VARIATION IN" MORNING AVVAKEIQIW
TIME 05 BIRDS IN sow-CENTRAL MICHIGAN

TIMI: I3: fha Dogs-o. a5 M. 5.
MICHIGAN STATE UNIVERSITY

George F. Féslar
I956

THESIS

VARIATION IN MORNING AWAKENING TIME OF BIRDS

IN SOUTH-CENTRAL MICHIGAN

By
George P. {isler

AN ABSTRACT OF

A thesis submitted in partial fulfillment of the
requirements for the degree of Master of
Science in Michigan State University

Department of Zoology

East Lansing, Michigan
fl/ 1956 /7

—L4

' 5C.

CK

THESIS ABSTRACT

The purpose of this study was to determine at what times species
of birds awakened in the morning by using the first songs and calls
as the criterion of awakening time. Terms used such as night song,
true night, civil and astronomical twilight, and awakening time were
defined as to their use in this paper.

Data were collected during a one year period with a total of 120
morning observations used for comparisons. The awakening times of 12
species of birds were recorded. The daily awakening times, designated
by the number of minutes before sunrise, were averaged for each of
four arbitrary periods, and significant differences between these means
were used to determine the relationships both for one species over the
year and also relationships of means between the various species in each
period. Graphs included showed the daily awakening time for each species
through the year and the significant differences between the means both
within and between species.

The major conclusions reached in this study were that there was
a tendency for the awakening time curves of the Species to approxi-
mate the sunrise and civil twilight curves during most of the year
but with systematic deviations of the awakening time curves at cer-
tain seasons. Birds awaken earliest in the spring of the year in

relation to sunrise. All Species do not rise earliest at the summer

solstice. The order of awakening of the various Species through the

year was not static but changed through the seasons. Most species.

awakened at very similar times during the fall and winter'lonths,
but, when spring arrived, the earliest rising species became earlier

and remained so until late summer or falls

VARIATION IN MORNING AWAKENING TIME OF BIRDS

IN SOUTH-CENTRAL MICHIGAN

By
George F. Fisler

A thesis submitted in partial fulfillment of the
requirements for the degree of Master of
Science in Michigan State University

Department of Zoology

East Lansing, Michigan
1956

TABLE OF CONTENTS

ACIQ‘IOWLEDGMENTS
INTRomflIONOOIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOIOOOO0.0.0.0.... l
MHODSCOOCOOOOO0.0.0.0000...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0 5

DEFINITIONS OF TERMSOOOOOOOO0.000.0.0.0...I...OOOOOOOOOOOOOOCOOOOOO 8

MULTS’..C.............................O.................0........13
Treatment Of Data...............................................13
SpfiCies Accounts.............................................ooo16

umml.C........C................0.0.............0.........16
song Sparm...‘.......O0.....0....0.................0.......17
Crow................................o........................18
Whiteebreasted NuthatCheeeeeeeeeeeOeeeeeeeeeeeeeeeeoeeeeeeee019
Flicker..............o.......................................20
Rabin........................................................21
Phoebe.......................................................2h
Red-BYCd Vireo............................................o..2h
WOOd Pme.....................................C.............2S
Catbird......................................................27
Ian“ warbler......0...’...............0.0..................27
Rough‘W1n86d swallOWeeeeeeoeo.cooeeeeeeeeeeooeeoooeeeeeeeeee028

DISCUSSIONeeeeeeeeoeoeeeeeeeeeeeeeeeeeeoeeeeeeeeeeeeeeeeeeeeeeeeeee29
The Alakening Time Curve.....................................ooo29
Order Of Awakening.............................................o30
Changes in the Order Of Awakening.............................o033
Ranking of Species as Early or Late Risers................;.....35

M AND CONCLUSIONSOOOOO0.0000000000000000000000000.0.0.000000039

APPENDHOOOOOOOOCOOOOOOOOOO0.0.0.0...IOO00.0.00...COOCOOOOOOOOOOOOOM
I TablBSOOOO0.00000000000000000000000000000000000000000000000000hZ

II GraphBOOOAOOOOOOOOOOOIOOO.0.OOOOCOOOOOOOOOOOOOOOO0.00.00.00.0057

HTERAWRE CITEDOOOCOOOOOOOOO.0......OOOOOOOOOOOOOOOOOOOOOOO0......70

0 e e e
O 0

e a e .
I O O O

I u 6
I I a e
e I I O

1.

O
D O C O
I I .
O O O I
D C I I
I O O
I Q I D

ACIWOWLEIIIMENTS

The author wishes to express his deep appreciation to Dr. T.
Iayne Porter for his helpful criticisms and guidance during this
study. Dr. Porter's critical reading of this manuscript has im-
proved it measurably, and, without his thoughtful comments on
the paper, many improvements in the style of writing would not
have been made.

The author is also indebted to Dr. Don I. Hayne for advice in
the statistical handling of the data and for reading portions of
the manuscript.

A further note of thanks is extended to Dr. George J. Wallace
under whose tutelage this study was began and whose helpful criti-
cisms early in the project were much appreciated.

The author is extremely indebted to Dr. Jesse ll. Shaver (re—
tired) of George Peabody College for Teachers in Nashville, Tennessee,
for the loan of the thesis by Enily B. Walker. Without his kind

soaperation, personal study of this thesis would have been impossible.

INTRODUCTION I

The reasonably constant sequence of awakening times of birds has
beennwell known.fbrwmany'years. lany authors have commented upon the
regularity with which birds first sing in the morning, each species
apparently having its own particular awakening time. Rice (1910) re-
ferred to this phenomenon as the "bird clock". .According to Ialker
(1928), Jeny in 18h6 compiled.a list of the sequence of the awakening
times for the birds in his area. Other workers cited by Walker (1928)
included Altun (1868), Godfrey (1897), and Gynsell (1903) each of
who. kept lists of the first rising times of various species. The
first serious study of this nature was undertaken by'Iright (1912,
1913) who worked during a period of several.years in the mountains
of New Hampshire. Allen (1913) published some additions and correc—
tions to Iright's (1913) work and added some more of his own records
in 1915 and in 1922. Allard (1930) listed rising times of about 11;
species and attempted.to evaluate his records for the purpose of'finde
ing why the birds arose when.they did. His work was the first serious
effort in this country to establish reasons for’a definite awakening
thee for the many species studied. Also, Allard's (1930) paper, Cl
far as this author could determine, is the only one concerned with
with the problem.of awakening time in which several species of birds
‘were systematically studied. Many authors have listed their impres-
sions as to the awakening times of various species, while other‘authors

have done serious studies and recorded the actual awakening times.

The best known of these is the work of Nico (19h3) on the Song Spar-
row. Others include Craig (1926, 1933. 1910). Taylor (1929). men
(1937).. Incense: (1950). m (1951). locate (1951), and King (1955:).
There are now many papers in the literature on the roosting flights
of several species, particularly herons (Seibert, 1951), and several
concerned with Robin and blackbird rcosts (Abel, 1911;, Nice, 1935,
Howell, 1910, for example) . However, few workers have conducted their
awakening time studies over a full calendar year and noted the changes
which occur in the rising time. Allard (1930) was the first to pub-
lish a report of awakening times based on an extended period of time
(January 15 to September 1). .Previously, an unpublished thesis by
Ialker (1928) presented a study of the problem from October to July.
line's (19113) work contains records taken at all seasons of the year
and over several years' duration. A

law authors, using a limited number of records, attempted to em-
plain'why birds rise at a particular tine of morning. However, this
author believes that it is necesaary to have a large series of records
covering all seasons of the year in order to draw reliable conclusions.
There are difficulties involved in‘ detemining what the awakening time
may be. No attupt will be made in this paper to offer conclusions
concerning the reasons for the species arising at the times they did.
The results obtained in this study will be presented concerning only
the awakening times observed over the seasons.

The study area was a beech-maple-oak woodlot near the campus of

Michigan State University. It was bounded on the west side by the

MAP 0F sruor AREA

RED CEDAR

 

WOODLAflD

 

 

 

\\\\\\s. .

 

 

 

or

 

3% 2. ch

we u<dw

hnnx
t0Hh<><mnbO

 
    

scscm :uxua

Whtiuxorxat

$

chances; II:

ozmuwg

 

 

Red Cedar River, on the north and east by well-traveled roads, and on
the south by open fields and a university housing project. The obser-
vation post was located on the south bank of the river in a clearing
out through the woods. Therefore, the observer had within his range
of hearing an open, grassy field, an edge habitat along the woods on
both sides of the cleared field, the woods habitat, and the river edge
habitat, including the woods and the cpen field. The accompanying
map illustrates the area. I

The study was begun on January 11, 1955, and terminated at the
end of February, 1956. The records of January, 1955. were not used.
This period was used to develop techniques which were used later dur-
ing the rest of the study. The records of 120 morning field trips

are presented in this paper.

