determined as follows: the lower limit of the
male distribution for both tarsus and culmen was set above
the measurement of the largest female, and the upper limit
of the female distribution was set below the measurement of
the smallest male. The results of the application of these
techniques are shown in Table 7.

Tarsus and culmen measurements were available for 18
young 11— or 12-days old from fieldwork in 1971. Application
of the sexing criteria to these birds allowed identification
of 13 of them (72%) into known sex categories, nine females
and four males; five of the young were intermediate. The

proportion of unidentifiable young in the 1970 and 1971

Figure %. Tarsus and Culmen measurements and color phase
of nestling grackles. One individual, an "Intermediate"

female, Tarsus = 25 mm, Culmen = 12 mm, has been omitted
from the graph to conserve space.

Tarsus,

Figure %.

37

.36

35

34-

.33

.32

31

30

129

 

36

 

II
D
I
I
o In 9 I:
9
e o 9
I I
00 I
I
O
O 00
Color Phase Sox
6 9
Charcoal I I
lnformodiato Q G
F 0 light Gray D O
Q
13 14 15 16 17
Culmen, in mm

37

Table 7
Criteria for Sexing Nestling Grackles 11 and 12 days 01d

 

 

Criterion Component Sample
Classes ‘ Differentiated

1. Tarsusa > 35 = Mb % M
measurement 32 < x‘g 35 = U 10 M, 6 F
alone 3,32 = F 8 F

identified 12 - %3%

unknown 16 - 57%
2. Culmen 3,16 = M 6 M
measurement 15 g_x < 16 = U 8 M, 8 F
alone < 15 = F 6,F

 

identified 12 %3%
unknown 16 - 57%

3. Combination of Tarsus and Culmen measurements

Tarsus > 35 and/or Culmen 2 16 = M 9 M
Tarsus intermediate AND Culmen intermediate = U 5 M, 5 F
Tarsus $.32 and/or Culmen < 15 = F 9 F

 

identified 18 6%%
unknown 10 - 36%

 

%. Tarsus length .2 33 = M - 13 M, 6 F

Illinois 31,5 x < 33 = U 1 M, h F

(Willson et al., 1971) < 31 = F % F
identified 17 - 61%
unknown - 18%

MISIDENTIFIED 6 21%

 

a — all measurements expressed in millimeters
b - M = male; F = female; U = unknown sex

38

samples is approximately one-third of the total number, not
a very high amount considering that about 20% of the adults
are intermediate (Wood, 1969). Mere females than males are
represented in the 1971 sample because some nestlings had
fledged from the nests before the measurements were taken.
As males are generally larger than females it is not unlikely
that the early nest leaving group is biased in favor of males.
The broad overlap of the measurements of the two sexes necessi-
tated that these criteria have a wider range of intermediacy
than Willson et a1. (1971) allowed, based upon data from
Illinois birds.

Data from this study only permit comparison of length
of tarsus criteria used in the two sexing schemes. The tar—
sus criterion of Willson et a1. (1971) is included in Table 7.
The sample of nestlings used was the sample collected at Rose
Lake in 1970. Because these birds were younger than the 13-
day olds for which the criteria were designed it might be ex-
pected that slowly maturing males would be identified as females.
However, the scheme incorrectly identifies six females as males,
more than are correctly identified as females, and no males are
misidentified. It is not clear how the sex of the Illinois
nestlings was determined. The Common Crackle in Illinois is
the same subspecies, Q. g. versicolor Vieillot, as that in
Michigan (A.O.U., 1957), and as this subspecies is very con-
stant throughout its large range (cf. Chapman, 1892) it seems
unlikely that differences in measurements at locations within

several hundred miles of each other should exist. Therefore,

39
extreme caution should be exercized in applying the criteria
of Willson et a1. (1971) in sexing young grackles.

Contingency analysis based upon sex and color phase of
nestlings using the C-test is shown in Table 8. The three
color phases refer to the pinfeathers of the crown and upper
breast viewed in good light. They are regions along a color
continuum rather than discrete sets. "Charcoal" birds are
very dark, almost black in coloration and appear rather
glossy. "Light gray" birds are much paler, gray or tan gray,
and appear dull rather than glossy. Two of the 70 nestlings
had blurred vertical breast streaks in addition to being
"light gray." Two of 3% nestlings handled in 1968 at Man-
hattan Marsh were also light gray with streaks. "Intermedi-
ate" birds are an intermediate gray color and not glossy.
Color phase does not appear to have any direct relation to
size (Figure %), but it is significantly related to sex
(Table 8). The probability of obtaining at random the dis-
tribution in Table 8 is less than 0.05. This character is
not an accurate one for sexing the young since all three
phases occur in both sexes. It is not known if this color
phasing is related to the iridescence of the adult plumage
but the possibility seems worthy of further study.

