1387
Postemergence
northern
turtle,
behaviour
peripheral
.Abstract:
.:
population
Emydoidea
K. Lorraine
Standing,
Ian P. Morrison
of neonates
of Blanding's
blandingii,
Thomas
in a
in Nova
B. Herman,
Donna
Scotia
D. Hurlburt,
and
The early postemergencebehaviourof neonateBlanding's turtles, Emydoideablandingii, was studied in Nova
Scotia in 1994and 1995to test the hypothesisthat hatchlingsgo to water immediatelyupon emergencefrom the nest.
At the northeasternlimit of the species'range,hatchling turtles should seekaquatichibernaculato avoid freezing;
where emergencefrom the nestis delayeduntil late autumn, selectionpressurefor water-seekingbehaviourshould be
strong. Using pigmentedpowder that fluorescesunder UV light, 78 hatchlingsfrom 12 lakeshorenestsand 2 inland
"
nestswere tracked for a maximumof 11 days. On the first day after emergence,hatchling movementwas random with
respectto water. Hatchlingsdid not orient themselveswith respectto slope, vegetation,or nestmates,nor did
individuals follow a consistentcompassbearing amongdays. Often trails were convolutedin the openand relatively
straight underdensevegetation.Evidencesuggeststhat hatchlingsin this populationmay overwinter in terrestrial as
well as aquatichibernacula.
Resume: Le comportementde Tortues mouchetees,Emydoideablandingii, fraichementemergeesdes nids a ete etudie
enNouvelle-Ecosseen 1994et 1995pour eprouverl'hypotheseselonlaquelleles petitestortuesgagnentl'eau immediatement
apres leur sortie du nid. A la limite nord-estde la repartitionde l'esp"ece,les petitestortues devraienttheoriquementse
chercherdes hibernaculumsdans l'eau pour echapperau gel; la oil l'emergencedu nid est retardeejusqu'a la fin de
l'automne,la pressionde selectionqui favorise Ie comportementde recherchede l'eau devrait logiquementetre forte.
La trajectoire de 78 tortues neonatesa ete suivie durant un maximumde 11 jours a partir de 12 nids sur les rives et de
2 nids loin des rives, par utilisation d'une poudre pigmenteefluorescentea l'ultra-violet. Au ler jour de I'emergence, la
trajectoire des tortues etait aleatoire par rapport a la positionde l'eau. Les tortues ne s'orientaientni en fonction de la
pente, ni en fonction de la vegetationou par rapporta leurs congeneres,et aucunetortue n'a suivi une trajectoire selon
une direction geographiquedefinie pendantplusieursjours. Les trajectoiresetaientsouventcirconvolueesen terrain
ouvert et relativementrectilignessousla vegetationdense.II sembleque les tortues neonatesde cette population
puissentpasserl'hiver dansdes hibernaculumsterrestresaussibien que dansdes hibernaculumsaquatiques.
[Traduit par la Redaction]
Introduction
Blan dmg
.,
f
res
h
wa
th
f
cally reproductive females. The paucity of information on
turtl
t
s
er species
th
G
(E
..
e
my
,
7
wes
em
ova
th
WI
t L
k
sou 0 e rea
th
t
N
sou
th
d
.d
01
ea
a
narrow
Th
a es.
S t.
co
bI.
la,
a
nd
a
."
mgll
)
1 t ' t d. al
a
all
e sm
t th
e no
" th
N
'
IS
I u m
d "
a
W
ew
range
t
Id
t
or
cen
ere
1 t "
d
, IsJunc popu a Ion m
rth t
t
.t f
eas
em
d '"
ex reml
yo.
1993
b
th
young
turtles
makes
it
difficult
to
protect
the
younger,
more
.
vulnerable age-classes effectIvely. Consequently, we have
...
directed
h..
f E b ' nd...,
N
S .
IStOrystrategies 0 , ta mgll m ova cotIa.
our
research
efforts
at
understandmg
the
early
lIfe
..
mformatlon
especles range, was ecare
y e
C
' tt
th Stat
f E da reaene
d W ' ldl m
' ti
'
c ana
d
omml
ee
on
e
us
0
n
ngere
I
I
e
m
a
(Herman et al, 1995). The resulting Recovery Plan for the
th
1 l.ti h.
fE bl
on e ear y I e IStory 0 .an
poI?~lation (Herman et al,)2 identifies research priorities to
facIl!tate man~g~~ent. As part of the~e effort~ sever~
studies were mltIated (McMaster 1996; McNeIl 1996;
Mockford 19965; I,P. Morrison and K,L, Standing, in
preparation), Despite these recent contributions, our understanding is primarily restricted to adult age-classes,specifi-
2 T.B. Herman,J.S. Bleakney,J.S. Boates,J. Dickie, C,
Drysdale,J. Gilhen, I. Morrison, T. Power, and K.L.