KETHODS ,

The materials required far this type cf? a study were few and
simple. A clipboard with sheets of paper preperly prepared for re-
cording the data, a wristwatch, pencils, and a pair of 6:30 binocu-
lars were used. A Isston Sight-light meter which measured light in—
tensity from zero to 75 foot-candles was taken on every field trip,
and the light intensity was recorded for possible further correlation
nth the awakening times. This meter was not accurate enough for
small readings, but it did allow recording of the light intensity up
to twenty minutes before sunrise on clear mornings. Temperature,
general notes on the sky conditions, wind velocity and wind direction
were the only weather taken on the site. The wind velocity was estimat-
ed by use of the Beaufort scale. Other weather data were taken from
the local monthly climatological data sheets issued by the United
States Ieather Bureau in East Lansing.

Data were recorded in the following manner. The author generally
reached the place of observation about one—half hour before the birds
began to sing. Ability to predict the time when the birds would begin
to sing became quite accurate after a few mornings of study. This
one-half hour time interval was adjusted as the sunrise time varied
through the year. However, the awakening time curve of the birds was
not in strict accordance with the sunrise curve through the year so
the adjustment of the tims interval had to follow the awakening time

curve 'ch the birds also. Recognising these conditions, the observer

5

was able to maintain the one-half hour period between his arrival in
the area and the first song of the earliest species. The possibility
d the records being biased by the author was eliainated by the selec-
tion of the criterion of awakening time.

a record of the number of individuals of each species singing or
calling was made during each one nimte interval. Observations con-
tinued fro- the time of arrival until sunrise. The observer ruained
in the area after sunrise to record any late risers. In compiling
the data, all records of rising time exceeding ten minutes after sun-
rise were discarded in order to make all records constant. The ten
nuts period represented the minimal time after sunrise in which the
author renained in the field on any given day during the year. The
records of each day were transcribed on to permanent charts and graphs
after returning from the field trip.

some difficulties in recording the awakening ties were encounter-
ed due to extraneous noises. First, the noise of from one to four
nimtes duration produced by trains on a near-by railroad sometimes
rendered hearing in the vicinity quite poor. A loud, shrill stag such
as that of the Cardinal could be heard over the din, but the much
weaker vocalization of the Tallow Warbler was difficult to hear. This
factor will have to be overlooked because determination of the exact
time «t the fifet call or song was impossible. The nacdm error
possible due to the noise of trains in recording the data was four

minutes. The noise from trains did not interrupt the observations

nere than 20 times during the year. The second disturbance was the

noise of auto traffic from the adjacent roads. The observer believes

that this disturbance was insufficient to cause any appreciable change
in the ability to hear the awakening songs or calls. a third factor
as the simultaneous and loud singing of all species. The amount of
noise made by the noisier birds, especially during the spring song
period, was sufficient to obscure the first songs of the less vocifer—
ous species. The true error here cannot be determined. wright (1913)
and Saunders (1929) also discussed the problem of noisy birds obscur-
ing the songs of quieter species. a fourth extraneous factor may be
the bias “of the observer in listening more attentively for a certain
species to sing at a time which experience had taught him to expect
it. This is a problems inherent in work of this nature and care must

be taken to avoid it.

DWINITIONS OF TERMS

A number of terms used in this paper require definitions. The
first is the tens ”awakening time". Selection of a definite time as
that time when a bird awakens is almost impossible. Therefore, some
criterion must be designated which will more or less indicate the
time when the bird becomes truly awake and active. Allard (1930)
stated that it was impossible to tell when the first light reached a
bird whose head was under its wing thus awakening the bird (seeming
that light is the critical factor in determining awakening time).
Therefore, he chose as a standard any morning song heard after astro-
nomical twilight as the true awakening time. However, even Allard
(1930) called this method only a "convenient denarkation". This
author believes that Allard's (1930) method did not recognize that
night songs are not limited to the period of true night. There was
no indication in my study that birds recognise the termination of
true night as the beginning of day by starting their activity at
astronomical twilight. Rather, this study indicated that the songs
uttered between astronomical twilight and the beginning of the dawn *
chorus were also night songs. The songs of true night and those
uttered between astronomical twilight and the beginning 3 the dawn
chorus are all sung in a like manner ( a few, scattered, single utter-
ances fro- one or more birds of the same species) and no distinction
should be made between them. They are night songs. Thus 'Allard's

(1930) awakening time criterion was dealed unsuitable for this study.
8

Wright (1913) stated that awakening time was that time when more
than one bird of the same species sang at the same time. However, a
bird that sings during the night will often be quickly followed by
the vocalization of one or more individuals of the same species, the
first bird apparently serving as a stimulus for the succeeding singers
(Saunders, 1929). After the completion of this singing, these birds
will not be heard again for possibly half an hour or more. Wright's
(1913) criterion of awakening time did not recognize these utterances
between astronomical twilight and the beginning of the dawn chorus as
night songs. Craig (19143) stated that the awakening song of one bird
did not influence the time of the awakening song of another. Saunders
(1929) stated that birds will begin singing continuously within a few
minutes after awakening, and Nice (19h3) expressed this same Opinion.
This writer considers these early sporadic outbursts between astronom-
ical twilight and the beginning of the dawn chorus to be night songs
rather than awakening songs. Allen (1913) stated that the true awaken-
ing time occurred when more than one individual of the same species
was singing regularly. This method of Allen's (1913) supplemented by
Saunders' (1929) views that birds will sing continuously shortly after
awakening constitutes the awakening time criterion followed in this
study. However, this criterion will hold only during the song season.
Later in the year when the birds become more or less silent as far as
song is concerned, it is necessary to substitute the calls of the birds
for the full song in the awakening time definition. Also, this defini-
tion is not acceptable for such irregularly singing species as the

Crested Flycatcher (M crinitus) and the woodpeckers. One can

acme that the first utterance cf the individual is its awakening
time fc these birds. However, the probable awakening time of the
flickers is known because the birds roosted near the observation post
ef the author, and he could see the birds when they emerged from their
roosting holes. But it must be remembered that the true awakening
time is not known for any species discussed in this paper. The first
songs and calls are only an indication of the time that the bird de-
finitely began activity in the morning.

The phenomenon of night song was recognised by Fright (1913).
but he did not believe that it played a part in the time of awakening.
Allen (1913) also recognized night song and made corrections for its
tandem. on the awakening time. Allard (1930) wrote that “night
song is not a common and regular behavior". There is some diagree-
msnt between this author and Allard (1930) on the amount of night
song because of the difference in methods of choosing what constitutes
the true awakening time. Even using Allard's (1930) criterion of a-
wakendng time this author feund that, for the birds under his surveil-
lance, night song was of common and regular occurrence. Other obser-
vers have published on the proneness of some species to sing at night.
The frequency of night songs apparently depends on the stage of the
nesting cycle and the phase of the moon. There is much night singing
during the peak of the song period in the spring. There is little
during the fall revival of singing and little or none during the mid-
sumer and winter periods. More night singing was evident when the
moon was full or when the sky was very clear and thus brighter than

it was on less clear but moonless nights. Thus the definition of

night song derived for use in this work was as follows: night song
{includes those individual songs which a bird(s) will sing during the
time between the end of the song period of the evening and the com-
ment of regular and continuous singing the next morning.

The definitions of true night, civil twilight, nautical twilight,
and astronomical twilight were taken from the Smithsonian Heteorolo-
gical Tables (1951). True night is the time delimited by astronomical
twilight when the sky is completely dark. True night begins in the
evening when the sun is 18° below the horizon and ends in the morning
when the rising sun again reaches 18° below the horizon. Civil twi-
light is defined as the time after sunset and before sunrise when the
sun is six degrees below the horison. Originally, it was "the time
after sunset or before sunrise during which, on clear days, there is
W light for ordinary outdoor occupations". Similarly, nautical
twilight is defined as that time after sunset and before sunrise when
the sun is 12° below the horizon. Astronomical twilight is that time
when the sun is 18° below the horizon, and its limits form the bounda-
ries of true night.

Previous workers (Allen, 1913, Wright, 1913, Allard, 1930) have
used the tern |'awakening time" rather freely. The connotation of the
term is that the songs designated as awakening songs indicated the
actual rising time of the bird. However, these authors presumably
meant these utterances were the first songs of the new day. Allard
(1930) recognized that the awakening time of a bird was almost ilpos-

sible to determine accurately as was stated earlier. The bird may be

12

awake for some time before it first sings. Therefore, it is incor-
rect to call the first songs the "awakening time". walker (1928)
tamed this first song the "awakening song". Nice (1910) used this
term in her work on Song Sparrows. Both workers imply that this is the
birds' first song of the morning but that it is not necessarily the
time at which they awakened. However, for lack of better terms, this
author will continue to use both terms. "Awakening song" will indi-
cate the first songs of the morning and "awakening tine" will be that

time when the "awakening song" was uttered.