These criteria for sexing nestling grackles are not
statistically significant and they are based upon limited
sample sizes. They must therefore be used with a great deal
of caution. They are presented here as is because collecting

large samples of young would seriously hinder my continuing

%0

aooama
ee mmaaaemea eao ae6-ma one -HH no aoeoseaaewao one maeaaeeoo no mo.ova - e
aoocwa pm

 

 

 

coausnaaumflo Hmuou map mafiaamuno mo moo.ovq momma .mxmq omom Bong some n o
mmnflaumon 0H0 amcuNH com IHH oumowocw mommnucoawm CH maonasn u h
mHmScH>aonH He mo 03» ca ummwah one mo mnflammaum coondononn now: a m
Ammo um any :a Amav rm on m c.6Heeoa
Area mm are an Ana ma lav m . sateen
Arno em lav m has me aims 6 one:
Hmuoa mhmaw unmaq mumaomanopaH Hmooamno
omega aoaoo New

mmaxomno mcHHpmoz ca :oflusnfinumfim ommnm aoaoo
w magma

%1

studies of grackle population structure and recruitment in
central Michigan. I will greatly appreciate any feedback

from others who attempt to use these criteria.

SUMMARY'

Social behavior of Common Grackles was studied in three
locations in central Michigan with the purpose of evaluating
whether or not this bird is nest-site limited, whether or not
breeding aggregations of the birds are formed because the
birds are actively colonial or simply distributed as a func-
tion of available nest sites, and to determine changes in a
population censused 20 years previously. In addition, a
study was conducted to determine criteria for sexing nestlings
of known age.

Breeding Common Grackles are actively colonial rather
than passively distributed as a function of available nest
sites. Breeding display between males and females in colony
groups is highly contagious. It stimulates synchronized
initiation of nest building by the females in terms of nesting
date and nest-site choice as well. A brief discussion of the
potential effect of age on breeding time is made.

A census of the breeding population in East Lansing,
Michigan indicated that nest sites are not now limiting
grackle populations as a whole, and that the population in
the area has increased six-fold in the past 20 years.

Tentative criteria are presented for sexing 11- and 12-
day old nestlings on the basis of tarsus and culmen measure-

ments. The potential significance of a previously unrecorded

%2

color phase series among the nestlings is discussed.

LITERATURE CITED

LITERATURE CITED

American Ornithologists' Union. 1957. Checklist of North
American birds, 5th ed. A.O.U., Baltimore, 697 pp.

Attwater, H. P. 1892. List of birds observed in the vicinity
of San Antonio, Bexas County, Texas. Auk, 9: 229-238.

Austin, 0. L. 1951. Group adherence in the common tern. Bird
Band., 22: 1—16.

Axtell, H. H. 1955. Bronzed grackle nesting on beaver lodge.
Auk, 72: 8%.

Bent, A. C. 1958. Life histories of North American blackbirds,
oréfiles, tanagers, and allies. U. 8. Nat. Mus. Bull., 211:
1" 90

Cahn, A. R. 1920. Bird notes from Itasca County, MHnnesota.
Wilson Bull., 32: 103-122.

Cameron, E. S. 1907. The birds of Custer and Dawson Counties,
Mentana. Auk, 2%: 389-%06.

Chapman, F. M. 1892. A preliminary study of the grackles of
fihe subgenus Qgiscalgs. Bull. Amer. Mus. Nat. Hist.,
: 1-20.

Chapman, F. M. 1928. The nesting habits of wagler's oropendola

(Zarhynchus wagleri) on Barro Colorado Island. Bull. Amer.
Mus. Nat. HiSto, : 123-1660

Coulson, J. C. and E. White. 1956. A study of colonies of the
kittiwake Rissa tridactyla (L.). Ibis, 98: 63-79.

Coulson,_J. C. and E. White. 1960. Effect of age and density
of breeding birds on the time of breeding of the kitti-

wake, Rissa tridagtyla. Ibis, 102: 71-86.

Crook, J. H. 1961. The basis of flock organization in birds.
pp. 125-1%9. 12 w. H. Thorpe and 0. L. Zangwill, eds.,
"Current problems in animal behavior," Cambridge Univ.
Press, Cambridge.

Darling, F. F. 1938. Bird-flocks and the breeding cycle: A
contribution to the study of avian sociality. Cambridge
Univ. Press, Cambridge, 115 pp.

%3

in,

Dawson, W. L. 1923. The birds of California. Vol. 1, South
Moulton, Los Angeles, 522 pp.

Dennis, J. V. 19%8. Observations on the orchard oriole in
the lower Mississippi delta. Bird Band., 19: 12-21.