Standing. National RecoveryPlan for the Blanding's Turtle
(Nova Scotia Population),Recoveryof Nationally
EndangeredWildlife Committee,Ottawa, OntoSubmittedfor
publication.
3 McMaster, N,L. 1996. Age structure, distribution, habitat
selectionand movementpatternsof juvenile Blanding's turtles
(Emydoideablandingii) in Kejimkujik ~ational Park, Nova
.,
d
1
d
t
ReceivedNovember 13, 1996. AcceptedMarch 26, 1997.
K.L.
Standing
T.B.
, ..'...
Herman
I and
D.D.
Hurlburt.
Departmentof Biology, AcadIa UnIVersIty,Wolfville,
NS BOP lXO C d
In M
' , panka ac'
da K " mku ' ik N t. al P k
ar sana,
eJI ~
a Ion ar,
M.c.. I orrIson.
d B .d
A
I. C ty NS BOTIBO C da
alt an n ge, nnapoIS oun ,
, ana.
I Author to whom all correspondence
shouldbe addressed
(e-mail: [email protected]).
Can. J. Zool. 75: 1387 -1395 (1997)
As
4
for
Scotia:
McNeil,
many
freshwater
turtle
species,
most
d ...,
I' .
mgll IS Imlte d to reports
B.Sc. Hons.
thesis, Acadia
University,
Wolfv~lle,
J. 1996. Post-emergent
movement
of hatchlIng
N,S,
Bl d .,
I
(E .]..]
bl .]. .. (H Ib k» '
an mg s turt es myuoluea anumgll 0 roo
m
Kejimkujik National Park, Nova Scotia,B.Sc. Hons. thesis,
A cad.Ia Umverslty,
..
WOle,
lfv ' ll N..S
5 M k l' d S 1996 A
..
f
al
I .
oc lor"
.genetIc comparIson0 sever popuatIons
of Blanding'sturtle (Emydoideablandingii). B.Sc. Hons.
thesis,Acadia University, Wolfville, N.S.
@ 1997 NRC Canada
1388
Can. J. Zool. Vol. 75, 1997
of clutch size and survivorship and hatching success
(Herman et al. 1995; Power 1989; MacCulloch and Weller
1988; DePari et al. 1987; Petokas 1986; Congdon et al.
1983; Bleakney 1963; Brown 1927; Snyder 1921). Few
studies have investigated the behaviour and habitat requirements of immature E. blandingii (McMaster 1996, see footnote 3; Pappas and Brecke 1992) and virtually nothing is
known about hatchlings (McNeil 1996, seefootnote 4; Butler
and Graham 1995).
To date, the most significant contributions to our understanding of the behaviour and habitat requirements of
neonate E. blandingii have been provided by Butler and
Graham (1995). These authors concluded that newly
emerged hatchlings in Massachusettsmove from their inland
nests to wetlands, likely in search of aquatic hibernacula.
However, some of their results appear to be at variance with
ttIis conclusion, primarily because of ambiguous terminology the location of their stud site (inland) which did not
.'.
...~.
'.
facIlitate mvest!gatlon ?f th.e ~nentatlon "!echarusm, and the
s~all saD1plesIze, :whIch lImIted conclusIons at the populatIon level and specIes level.
While there are many similarities between the Nova
Scotia population and those elsewhere, Herman et al. (1995)
and Power (1989) have identified behavioural differences in
this northern population. To date, all differences noted are
associated with adult behaviour, and it is suspected that they
are responses to environmental constraints at the northt
I.. t f th
.,
(H
t 1 1995
,eas ern Imi 0
e specIes range
erman ea..
'
P?wer 1989; Bleakney 1958). The POSSI~IlItyofbehavI~ural
dIvergence m younger age-classes, partIcularly hatchlIngs,
must not be overlooked in the development of management
strategies. Since the protection of hatchlings is a crucial
component of the successful recovery and stabilization of the
E. blandingii population in Nova Scotia (Herman et al., see
footnote 2; Iverson 1990), a greater understanding of the
early postemergence behaviour of neonates in this northern,
isolated population is required.