RESULTS

Treatment of Data

During the year of study, the author obtained complete records
for approximately 25 species. Of these, 12 were selected for dis-
cussion. They are as follows: Cardinal (Richmondsna cardinalis),

 

Gatbird (Dumetella carolinensis), Crow (m bracMchos ,
Flicker (Colagges auratus) , Phoebe (Sazgrnis M , Red-eyed Vireo
(11:93 olivaceus), Robin (Tur___£l_u_s_ mifltorius), Rough-winged Swallow
(StelgidoPteryx ruficollis), Song Sparrow (Helogpisa melodia), White-
breasted Nuthatch ('S__it_t_a_ carolinensis), Wood Pewee (ContoEus 23.1139.) ,
and Yellow Warbler (Dendroica petechia . In Table I will be found a
list of these birds giving their status in the woodlot and migration
dates where applicable.

In discussing a subject as variable as the awakening time of
birds over an entire yearly cycle, grouping of the data in some mean-
ingful manner is necessary. Since the awakening time apparently varied
with the seasons, the author presumed, erroneously, that seasonal
grouping of the data would be wisest. However, the change of the a-
wakening time over the year did not follow the change in the seasons.
Thus the data had to be grouped in another manner in order to illustrate
the change in the awakening time. The four periods selected were de-

signed to illustrate the change in the awakening time through the i

13

11;

year. The periods were derived by careful inspection of graphs con-
structed for each species showing the awakening times , the sunrise
curve, and the civil twilight curve (Grapna I through VI). The data
for each species was computed for these periods. All species could
not be expected to exactly fit these four periods in their awakening
time trends. There were three species where the defined periods did
not adequately illustrate the true awakening time trend. These ex-
ceptions will be discussed later in the species accounts. Although
not completely accurate in showing the awakening time trends, group-
ing of the data for each species into the same four periods was nec-
essary in order to permit comparison between “species statistically.
Table II lists these periods with the inclusive dates. Hereafter,
references to these periods, where practical, will be made by number
rather than by dates. .

Each daily awakening time or each average awakening time for an
period is expressed in the number of minutes before sunrise that the
song or call was heard. Previous workers (Allard, 1930, Nice, 19143)
have objected to use of the sunrise curve as the basis for comparison
stating that the awakening time curve more closely approximates the
civil. twilight curve. However, by inspection of graphs I through VI,
it can be seen that the awakening time curve of birds in this study
did not follow rigidly either the civil twilight or the sunrise curves
but deviated from both these curves during the year. Therefore, the
data will be treated as "minutes before sunrise", a term which this
author believes has more meaning and is more quickly understood than

the term "deviation from civil twilight".

15

The daily awakening times for a period were averaged resulting
in only one figure representing the awakening time for each species
in a period. Previously, averaging awakening times has been objected
to by Allard (1930) because it tends to disguise trends toward earlier
or later awakening times. While this may be true, for purposes of com-
parison, reducing the volume of data by the use of averages of daily
awakening times is necessary in order to. more easily compare the re-
sults both within and between species. This was the author's reason
for selecting the four periods as they are set up. The periods show
adequately the trends in the time of the awakening song for most species.

This author believes that the objection by others to averaging a-
wakening time data was initiated because pioneer workers in this field
(Iright, 1913, Allen, 1913, 1915) averaged their results indiscrimi-
nately without regard to changes in the awakening time over the season.
These workers than stated that this average represented the awakening
time of the species. when Allen (1913, 1915) compared his results
with Iright's (1913). he found there was some disagreement, especially
over the awakening order of the Robin and the Song Sparrow. Allard
(1930) explained the disagreements over awakening times by assuming
that the difference lay in the "fact‘I that species arise at the same
times in minutes before sunrise on am given parallel of latitude but
would not be the same on different parallels of latitude. Therefore,
Iright's (1913) New Kanpshire awakening time averages would naturally
differ from Allen's (1913 , 1915) Boston averages. However, this ob-
server believes that the pertinent difference in the research of Allen

and Fright is in the fact that Allen's averages represent a longer

16

period of the year, and, also, his interpretation of awakening tine was
different. Thus it seems desirable for workers in this field to define
their use of an awakening time standard. Dates of the period of the
year covered and a graph which shows the change in the time of the first
awakening song over that period should supplement the averages employ-
ed. Averages of daily awakening times that have been computed from
carefully selected periods are reliable and do show the changes in the
awakening time during the year.

The standard deviation and the standard error were computed for
each species during each of the four periods that it was present in
the woodlot. These will be found in Tables III through XIV. The
graphical method (Dice and Leraas, 1936) was used to test the signi-
ficant differences between means. This graphical test was supplasent-
ed by the use of "Student's” 't" test where necessary. Throughout
the study, the 95% confidence level is considered significant. Graphs
VII through 11 illustrate these data. No attenpt was made to distin-
guish the difference (if any) in the awakening times of males and fe-
males. Also, there was no distinction made between bright and dark or
wall and cold mornings or any other variables in the weather. The re-
sults below are discussed for each species, and then the different
species are compared. Wherever the negative sign (-) precedes a num-
her, the number of minutes before sunrise is indicated. Similarly, a

plus (4-) indicates the number of minutes after sunrise.

Species Accounts

Cardinal (Richmondena cardinalis). This species was a permanent :w -

17

resident of the woodlot. The usual number heard singing during the
height of the song season was five. During the winter, three or four
birds could be heard caning each morning. Some of these data in-
cluded the singing of the female as well as that of the male. Fueles
were observed singing on several mornings. Laskey (19bit) reported
that the female Cardinal frequently song. This same author also stat-
ed that the Cardinal sang considerably at night. This observer found
night singing to be more irregular and somewhat less in number of
songs than reported by Iaskey (19M).

. The mean awakening time of the Cardinal became progressively
laterubeginning in the spring and continuing through the winter as
can be readily seen from Table III and Graph VII. From an early aver-
age in period I of 440.6, the average dropped to a low of -23.2 in per
iod IV. This MP in the awakening time mean was not steady but re-
gressed by "jumps". From the graph, one can see that periods I and
II overlap as do periods III and IV also. Neither of these were sig-
nificantly different when the means were tested. Therefore, the a-
wakening time of the Cardinal can be divided into two periods of the
year, a spring-summer early awakening and a fall-winter late awaken-
ing. The means of the four periods clearly illustrate that this spee .1
cies began its song activity at different times before sunrise during

the year.

Song Sparrow (Helogisa melodia). The number of Song Sparrows heard
singing or calling varied from ten in the spring to two in the winter.

The two birds that remained in the area during the winter months

18

habitually roosted near the observation post of the author. Thus
establiahnent of their awakening time was easy and quite accurate.
This species showed a decrease in the mean awakening time from
period I through period IV as did the Cardinal. The means of the
periods well illustrate the change in the awakening time over the
year in relation to sunrise. From Graph VII, the difference between
the means of the periods can be seen to be highly significant; that
is, there is a real difference between all the means of the four
periods (also see Table IV). The earliest awakening time was in the
spring of the year, and the latest was during the winter period (IV).
As with the Cardinal, the division of the data into the four periods
did not include any major fluctuations in the time of the awakening
song or call. To clarify Wright's (1913) and Allen's (1913) discrep-
ancies in means and in order of awakening, further discussion of the
awakening time of the Song Sparrow will be deferred until the dis-

cussion of the Robin.

Crew (9239.! bracmwos). Crows were observed in the woodlot all
during the year. The number of individuals varied from five or six to
30. The earliest awakening time of the Grow in relation to sunrise
occurred in the spring period. The means for the other three periods
were somewhat later. All the means of these latter periods fall close—
ly together and were not significantly different (Graph VII and Table
V). Therefore, the awakening time periods of the Crow could be divided
into two parts. The first part consists of the spring period only,

and the second encompasses the remainder of the year. This species

19

was the only species studied by the author that had such an arrange-
ment of its awakening time periods.

The awakening time periods for the Crow may not be completely
reliable, because the species did not roost in the woodlot at any
time during the study. The records were from birds flying overhead.
However, the records are more accurate than this discrepancy may lead
one to believe. The Crows always came from the same direction (south-
west) and were invariably heard calling before they were seen. The
roost was found to be only a short distance away. Therefore, ‘in
light of these factors, the awakening times as given are thought to
be quite accurate. There is some doubt in the author's mind as to
whether Crows call very soon after beginning their day's activities
or wait some time before vocalising. Zia-er, quoted by Craig (19113).
was of this same Opinion. Thus the author believes that the data as
recorded for the Crow is sufficiently accurate to warrant including

in 131118 paper.

Ihite—breasted Nuthatch (m carolinensis). At least twc individ-
uals of this species were heard and recorded throughout the study,

and at times as many as five were present. The graph for this species
(Graph VIII) resembles that of the Song Sparrow except that the means
of the four periods are closer together. (See Table VI for the mean,
standard deviation, and the standard error computed for the Ihite-
breasted Nuthatch.) This species had an awakening time mean that be-
came later and later over the seasons beginning with period I, but the

differences between periods were not as great as the means in the other

20

species. Period IV was the only period that differed significantly
fro. all other periods. Thus the White-breasted Nuthatch definitely
awakened later in the winter than at any other time of the year.
However, during the other three periods, no distinction could be
made which showed this species to arise at different times during
these three periods. By referring to Graph :VIII, the general trend
of the means of these periods to become later as the year progressed
can be seen. The only other significant difference that could be
shown was between periods I and III , thus strengthening the state-
aent that the bird arose later in relation to sunrise as the year

progressed.