Dunham, D. W. 19 1. Interspecific cacique colonies in Surinam.
Auk, 88: 17 .

Eyer, L. E. 195%. A life history study of the bronzed grackle

Qgiscalus guiscula versicolor Vieillot. Unpubl. Ph.D.
dissertation, Michigan State Univ., 158 pp.

Ficken, M. S. 196%. Nest-site selection in the American red-
start. Wilson Bull., 76: 189-190.

Ficken, R. W. 1963. Courtship and agonistic behavior of the
common grackle Qgiscalus ggiscgla. Auk, 80: 52-72.

Cibbs, M. 1903. The grackles change in nesting habit. Bull.
MiCho ornitho Clubo, "1’: 57’s 0

Colsan, L., and E. C. Holt. 191%. Birds of Autauga and
Mentgomery Counties, Alabama. Auk, 31: 212-235.

Haartman, L. von. 1956. Territory in the Pied Flycatcher
Mgscicapa hypoleuca. Ibis, 98: %61-%75.

Holcomb, L., and C. Twiest. 1968. Ecological factors govern-
ing nfist building in redwinged blackbirds. Bird Band.,
39: l -230

Horn, H. S. 1968. The adaptive significance of colonial

nesting in the Brewer's blackbird (Egphaggs gyanocepga-
13g). Ecology, %9: 682-69%.

Horn, H. S. 1970. Social behavior of nesting Brewer's black-
birds. Condor, 72: 15-23.

Lehrman, D. S. 1959. Hormonal responses to external stimuli
in birds. Ibis, 101: %78-%96.

Lloyd K. 19%3. An unusual nest of the bronzed grackle. Wilson
Bull., 55: 56.

MeIlhenny, E. A. 1937. Life History of the boat-tailed grackle
in Louisiana. Auk, 5%: 27%-295.

Meyerreicks, A. J. 1958. Magpie-jay robs Mexican cacique
nests. Condor, 60: 67.

Nero, R. W. 1960. Identification, redwinged blackbird. In
U. S. Bureau of Sport Fisheries and Wildlife. 1961,
"Bird-Banding Manual", mimeo, 88 pp.

%5

Orians, C. H. 1961a. The ecology of blackbird (Agelaius)
social systems. Ecol. Monogr., 31: 285-312.

Orians, C. H. 1961b. Social stimulation within blackbirds
colonies. Condor, 63: 330-337.

Peterson, A., and H. Young. 1950. A nesting study of the
bronzed grackle. Auk, 67: %66-%76.

Schafer, E. 1957. Les conotos: etude comparative de

Esarocolius an ustifrons et P. decumanus. Bonn. Zool.
Beitr.,sonderheft 1957: 1-1%7.

Schaller, C. B., and J. T. Emlen, Jr. 1961. The development
of visual discrimination patters in the crouching re-
actions of nestling grackles. Auk, 78: 125-137.

Selander, R. K. 1965. On mating systems and sexual selection.
Amer. Nat., 99: 129-1%1.

Selander, R. K., and D. R. Ciller, 1961. Analysis of sympatry
3f great-tailed and boat-tailed grackles. Condor, 63: 29-
6.

Selander, R. K., and R. J. Hauser, 1965. Conadal and behavioral
cycles in the great-tailed grackle. Condor, 67: 157-182.

Skutch, A. F. 195%. Life histories of Central American birds:
Fringiligdae through Coerebidae. Pacific Coast Avidauna,
31: l- .

Synder, L. L. 1937. Some measurements and observations from
bronzed grackles. Can. Field Nat., 51: 37-39.

Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman,
San Francisco, 776 pp.

Stockard, C. R. 1905. Nesting habits of the birds in Mississip-
pi. Auk, 22: 273-285.

Thomas, R. H. 19%6. An orchard oriole colony in Arkansas.
Bird Band., 17: 161-168.

Tracy, H. C. 1896. The bronzed grackle, a change in nesting
sites. Wilson Bull., 8: %.

Tutor, B. M. 1962. Nesting studies of the boat-tailed grackle.
Auk’ 79: 76-81}.

Williams, L. 1952. Breeding behavior of the Brewer blackbird.
Condor, 5%: 32-%7.

Willson,M. F., R. D. St. John, R. J. Lederer, and S. J. Muzos.
1971. Clutch size in grackles. Bird Band., %2: 28-35.

%6

Wood, J. 1969. A bird bander's guide to determination of
age and sex of selected species. Coll. of Agriculture,
Penn. State Univ., College Park, 181 pp.

Youngworth, W. 1932. Notes on the nesting of the bronzed
grackle and Say's phoebe. Wilson Bull., %%4 %l.

"11111111111111.1171“