It is generally assumed that hatchling freshwater turtles go
to water upon emergence from the nest (Ehrenfeld 1979).
However becauseof the limitations of Butler and Graham' s
(1995) s~dy and the possibility of behavioural divergence in
h hI ...remaining
atc figS, we InvestIgated the early postemergence behav...,.
loUr of hatchlIngs m the Nova ScotIa populatIon.
The unique nesting habits of E. blandingii in Nova Scotia
provide ideal cir~u~stances ~or inv.estigating neonate behavlour. Elsewhere m Its range,
IncludIng Massachusetts
."
.0 (Butler
and Graham 1995), thIS specIes typIcally nests Inland and
away from open water (Butler and Graham 1995; Ross and
Anderson 1990; Petokas 1986; Congdon et al. 1983); the
Nova Scotia population is unique in that females tend to
nest along cobble lakeshores. Though some females in this
population nest inland more than 200 m from open water, the
majority nest within 5 m of open water (Power 1989), and
though lake levels f1uctuate,hatchlings emerging from beach
nests have an unobstructed
.'
view
of the lake.
This
and the
duratIon of the study (two seasons), enabled us to study a
large! samp~e.than that of Butler. an? Graha~ (.1995),and to
~rovlde st~tIStIcalas well as qualItatIve deSC~IptIons
of hatchlIng behavIour. We were also able to eXaD1memore closely
the water-seeking strategies of E. blandingii hatchlings. Our
research is intended to complement the work of Butler and
GrahaD1(1995), providing further insight into the adaptations
of E. blandingii as a species, and the particular adaptations
of young turtles in the Nova Scotia population. We reasoned
that at the northeastern limit of the species' range, hatchlings
should seek aquatic hibernacula to avoid freezing, and that
because of late emergence from the nest, selection for such
water seeking should be strong. Here we address the explicit
hypothesis that hatchlings go to the nearest water immediately upon emergence from the nest.
Materials
and
methods
Study site
This studywasconductedin Kejimkujik NationalPark, Nova Scotia
(44°15'-44°30'N, 65.o00'-65°30'~), during summerand autumn
1994 and 1995: Nestl~g and hatchlIng m?ve~ent data were collectedat~~ majorn~stmgcenter~(asdescr~bed
m Power 1989)and
at an ad.dltlonalnestIngbeachdiscoveredm 1995..
.
Nestmg beacheshave a gradual slope and typically comprIse
fist-sizedcobbleand large pebbles.Beachesare sparselyvegetated
with grasses, rushes, and cranberry (Vaccinium macrocarpon)
(Roland 1945). The upperbeachesare bordered with huckleberry
(Gaylussacia baccata), blueberry (Vaccinium spp.), leatherleaf
(Chamaedaphnecatyculata), sweet gale (Myrica gale), red and
white pine (Pinus resinosa,and P. strobus),and maple (Acerspp.).
Inland nestswereconstructedin the gravel shoulderon the north
s~deora two-lane pavedroad. .Theshoulderis 2 m wide on both
sidesof the road and descendsInto a more steeplysloped bank of
mowed vegetation. The lower banks are bordered by a steep
forested...
InclIne (predominantlyP. strobus). A densemat of saturated Sphagnumspp, (0.5-1.0 m wide) occurs along the base of
the south-sideslope. The nearestopen water is a small marsh
located north of the road, 100-150 m west of the sampled nest
sites.
Nestiug and emergence
Beachesand roadsideswere surveyed for nesting E. blandingii
throughoutJune in 1994 and 1995. Nestswere protected against
predationwith a raised wire-meshcageonce nestconstructionwas
complete;this cagealso servedas a pen for emergenthatchlings.
BeginningSeptem~r I in both years,nestswere ch~~kedregularly
for emergenthatchlings.In late Octoberall nestcavitieswere excavatedto determinethe numberof unhatchedeggsand to free any
hatchlings. Upon discovery,hatchlingswere measured
(
.
I
th CL)
t h d s: .d t. fi t.
maximumcarapaceeng,
, no c e lor I en I Ica Ion (Stand
ing et al. 1995) dustedwith tracking pigment and released.