Flicker (Colaptes auratus). The Flicker was present in the woodlot
during all four periods of the year. Only one bird was there during
period IV, however, and no recordings of the awakening time were
taken during that period. Ole possible explanation for the lack of
observations of this species in period IV is that the Flicker may be
an extremely late riser during that period. Thus the awakening time
for period IV may not have been recorded because the observer had left
the am before the Flicker awakened. Shaman (1910) stated that the
flicker generally arose " soon after sunrise '. The bird was seen in
the woodlot in period IV! during the afternoon only. Another possible
explanation is that this species is almost silent during period IV in
this regionso that the chances of hearing its awakening calls (if it
gave one) are negligible. The roosting place of this bird was not

located during period IV, nor was it ever seen entering or leaving

21

the holes that were occupied during the rest of the year. Quite pos-
sibly the Flicker did not roost in the area. This would account for
the lack of arv morning records for this species during period IV.

This species is the first example to be discussed where the use
of arbitrary periods in computing the mean awakening time was mis-
leading in that it partially disguised the awakening. time trend. while
the mean awakening times of the Flicker were considerably different,
there was no significant difference ‘of the means between any of the
periods (Graph VIII and Table VII). This lack of significance in-
dicated that the Flicker arose at about the same time during all three

periods. However, this is not entirely true. Study of Graph II of
the awakening time curve and the sunrise curve shows that the division

between periods I and II split the Flicker records at the height of
the earliest awakening time. Careful study of this graph (II) will
show that the Flicker actually arose earlier and earlier from about
larch 22 until its earliest awakening time was reached on May 1;. after
this latter date, there was a decline in the earliness of awakening.
The birds arose very irregularly through June to December and never

again approached the early awakening time of spring.

Robin (T_u_1@_u_s_ migratorius). The number of Robins heard and» recorded
each morning varied but was usually about 12 except late in period III
when there were fewer birds present. This species followed the pat-
tern & awakening time means previously described for the Cardinal,,
except that there were no data recorded during period IV because the

species was absent from the woodlot during that period.

22

The means of periods I and II are not significantly different,
but both differ greatly from period III. The difference in awaken-
ing time means of periods I and II from the mean of period III is
apparent from Graph VIII and Table VIII. The species exhibited the
greatest difference of the means between periods of any of the species
studied. The Robin arose at approximately the same time during the
spring- summer period, but the birds no longer arose as early with the
subsidence of the song season and the approach of fall. The mean
awakening time became very close to that of the other species. How-
ever, Vaurie (191:6) in Pennsylvannia stated that the Robin arose ear-
liest during mid and late manner with only the Mourning Dove
(Zenaidura macroura) rising earlier. Vaurie's statements were not
corroborated in this study.

According to Bent (19h9) male Robins roost in large flocks during
the nesting season rather than individually near their nests. This
kind of activity was not observed at any tine during this study. Each
morning that records were made the songs of the various individuals
always came from the different nesting areas. If these male birds
did roost at some site other than the woodlot, they apparently flew
into the area long before they were in song. The supposition that
they flew into the area before singing is very doubtful, because
little movement of birds was observed in the early morning before
cinema began.

The Robin is a species that did not sing much at night. The

Song Sparrow indulged in much night song (Nice, 191:3). and thus by

23

using'lright's (1913) criterion of awakening time one would find

that the Song Sparrow arose earlier than the Robin. This author
believes that this may be the reason for the disagreement between
wright (1915) and Allen (1913) concerning the order of awakening of
the Robin and the Song Sparrow. Nice (1910) was of the Opinion that
both'lright's (1913) and illard's (1930) records included night songs
of the Song Sparrow. ‘Wright's (1913) average awakening time for the
Song Sparrow was -73 while this author's was -33. Allen (1915) gave
-29 3/8 as an average for the Song Sparrow. ‘lright's (1913) average
for the Robin was ~6h, and.Allen's (1913) was -63. (Allen arrived at
a.-53 2/3 average for the Rebin in 1915.) This author found the aver-
age awakening time for a period comparable to the data of these authors
to be -62 for the Robin. The differences between the observations of
these authors in the awakening times of the Song Sparrow and the Robin
and also the order of awakening of the two species seems to be the
discrepancies in the averages for the Song Sparrow. The average for
the Robinnwas constant between the three sets of data. Allard 11930)
stated that the disagreement over which species was the earlier*riser
was the 'fact' that birds of higher latitudes (wright's records)
would sing earlier in the morning than birds of lower'latitudes (Allen's
records). However, this author believes that the major difference
between the awakening times of these two species is found in the dif-.
ference of the awakening time criterion used and the much greater
amount of night song indulged in by the Song Sparrow. Therefore, this

author also believes that the Robin is the earlier riser of the two

species.

2h

Phoebe (Sayornis phoebe). The Phoebe was present in the woodlot dur-
ing periods I and II only. The reason for the birds' absence dnring
period III is not known. The number of birds during periods I and III
was usually three but occasionally only two could be heard.

This species was the anomaly among the birds studied because it
arose earliest during the second period of residency in the woodlot
which was during period II. All other species studied arose earliest
during their first period in the woocnot. The average awakening time
for the Phoebe during period I was -20.0, while the average for period
II was -31.h (Table II). The means of these two periods are signi-
ficantly different (Graph II). At present the author is unable to
determine why the Phoebe should differ from the other species studied
in this respect. There were no notable fluctuations in the awakening
times during the periods except near the end of period II when the

birds tended to give their awakening song a little later.

Red-eyed Vireo (1723.9. olivaceus). The Red-eyed Vireo was present in
the woodlot during periods II and III. This species arose earliest
during period II (Table I and Graph II). A slight overlapping of

the lines representing periods II and III will be noted on the graph.
The test for significance of the means between the tw0 periods showed
the means to be significantly different. After reaching their latest
awakening time in August and September, the Red-eyed Vireo arose
earlier in October, thus exhibiting a very definite fall revival of

song as well as an earlier awakening time. The similarity of periods

25

II and III was caused by the birds exhibiting this decided change in
their fall awakening time. While this phenomenon occurred for several
species, nowhere was it so marked as with the Red—eyed Vireo. For
example, mid-August awakening times were about +1; to -5, while for

the first half of October the records were -11, -6, —1h, and --12.

After these four records were taken, the species was no longer heard
in the area. The above information shows the change in the awakening
time of the Red-eyed Vireo within one period (III), and, along with

the Flicker, illustrates inaccuracy in the periods as defined in deter-
mining the trend of awakening time over the seasons. In period II

there was no fluctuation of the awakening time.

Iood Pewee (Contogs gig-erg). The Wood Pewee was present in the wood-
lot during periods II and III. It followed the pattern of rising
earliest in the first period during which it was present. The mean
of period II is significantly different from the mean of period III
(Table II and Graph II). This bird, as did most other species, awaken-
ed later as the season progressed into fall.

The song of the Hood Pewee in the early morning and late evening
is unique. This song was first described as such by Oldys (1902, 1901;)
who applied the term "twilight song" to it. Oldys (1901.) stated that
this twilight song lasted about 20 minutes and was not restricted to
any particular season, at least while the species was in his area.
Saunders (1929) wrote that the twilight song lasted ho minutes or
more and was limited to the time from June to late July. Craig (l9h3)

in his extensive work on the song of the Wood Pewee, divided the

26

twilight song into a prelude, a rhythmic portion, and a postlude.
The rhythmic portion which is the true twilight song lasted from 16
to 32 minutes (average 214%) and was sung from the time of the birds'
arrival in the spring until August and September. This observer
found that the twilight song was sung from the time of arrival until
Just before it left the area in September. hiring the morning the
song lasted from the time the bird began singing until about ten min-
utes before sunrise, a period varying between five and 50 minutes.
There were no major fluctuations of the awakening time within any
given period, but there was a tendency toward later rising during the
latter part of period III.

Wright (1913) found the Wood Pewee to be the earliest riser of
all the species he studied. However, Allen (1915) aptly pointed out
that the wood Pewee sings much at night and so night's (1913) records
undoubtedly reflect his use of night songs as awakening songs. Craig
(1910) stated that the Hood Pewee was one of the earliest birds to
sing in the morning. He also wrote that all WOOd Pewees for which he
had records awakened at about the same time with only mild variations
in this time. There was no difference in the east-west awakening
times and only slight variations in the north-south data (records
taken between 3h° to 15" north latitude and 68 3/h° to 97 1/2° west
longitude). Craig (1910) also stated that the awakening time became
earlier in relation to sunrise as the season advanced, and then,
after reaching its earliest rising time, became later in relation to
sunrise until the birds left on their fall migration. This author

obtained the same results from this study.