'
,
Dusting and tracking
d
d
h. h fl
d UV I ' ht (R d.
oure power, w IC uorescesun er
Ig
a Iant C01or,
LBRS Series,Richmond,Calif.), wasusedto track emergenthatchlings (Butler and Graham1993). Hatchlingswere dustedusing a
cotton swab,covering the carapace,plastronand leg sockets,but
particularcarewas takento avoid contactwith the eyesand nostrils
(Stapp et al. 1994), Nestmateswere assigneddifferent colours
wheneverpossibleso that individual trails could be distinguished
du:ing t:acking, Hatchlingsw~re releasedat the nestsite in random
onentat~onand were le.ftundisturbed. .
C 1
At night, powder trails were follo~ed usm~ hand-held UV lam.ps
(Ray tech Raytector) and marked with flaggIng tape secured with
pebbles.Flagging tape was superimposedon the entire length of
mosttrails, thoughin somecasesit waslaid at intervals. If a hatchling was found at the end of a trail, its locationwas markedand it
wasredustedwith the originalpigment.To minimizedisturbanceon
suchoccasions,the area immediatelysurroundingthe turtle, rather
thanthe turtle itself, was powdered;hatchlingswalked throughthe
@ 1997 NRC Canada
.
\
-,
Standing et al.
1389
powderas theyresumedtheir trek, facilitating the collectionof data
over severald~ys.
..Scotia
Flaggedtrails were mappedusIng a compass(azimuthto the
nearestdegree)and a measuringtape or meter stick (to the nearest
0.1 cm). Trails w~r~mappedin relat~onto water, vegetation,topography, nest of OrIgIn,and other trails.
Table 1. Orientationof hatchlingBlanding's turtles in Nova
with respectto the water nearestto the nest; analyses
include first-da routes onl
y
y.
No. of
Orientation
Angle of
Nest hatchlings Distance to nearest Length meanvector
T it al .No.
ra an YSIS
!
,
Trail datawere analyzedusing circular statistics(Batschelet1981,
1965; Zar 1984).The V testis usedto testwhethera set of angles
tendsto clusteraround a hypotheticaldirection (00). In this study,
we wanted to test not only whether hatchling movement (y) is
directed (nonrandom),but whetherit is directedtoward water (°0)
(Batsche1et
1981).The null hypothesisis thatthe parentpopulation
from which the sample(YI' Yz, ..., Yn) is drawn is randomly
(uniforrnly~dist~ibutedaboutthe hypotheticaldirection (00). .10"
The orlen~tl~n
from nest to near~st w.ater was standardized
among nests WIthIn years so that the Orientation (y) of the resultant
vector (R) of eachhatchling's first-day trek was convertedto the
polar angle y':
LetYI,Yz, ...,'tnbenobservationsofhatchlingorientation
(azimuthclockwise from north) on day 1 from a single nest; let fJ
be the direction to the nearestwater from this nest. Rotatingthe
originalzero direction (north)so that fJbecomesthe newzero direction, °0 yields the new polar angle 't' from the original 't. The
new polar angle is defined by
[1] 't' = Y -fJ
and its value is eitherpositive or negative,correspondingto either
a clockwise or a counterclockwiserotation from 00 (Batschelet
1965). The new sample ('t'I' 't'z, ..., 't'n) for each year was
plotted on a unit circle andthe meanvector(m '), its length (r'), and
its polar angle «1>')were calculated. Then the componentof the
meanvector with respectto the direction to the nearestwater, or
"the homewardcomponent" (v), and the test statistic (u) were
calculated(Batschelet1981). If the critical value u(a) is less than
or equalto the teststatistic (u), the null hypothesisof randomness
is rej~ted.
To Investigatethe possibilitythathatchlIngswereorgamzedWith
respectto somefeatureother thanwater,.a ph.enomenon
thatwould
be maskedby the Vtest,nestswereexamInedIndependently
for patterns in first-day movementamongnestmates.
By plotting the original 't samplesseparatelyfor each nestand
calculatingthe meanvector (m) and its length (r), the length of the
meanvectoris usedas an index of angulardispersionamongnestmates(Zar 1984).Sincethe samplesize (n) within nestsis small,
we arbitrarily definednestswith r > 0.6 as "directed." If the polar
angle ("') of the meanvectoris in the direction of the nearestwater
(fJ), so that
[2] IfJ-"'I < 25°
.productive,
and r > .O.?, then hat~hlmg ~ovement on the firs,t,d.aY after emergence WIthIn the nest IS considered to have been dtrected toward
th
t
e neares
t
wa
"
er.