27

Gatbird (Dumetella carolinense). The Catbird was present in the wood-
lot during periods II and III, and its earliest awakening time average
occurred in period II. The mean of this period (Graph I and Table
III) is significantly different from the mean of period III.

The awakening time of the Catbird was quite regular, and it
appeared to be the most regular riser of the species studied. Dur-
ing one portion of period II, the daily awakening times were -h1, -h0,
-39, ~10, and --38. In period III, there was one sequence of awakening
times as follows; -28, ~29, —29, -29, -23, and —28. During the same
days that the above records were taken other species were much more
variable.

The termination of the awakening time records of the Catbird
differed from that of the other species. Most of the other species
studied gave their awakening songs or calls later immediately before
they left on the fall migration. The awakening time of the Catbird,
in contrast, remained constant until the birds departed from the area.

No less than four birds were recorded as singing or calling in

detemining the awakening time of the Catbird.

fellow Iarbler (Dendroica petechia). The Yellow Warbler was present
in the woodlot during the very last part of period I (two awakening
time records), all of period II, and part of period III. Because of
the lack of sufficient records in periods I and III, no attempt will
be made to present conclusions concerning the awakening times during
these periods. The reason for the lack of data in period I was the

absence of the birds. During period III, however, the species was

28

present but did not sing or call often enough in the early morning
to permit the recording of the awakening tine.

The Yellow Warbler was a reasonably constant riser (Graphs VI,

I and Table XIII). However, this species is the last of the birds
included in this study that exhibited extreme variations of the
awakening time within a period. After June 29, the Yellow warbler
began rising several minutes later than it had been rising before
that date. From June 29 until the last reliable record was taken on
August 2, the awakening time of this species was consistently later
than it had been during the first of the period. ‘

Ghapnan (1907) quoted Jones as stating that there was a decided
decrease in the amount of singing of the Yellow Warbler after mid-July.
A similar decrease was noted in the present study after August 1 after
which time the birds almost ceased singing. Until that tine, there

had been five birds whose awakening times were recorded every morning.

Bough-winged Swallow (midoptery; ruficollis). This species was
present in the woodlot during period II only. Although there were few
records for this bird (12), they were included because early rising
- tendencies have been attributed to members of the swallow family.

The birds roasted about 150 yards up-stream from the observer, and it
was their habit to fly back and forth over the river in front of the

observer. The author is quite certain that these calls were the first

morning utterances of this species as their twittering could be heard

quite some distance. Graph I and Table 11v present the data on the

awakening time of the Rough-winged Swallow.

DISCUSSION
The Awakening Time Curve

It is a well known fact that birds rise earlier in the spring
and summer than at any other time of the year. Any field worker can
observe that birds do not rise as early in the winter as they do dur-
ing the winner. The data presented in this paper has shown that the
awakening time curve of several species of birds became not only
earlier in relation to clock time but also in relation to the sun-
rise and civil twilight curves. Then, after the birds reached their
earliest rising time, the awakening time curve again more closely
coincided with the sunrise and civil twilight curves as fall approach—
ed. Other authors have concurred in these findings, notably Walker
(1928), Allard (1930), Craig (l9h3). and Nice (19h3). However, Emlen
(1937) has observed a female Mockingbird (Minus; polzglottos) that a-
wakened later in relation to the time of sunrise as the nesting sea-
son approached.

Allnrd (1930) and Nice (l9h3) have stated that the awakening time
curve more closely approximates the civil twilight curve than it does
the sunrise curve. Allard (1930) stated that the earliest rising time
occurred at the summer solstice. This is what lead him to believe
that the awakening time curve was a duplicate of the civil twilight
curve. Careful scrutinizing of graphs I through VI will show that

this is not true, at least for the birds in this study. The awakening

29

30

time curve did not follow the civil twilight curve any more closely
than it followed the sunrise curve. ' The earliest awakening times
were reached either before, at, or after the summer solstice.

The period of the earliest mean awakening time for the perh-
manent residents and the earliest migrants (except the Phoebe) for
this study was period I. The remaining species had their earliest
awakening tine means during the first period in which they were present.
The general statement may be made that the mean awakening time of a
migrating species will be earliest in that period during which it is
first present in the area while the mean awakening time of a perma-
nent resident will be earliest in period I. It is obvious, not only
from this statement, but, in addition, from perusal of the limits of
the periods as used in this paper (Table II), that some birds will
have their earliest awakening times before the summer solstice. OthC
species will have their earliest awakening times either corresponding

with or after the summer solstice.

Order of Awakening

Since there is a definite order of awakening, an analysis of this
order within each period would be of interest. For each species dur-
1118 each period, the graphical method (Dice and Leraas, 1936) was used
to test for significant differences of the means. A range of four
standard errors with the mean as the mid-point was plotted on the
graph for each species. Significant differences of the means can be

discerned by the lack of overlap of the line for one species with the

31

line of another. Broad overlap of these lines indicated no signifi-
cant differences of the means. Slight overlap or any doubtful cases
were tested by the use of the "t" test. Graphs VII through XI were
constructed in this manner.

The difficulty of discerning which species of birds arose before
other species was enormous during certain periods because of the ex-
tent of the overlapping of the lines representing four standard er-
rors. To facilitate reading of Graph XI, Tables XV through XVIII
were made. EXplanation of these tables is presented with Table XV.

The awakening order of the various species was easily dissem-
ible during period I. Graph II and Table XV show that the Robin was
the earliest riser. The Song Sparrow and the Cardinal, although hav-
ing very different means, were not significantly different. The means
of the Crow and the Phoebe were significantly different when tested,
but the means of the White-breasted Nuthatch and the Flicker were not.
Therefore, the species observed in this stud], for the most part, did
rise at significantly different times during period I.

Periods II and III were the two most complicated periods in dis-
tinguishing the order of awakening. Written descriptions of the order
of awakening of the Species during these periods would serve no pur-
pose. One can discern the general pattern of the awakening order by
studying Graph XI and Tables XVI and XVII.

Period IV is depicted in Graph XI and Table XVIII. This period
is much easier to evaluate than the tw0 preceding periods mainly be-

cause there were only four species recorded. The Cardinal was the

32

earliest riser, while the mean of the Song Sparrow did not differ
significantly from that of the Crow. Thus these two species arose
essentially at the same time. The White-breasted Nuthatch followed
as the latest riser.

The above comparison of the order of'awakening times of the Species
by periods has shown the apparent futility of trying to establish an '
exact,'unchanging order of awakening. However, one should notice that
the total range of awakening times, as computed by allowing two stan-
dard deviations above and below the mean for the earliest and the latest
rising species, was from +9 to -103 minutes from sunrise in period I,

a range of 112 minutes. In period II, the range was 97 minutes (+h to
993). In period III, the range was reduced to 63 minutes (*7 to -56),
and in period IV; the range covered only to minutes (+5 to -35). This
reduction of the range of awakening time indicated that there was a
tendency for the various Species to reduce their awakening times to a
narrower range as the season progressed. All of the species studied
arose at more similar times in the winter than they did in the spring.
This statement can be shown further and more conclusively'by computing
the range of awakening times for those Species that were permanent
residents. In periodiI, the range was 73 minutes for these qiecies,
in period II 50 minutes, in period III 50 minutes, and in period IV
no minutes. -From these data, the pattern for the Species of this study
appeared to be that birds vary considerably'over the seasons in their
awakening times, but a definite pattern existed in which all Species

tended to rise within a shorter period of time during each

33

naming as the seasons progressed (beginning with period I). The species
awakening times were reduced to a "base level" in the late fall and win-
ter. As spring approached, the earlier rising species arose much ear-
lier than the later rising species and thus widened the range of awe.-
kening times once more.

In the comparison of all species, the failure of the ranges of
periods I and II in differing from each other was due to the Robin.
This Species remained at its earliest awakening time during period II
and thus equalized the total ranges of periodsI and II. For Species
that were permanent residents, the difference in theranges of periods
I and II was 23 minutes. Periods II and III for these Species were
identical, however, indicating a Stunner-fall equilibrium of awakening
time. The range of awakening times during period IV was the smallest

of any of the periods.

Changes in the Order of Awakening

Walker (1928) was of the Opinion that the order of awakening

times between species did not change through the year. Allard (1930),
having a different opinion, stated that the order did change. This
observer believes that some species change their relative sequence of
awakening times through the seasons. In Table III will be found list-
ed the periods and the relative positions of the four species that
were permanent residents. The means of the Song Sparrow and the Car-
dinal were not significantly different during the year except in period

IV when the Cardinal arose earlier. However, in period I, the average

3h

of the awakening times of the Song Sparrow“was five minutes earlier
than the average of the Cardinal. The original records revealed that
in.alnost every instance the Song Sparrow arose before the Cardinal
during this period. Therefore, the Song Sparrow and the Cardinal did
change their order of awakening during the year if one considers the
field Observations valid. The "t" test between these periods showed
no significant differences in the means of these two species. The
Crow and the‘White—breasted Nuthatch maintained their same order of
alakening throughout the year.

Other changes in the order of awakening times are shown in Table
II. The Robin.and the Catbird reversed their sequence between.periods
II and III, and the Phoebe and the Crow exchanged places in the m-
kening order between periods I and II. Records indicated that there
may have been a change in the order of awakening between the Bed—eyed
Vireo and the Crow during periods II and III. The difference of the
means‘of these two species was so great during period III that the
author believes that the Red-eyed Vireo and the Crow changed their
order of awakening. However, this change was represented in the means
only, for the means of period III did not test significantly different.

The comparison of the change in the order of awakening of species
during the year did not show the gradual transition that occurred as
these birds changed their order of awakening. There were days during
the observations when the species changing their awakening order were
rising at the same time. This did not last long but was observable in

the field. Thus one can not only observe a change in the order of

35

awakening from the data, but one can also observe this change as it
is accomplished in the field.

wright (1913) and Allen (1913, 1915) did not indicate that any
change in the order Of awakening took place. This is undoubtly due
to the paucity of data that they had. Also, their data were collected
during a number of years and over such a small portion of the year
that any change in thw awakening order was probably not noticeable to
either author. Walker (1928) did not detect any change in the awa-
kening order from her data which were collected from October through
July. Allard (1930) stated that Species did vary in awakening order
through the seasons. His work extended from January through August.
The results Of this study showed that the greatest change in the order
of awakening occurred between the months of February and November
while December through January saw no change in the order Of awaken-
ing. Thus Walker (1928) could easily have overlooked any change in
the order of awakening that may have occurred. The results Of this

study agreed in general with Allard (1930); that is, the order of

awakening changes over the year for many species.

Ranking Of Species as Early or Late Risers

Many authors who have written on the awakening times of birds
have expressed Opinions as to the relative earliness of lateness of
this time for different Species and families. For the most part they
have agreed. However, most Of them were considering only the period

0! intense singing. This will introduce error into their statements

36

because the time Of awakening not only becomes later and later in re-
lation to sunrise during the year, but, also, the order Of awakening
changes. Therefore, any statements as to the earliness or lateness Of
the awakening time Of a Species should include the time Of year during
which the data were collected. Since most Observers have gathered their
data in the Spring and early summer, comparable records in this study
include data form.periods I and II. In Table III the author has ar-
bitrarily divided the species into four categories according to his
computed means and his beliefs as to the relative awakening times dur-
ing the seasons. The terms "earliness“ and."lateness" of awakening for
all periods are based on period I-II. Consequently there were no early
risers in periods III and IV. The following comparisons with the re-
sults of other authors were based on this author's period I-II.

There is considerable agreement concerning the species which are
extremely early risers in relation to sunrise. 'Wright (1913): Allen
(1915). Allard (1930), and Wallace (1955) all consider the Robin to
be a very early riser. The Song Sparrow is considered to be an early
riser (wright, 1913, Allen, 1913, 1915, Nice, 19143), although its posi-
tion in the 'bird clock" has been diSputed. Oldys (190h),‘wright (1913),
and Craig (19h3) considered the WOOd Pewee to be an early riser. NO
statements as to the position Of the Catbird or the Cardinal were
fOund in the literature.

The Roughdwinged Swallow and the Phoebe were in the "middle" ris-
ing group. Published evidence indicates that both the swallow and the

flycatcher families are early risers. ‘Wright (1913) and Allen (1913)

f!

37

stated that swallows were early risers. Allen (1922) stated that the
Kingbird (Ezra—113113 Ema) was one of the earliest risers, and
Wright (1913) and King (1955) stated that the Alder Flycatcher
(Eidonax traillii) was a very early riser. Wright (1913) consider-
ed that the Wood Pewee was the earliest rising Species. MacQueen (1950)
published records showing that the Least Flycatcher (widow minimus)
was an early riser. However, the results of this study Showed that
tw0 of the three species Observed representing these two families were
not early risers. Of course, it is not essential that all members Of
a family rise at the same or nearly the same times, But generalized
statements as to the awakening time of an entire family Of birds
should not be made until substantiated by sufficient Observations.

The Crow, Yellow Warbler, and the Red—eyed Vireo were in the "late"
rising group in this study. Allen (1913, 1915) stated that Crows were
early risers (average —h2) while Wright (1913) stated that they were
comparatively late (average -26). This study indicated that Crows
were late risers. lright (1913) considered warblers in general to be
early while Saunders (1929) considered them late risers. Allen (1913)
stated that warblers as a family were not late, but he also wrote that
the Yellow Warbler was one Of the latest rising of the family. Bent
(1953), however, stated that the Yellow Warbler was an early riser.
Little has been written concerning the awakening time of vireos. Wright
(1913) found the average Of 22 records of the Red-eyed Vireo to be -31.
This author found the average for this a) ecies to be -25 for 21; records.

The last group Of Species has been termed "very late" because this

/\

38

group differed so greatly in awakening time from the other three groups.
Many observers have agreed that woodpeckers were late risers (wright,
1913, Bussman, 1933), some even stating that certain species did not
arise until after sunrise (Sherman, 1910, Tanner, 19hl). Apparently
there is little known concerning the awakening time of the White-breast-
ed Nuthatch. ‘Iright (1913) gave one record Of-+lO, and Tyler (1916)

published one record Of +19 stating that this pecies was a late riser.

 

 

This author found the White-breasted Nuthatch was a late riser, but
considerably earlier in awakening time than the two records above in—
dicate.

The various Species did not retain their same ranking as early or
late risers as the season progressed. By referring to Table XXI, one
can see that only the late risers maintained the same rank throughout
the entire year. This again Shows the fact that the birds all tended
to awaken within a narrower range Of time in.minutes each day as the
seasons advanced toward winter. However, one Observer has found that
this is not true. Vaurie (19h6) stated that both the Robin.and the
Song Sparrow were early risers in mid and late summer. This study

did not corroborate these findings Of vaurie (l9h6).

SUMMARY AND CONCLUSIONS

1. The awakening time curves of most species of birds approximates
the sunrise and civil twilight curves during the year'but with system—
atic deviations during some seasons.

2. The average awakening time of a species is a useful figure in
illustrating the variation in awakening time during the year if deter-
mined by the use of periods of relatively constant awakening times.

3. It is obvious that birds awaken earlier in that time of year when
the sun rises earliest; that is, they rise earlier according to the :
clock.

h. It is not so obvious that Species of birds also awaken earliest
in the spring (periods I and II) in relation to the sunrise and civil
twilight curves.

5. All species do not awaken earliest in relation to sunrise at the
summer solstice.

6. Permanent residents rise earliest in relation to sunrise during
period I as designated in this paper (most of March and all of April).

7. Migrating species rise earliest in relation to sunrise for about
one month immediately after arriving on their nesting grounds. (There
are certain exceptions which do not sing for about a week after4arrival.
These species then quickly reach a "peak" of early rising.)

8. The order of awakening for certain species may vary through the
year.

39

no

9. The mean awakening time of birds varies most from species to
species during the spring and early summer, the range of difference
being least during the fall and winter periods.

10. Comparison of the observations in the present study'with work

fore sunrise regardless of moderate differences of latitude and lon—

of others suggests that some species arise at about the same time be- 7]
Q
l
I!
gitude. 1

I Tables

II Graphs

APPENDIX

 

I TABLES

ha

‘v—i

STATUS OF SPECIES IN THE WOODLO’I‘

 

 

__ species Status First Date Last Date
Cardinal permanent .. .—
Catbird smmner Kay 1; October 11
Crow permanent ~— .........
flicker permanent -- —--
Phoebe summer March 22 July 20
Red-eyed Vireo summer say 16 October 15
Robin summer March 13 November 12
Rough-winged

Swallow summer May 6 June 1;
Song Sparrow permanent -—-—- ....
White—breasted

Nuthatch permanent -- -—-
wood Pewee summer May 18 September 19
Yellow Iarbler stunner April 28 September 6

 

 

 

TABLE II

ARBITRARI PERIOIB OF AWAKENING TIMES
WITH INCLUSIVE DATES

 

 

 

Period Dates
I March 7 to April 30
II Kay 1 to July 30
III August 1 to November 30

IV December 1 to March 6

 

TABLES III THROUGH XIV'CONTAIN PEBTINENT
DATA FOR EACH SPECIES

 

 

 

 

 

 

 

 

TABLE III
CARDINAL
Period Number of Range“ Mean* Standard Standard
Records Deviation Error
I 21 17-68 10.6 114.176 2.69
H 31 26446 35.5 11.89 0.88
III 3h 12-h6 26.8 7.93 1.36
Iv 23 8-33 23.2 5.75 1.77
TABLE Iv
SONG spmow
Period Number of Range Mean Standard Standard
Records Deviation Error
I 21 26-76 1.5.9 13.51. 2.96
II 30 23-uo 33.6 17.60 0.81;
III 3’. 16-36 25.8 11.71; 0.81
Iv 20 +2-29 17.8 7.89 1.77

 

* The range and mean for each period in Tables III through XIV are

expressed in the number of minutes before sunrise except where a

plus (+) sign indicates minutes after sunrise.

h6

 

 

 

 

 

 

 

TABLE V
CROW 3:
Period Number of Range Mean Standard Standard
Records Deviation Error
IV 23 {8-27 16. 3 8 . 55 1.78
TABLE VI
MTE-BREASTED NUTHATCH
Period Number of Range Mean Standard , Standard
Records Deviation Error
I 22 «3-19 11.9 11.85 1.03
II 21 +6-20 9.7 6.50 1.112
III 21. +1o-21 6.2 6.62 1.35
IV 15 +5-10 1.7 3.59 0.93

 

117

 

 

 

 

 

 

 

TABLE VII
FLICKER
Period Number of Range Mean Standard Standard
Records Deviation Error
I 10 +6-16 7.2 8.01 2.53
II 9 +6-21; 1.3 7.75 2.58
III 1h +9-17 5.1 5.91 1.58
TABLE VIII
ROBIN
Period Number of Range Mean Standard Standard
Records Deviation Error
II 28 22-86 63.5 114.98 2.83
III 20 17-80 25.9 5.80 1.30

 

h8

 

 

 

 

 

 

 

 

 

 

 

TABLE II
PHOEBE
Period Number of Range Mean Standard Standard
Records Deviation Error
I 12 3-32 20.0 7.65 2.21
II 26 22-h2 31.1: 6.00 1.18
TABLE I
RED-EIED VIREO
Period Number of Range Mean Standard Standard
' Records Deviation Error
II 28 1-37 2h.8 10.79 2.20
III 10 «4-35 13.87 12.28 3.89 '
TABLE II
WOOD PEWEE
Period Number of Range lean Standard Standard
Records Deviation Error
II 23 +8-57 hh.6 17.h1 3.63
III 15 9-h8 26.7 18.76 3.81

 

 

., e...

h?

 

 

 

 

TABLE III
CATBIRD
Period Number of Range Mean Standard Standard
Records Deviation Error
II 29 2-63 83.1 12.82 2.30
III 211 18-38 29.5 11.21 0.86
TABLE XIII

YELLOW WARBLER

 

 

 

 

Period Number of Range Mean Standard Standard
. Records Deviation Error
II 28 10-31 21.9 6.1111 1.22
TABLE XIV

ROUGH-WINGED SWALLOW

 

 

Period Number of Range Mean Standard Standard
Records Deviation Error

II 12 13-h8 3h.5 10.98 3.17

 

50

TABLEJXV

RELATIVE SEQUENCE 0F.AWAKENING
FOR PERIOD I

 

 

ls Rabin (61e6)

2. Song Sparrow (u5.9) 3. Cardinal (no.6)

 

he CEO" (2509)

5. Phoebe (20.0)

6. White-breasted Nuthatch (11.2)--’7. Flicker (7.2)

 

Explanation of Tables IV, XVI, XVII, and XVIII.

The Species are in their approximate awakening order from.t0p to
bottom. No lines between species indicate: that their means are signi-
ficantly different. Lines between species indicate the means of the
species in question are not significantly different. Dotted lines be—
tween two species indicate that the means were tested.with the "t" test
and were found significantly different. The 95% confidence level was

considered significant. Figures in parentheses are the means during
the period in minutes before sunrise.

51

TABLE XVI

RELATIVE SEQUENCE OF AWAKENING
FOR PERIOD II

M
1e Rabin (63e5)

 

2. Woo‘d Pewee 0111.6)“. “.3. Catbird (153.1)

E
g .
"card%na1 (35.5)

Roughdwinged ’ ’,/” .
Swallow (3h S)‘-—- Sons Sparrow (33 6)..“~ i
. .pPhoebe (31.h)

Rad-eyed Vireo (2h.8)'

Yellow Warbler (21.9)

Crdw (18.5)

11. White-breasted Nuthatch (9.7)
I

1
12. Flicker (1.3)

 

52

TABLE XVII

RELATIVE SEQUENCE OF AWAKENING
FOR PERIOD III

 

 

Catbird (29.5)_?::::::°-'~-=

/Cardina1 (26.8)§

wood Pewee (26. 7)< E
I 1
\llobin (25. 9) E

|

Song Sparrow (25.8)

Red-eyed Vireo| (13. 8)

White-breasted Nuthatch (6.2)

 

 

Flicker (5.1)

 

53

TABLEIXVIII

RELATIVE SEQUENCE OF.AWAKENING
FOR PERIOD IV

 

 

1. Cardinal (23.2)

2. ‘Song Sparrow (17.8)

3e Crow (16e3)

h. ‘White-breasted Nuthatch (1.7)

 

Sh

TABLE XIX

CHANGE IN THE ORDER OF AWAKENING TIMES OVER THE FOUR PERIODS
AS ILLUSTRATED BY THE FOUR PERMANENT.RESIDENT SPECIES

 

 

 

 

 

 

2. Song Sparrow (17.8) Crow (16.3)

h. mite-breasted Nuthatch (1.7)

Period . Species O

I . 1. Song Sparrow (h5.9) Cardinal (no.6)
. Be or“ (25.9)
: h. White-breasted Nuthatch (11.9)

II . l..Cardina1 (35.5) Song Sparrow (25.8)
e Be or" (1805)
. 1.. white—breasted Nuthatch (9.7)

III . 1. Cardinal (26.8) Song Sparrow (25.8)
. 3. Crow (19.0)
I h.‘lhite-breasted Nuthatch (6.2)

IV : 1. Cardinal (23.2)

 

In this table note the gradual shift in awakening order of the
Song Sparrow and the Cardinal as the season progresses. Also note
the continuity of the positions of the‘Ihite—breasted Nuthatch and
the Crow. There was no significant difference in the rising time
between the Crow and Song Sparrow during period IV. Figures in

parentheses are the means during the period in minutes before

sunrise.

55

TABLE XX

CHANGE IN THE ORDER OF AWAKENING TIMES OF SEVERAL SPECIES

 

 

 

 

 

 

 

 

 

Crow (18.5) Red-eyed Vireo (13.8)
These do not test significantly
different at the 95% level, but
there is a great difference in
the means suggesting a change
in the awakening order.

Period II . Period III
Robin (63.5) . Catbird (29.5)
Catbird (h3.1) . Robin (25.9)

Period I : Period II
cm (25e9) e Phoebe (31.14)
Phoebe (20.0) . Crow (18.5)

Period II . Period III
Red-eyed Vireo (2h.8) . Crow (19.0)

 

Figures in parentheses are the means during the period in

minutes befbre sunrise.

TABLE XXI

RANKING or SPECIES BY RISING TIME

56

 

 

Period . '

Middle

Late

 

 

 

Early . . . Very Late

I Cardinal I Phoebe Crow I Flicker

I Catbird . Rough-winged . Red-eyed Vireo . White-breasted
I-II . . Swallow . . Nuthatch

. Robin , . Yellow Warbler ,

. Song Sparrow . . .

. wood Pewee . . .

O O O O

. (None) . Cardinal . Crow . Flicker

. . Catbird . Red-eyed Vireo . White-breasted

e e e 0 MthBtCh
III e e RObin e e

e e Song Sparrow e e

. . Wood Pewee . .

. (None) . Cardinal . Crow . Flicker

. . . . (presumably)
IV . . . Song Sparrow . White-breasted

e e e e N‘IbhfitCh

 

II GRAPHS

57

58
Explainations of Graphs I-VI

These-graphs illustrate the sunrise curve (dashed line), the
civil twilight curve (solid line), and the daily'awakening times of
12 species discussed. Symbols representing the awakening time are
circles (0) indicating singing, crosses (x) indicating calls, and
dots (°) indicating the awakening utterances when.no distinction‘was
made between calls and songs. The time of morning (E.S.T.) is given

on the left of the graph and the months of the year at the bottom.

 

59
GRAPH I

 

7:15‘

51; F

745

51/5

 

 

 

SUNRISE ArIO CIVIL TWILIGHT CURVES
WITH THE DAILY AWAKENING TIHE OF
A (Anorwnl
B. CRTBIRD
\ /’~
\\ //'m"l1
"‘ x x\ /
nib / x
x‘\ / I
no\ / H
a \ /
/ ',
A \ s/ ' I
‘ \ / I:
0 °. \ / 3'
0°" \ / m"
\ M.”
/
°o° .\ / ’l
o ‘ \ /
° \
° ./
o \ ‘°
0 \fl/
\ ° 0’ /'-1
\ 0 0° Do 0 /
\ ° ° . °°
\ 00 000° /
o a /
\ /
\ /
\ /
B \ /
/
\ / ..
\ / x I
\ /
\ ’ * ‘
/" .
\\ / '
/
\ / “
\.
\._,’/
0‘0
0 00
o o.
000‘0000 .006
O
o o o
l l L l 1 l l l l l L
TAN FEB MAR APR MAY TUN TUL AUG SEP OCT NOV DEC

GRAPH II

 

Suanxsf Ava CIVIL rerxaur cunves
WITH Tue DAILY swaneuzuc TIME OF
A Clow

a FLICKER

TH

 

 

OH

5315

TH

05

y”

 

 

 

l l l J l 1 l l l l l
TAN FEB MIR APR MAY TUN TUL AUG SE! OCT NOV DEC

 

61

GRAPH 1H

 

sunnrsE AND cxsz rerrcHT canves
WITH THE DAILY AWAKEIIIG True OF
4.PuoEa£
a REDfYED vrnEo

TU“

 

$5

5»: '

rw'

"-I!

I"!

 

 

11 1 1 1 1. 1 1 1 1 l 1
TAN fEB MAR AIR MAY TUN TUL AUG SEP OCT NOV DEC

 

 

62

GRAPH N

 

JUURISE AND CIVIL TerIGuT CURVES
WITH THE DAILY AWAKEMIUG TI“? OF
A ROBIN
L a Rouawwrusto SWALLOW

Tfl’

 

$5

$U

Tfl’

$5

$5

 

L 1 1 1 1 1 1 1 1 L 1
TAN FEB MAR APR MAY Tun TUL AUG SEP ocT NOV DEC

 

 

 

63

GRAPH I .

 

suuarsc AND CIVIL ?W1LIGHT canvas
WITH was 05:17 Ammxtwrwo Tru: OF
A.SOfl6 JPARROW

& wurrssncnsrcp
mnunrcu /“

7U5

615

$15

7-‘I5

(=15

Sils‘

 

 

 

J 1 1 1 L, 1 1 1 1 l 1
TM! FEB MAR APR MAY TUN TUL AUG SEP OCT NOV DEC

 

7d$

$3

545

TU

05

$3

61:
GRAPH II

 

 

L

l

 

SUNRISE Ana CIVIL TWILIGHT cunvs:
WITH THE DAILY AWAKENING True or
A.woop PEWEE
B. YELLOW WARBLER

 

l l L l I l I l l

 

TAN

FEB

MAR

APR MAY TUN Tut AUG SEP OCT NOV DEC

so

8 .6

MINUTES serous SUNRISE
8

GRAPH 12H

szouxszennr DIFF€A€~C£ 0F usafls

65

 

 

 

 

 

 

 

 

V
I
I 1‘
m m 1‘
I
I! + *
II I
IN
SONG SPARROW CARDINAL CROW

 

MINUTES BEFORE SUNRISE

GRAPH m

536N1F3¢ANY OIFFERENC‘ OF MEAHJ

 

70'-

60-

 

 

50’—

qo~

30-

 

 

 

 

 

1T

 

 

 

 

Kosxu WHITEfiREfiSTFD FLICKER
NUTHATCH

8 t

MINUTES BEFORE SUNRISE
8

l0

GRAPH 1X

SICNIFICAUT OIFFEREHCE OF MEAN:

67

 

 

 

 

 

 

 

 

 

 

 

PHOEBE REDTYED VIREO W000 PEWEE

. “same E,
1.

5a

-b
0

MINUTES OEFonE SUNRISE
’6‘ 8

'0

GRAPH X

SIGNIFICAHT orrrtntuce OF MGAUJ

 

 

 

 

5.4-

 

 

 

YELLOW WARBLER

CATBIRO

ROUGH-WINGE o
SWAI. L ow

 

69

GRAPH XI

SIGNIFICRNT OIFFEREHCE OF MGAA’S

 

.t\<~ad,.. ‘

:u..¢t~1.€ +

Leuflnxwwm :HI).

3.030 -

. keewcauezow.

 

eruemnuolll -

akin... Of); 1

:«uawtu_ —

. <11<tfi ©§QH

) Hurst _

b
corn: oiyl.ar: q

\fhwu -

UM! 4”.”
25.3.23; Ii

3,“. n( mam that:

 

«~ELI

#-

I
é-33e banana S :36? 4
‘
O'Nfihtvb
i¢>lne .hP.H¢ -

1‘

bath.) ._¥u .Abk .
$.44mP

- \ ..\.'aa

“whorl

«Siuaiwu —
trike.) 3. .. —

P
4

2:11. «1* 53:; 8‘
III|I’IIIL":’VW¥M. m-3))

3% u

 

 

.r “hat

4’

P

m... 1:..an a m

9‘ D" —

<-

 

th¢NLVW .
sonic»; 3.55 1

b p

4
I Cort.) £vhdk
:11,» 1 1.3.5

ski)

 

 

70 L-

50""

w u
m ”H53; wcoauu nuhzzut

Pl _
w

PERIOD 1!

PERIOD I

PERIOD I

PERIOD I

LITERATURE CITED

Abel, A. R. 19111. Notes on a northern Robin roost. Wilson 31111.,
26: 165-172.

Allard, H. A. 1930. The first morning song of some birds of Wash—
ington, D. C. Amer. Naturalist, 61$: 1136-1469.

Allen, RSI-I. 1913. More notes on the morning awakening. Auk, 32}
229-23 .

Allen, P. H. 1915. The status of the Song Sparrow and the Chipping
Sparrow as early birds. Auk, _3_2_: 110-113.

Allen, F. H. 1922. Some little known songs of common birds. Nat.
History, 22: 235-2142.

Are, J. 1951. Morning and evening bird calls. Jour. Bombay Nat.
Hist. Soc. , _5_C_)_: 1107-1113.

Bent, A. C. 1919. Life histories of North American thrushes, Icing-
1ets, and their allies. U. S. Natl. Mus. Bull. 196.

Bent, A. C. 1953. Life histories of North American wood warblers.
Ue Se Natle “113s Bulle 203e

Bussman, J. 1933. Experiments with the terragraph on the activities
of nesting birds. Bird-banding, g: 33-110.

Chapman, 1'. l. 1907. The warblers of North America. New York.

Craig, W. 1926. The twilight song of the Iood Pewee; A prelimi-
nary statement. Auk, I_1_3_: 150-152.

Craig, I. 1933. The music of the Wood Pewee's song and one of its
h'Be Auk, 59.3 17h-1780

Craig, W. 19143. The song of the Wood Pewee, Hyioohanes virens Linnaeus:
A study of bird music. N. I. State Mus. Bull. No. 3311.

Dice, L. R., and H. J. Leraas. 1936. A graphic method for comparing

seteral sets of measurements. Cont. Lab. Verb. Genetics, Univ.
“101113811, NO. ’, pp. 1.30

70

71

Emlen, J., Jr. 1937. Morning awakening time of a lbckingbird. Bird-
banding, Q} 81.

Howell, J. C. 19h0. Spring roosts of the Robin. Wilson Bull., 52:
19-23. "

King, J. R. 1955. Life history of Traill's Flycatcher. Auk, 72: 1h8-
173. . ‘—

Laskey, A. R. 19th. A study of the Cardinal in Tennessee. ‘lilson
We 3 £9} 27‘h-he

IacQueen, P. I. 1950. Territory and.song in the Least Flycatcher.
Wilson Ball-e, 933 19’4‘2050 ‘

McCabe, a. A. 1951. The song and song-flight of the.A1der Flycatcher.
Wilson Bull., 62; 89-98.

Nice, N. M. 1935. Some observations on the behavior of Starlings and
Grackles in.relation to light. Auk, 52} 91-92.

Nice, M. M. 19h3. Studies in the life history of the Song Sparrow
IIe Trans. Linne SOCe Ne Ye, 93 1.3290

Oldys, H. 1902. Parallel growth of bird and human.music. Harper's
magazine, August: h7h-h78.

Oldys, H. 19Gb. The rhythmical singing of the lbod Pewee. Auk, g1:
270-271.

Saunders, A. A. 1929. Bird song. New'York State luseum.Handbook 7.

Seibert, H. C. 1951. Light intensity and the roosting flight of
herons in New Jersey. Auk, éfi} 63-7h.

Sherman, A. R. 1910. At the sign of the Northern Flicker. ‘Wilson
Bu11e, 22} 135.166.

Smithsonian Meteorological Tables. 1951. Smith. Misc. 0011., vol. 11h.

Tanner, J. J. _19h1. Three years with the Ivoryhbilled'woOdpecker.
1nd. Mag., §23,5-1h.

Taylor,‘l. P. 1929. Order of awakening of some Arizona birds. Auk,
hés 399e

* Tyler,'w. n. 1916. A study of a Whiteebreasted Nuthatch. Wilson
Bu11e, gg} 18'250

vaurie, C. 19h6. Early morning song during middle and late summer.
Auk, 92.8 163‘171e

72

Walker, E. B. 1928. The relation of light to the awakening song of
birds. Unpublished M. A. thesis. George Peabody College for
Teachers, Nashville, Tennessee.

Hallace, G. J. 1955. An introduction to ornithology. MacMillan Co.,
, New,York. . .

'Wright, H.'W. 1912.'Morning awakening and evenpsong. Auk, 22; 307-
327.

wright, H. W. 1913. Morning awakening and evenpsong. Auk, 29}
512-537.

ROOM USE 02.511

 

 

“7111111 11117111111111“