To assessthe straightnessof anindividual's courseamongdays,
the resultantvector (R") and correspondingangle ('t") of each
day's trek were plotted. If the range (>')(i.e., the smallestarc containing all data in the distribution)was less than or equalto 22.5°
(an arbitrarily chosenlimit), the turtle is consideredto have followed a consistentbearingamongdays. Sinceseveralturtles were
stationarybetweendays,the first and seconddays of travel do not
necessarilycorrespondto the first and seconddays after release.
Results
Nesting and emergence
Fresh E. blandingii nests were protected betweenJune 15 and
trailed
(m)
water (deg.)
(r)
("')
84
289
230
240
230
225
0.30
0.90
0.18
0.82
0.89
0.35
280
356
180
317
253
114
180
0 35
.
190
0.38
0.18
-009
0.39
0.76
0.33
0.67
141
191
1994
1a
2b
7a
8b
9c
7
8
5
5
2
7
13a
4.80
-150.00
7.10
14.80
6.20
8.50
6
15.20
1995
1a
2a
6
7a
IOC
11a
12C
3
7
1
2
2
9
4
34.95
11.40
10.05
-150.00
11.00
16.01
15.60
172
276
222
289
101
187
228
057
098
273
214
Note: Nestswith a meanvectorlength(r) >0.6 areconsidered
to be
directed;nestswith r > 0.6 andwhosemeanvectorangle«(II)is within
250of thenearest
waterareconsidered
to exhibitdirectedness
toward
thenearest
water.
aNodirectmovement.
bDirected
movement;
not towardwater.
'Directedmovement
towardthenearest
water.
June 25, 1994, and between June 16 and June 29, 1995.
Emergence began on September 6, 1994, and September 13,
1995. Mean incubation time (number of days from oviposition to emergence) was 94.8 days in 1994 and 90.1 days in
1995 H t hI.
t.
dt
tu all fr
th
...a
c mgs c?n mue 0 emerge na r y om e nest
until mId-Octobe~ m both y~ars. The l~test .recorded ~atural
emerge~ce for thIS populatIon (excludmg lIve hatchlmgs at
excavatIon) occurred on October 25, 1995. Emergence
within nests was mostly asynchronous in both years, lasting
between 2 and 10 days in 1994 and between 2 and 11 days
in 1995. Live hatchlings were uncovered during nest excavations in October of both years.
In both years 93.3% of successfully protected nests were
that is, they produced at least one live hatchling.
Of
103 hatchlings
all
y
(..
I.e.,
observed
d d) f
unal
e
in 1994,
th
rom
e
80.6%
emerged
nes
,
.10
natur-
1.
t 12 6 07
were
a Ive
w h en
.
excavated, a~d 6.8% had successfully hatched but ?Ied
before emergmg from the nest. In 1995, of all hatchlIngs
observed (n = 119),83.2% emerged naturally, 14.3% were
alive when excavated, and 2.5% were dead upon excavation.
Hatchlings' body sizes were not significantly different
between years (1994: mean CL = 33.1 mm; SD = 1.6 mm;
n = 88; 1995: mean CL = 33.5 mm; SD = 1.9 mm; n =
100; 0.0587 < (0.05.(2),186,
P = 0.05).
Of the six beach nests sampled for hatchling movement in
1994, one faced northeast (azimuth 84°), one due south, and
four had a southwest aspect (azimuth 230°,240°,230°,
and
225°); the mean distance between beach nests and the
@ 1997 NRC
Canada
-
I
1390
Can. J. Zool. Vol. 75, 1997
Fig. 1. Movementof 40 E. blandingii hatchlingsin relationto the nearestwater to the nest in 1994. Each point on the unit circle
representsthe angle (v) (azimuthfrom north) of the resultantvector (R) of eachhatchling's first-day trajectory, which has been
convertedto the new polar angle (vi) (azimuthin relationto water)so that the direction to the nearestwater is standardizedamong
nests.The arrow is the meanvector (m') of the sample;its length (r') is 0.234 and angle (1/>')is 64.70 clockwise from the water. At a
significancelevel (a) of 0.05, we acceptthe null hypothesisthat the initial movementis randomwith respectto the nearestwater.
WATER
t
t
/
n = 40 hatchlings
r' = 0.234; <\>'
= 64.70
u = 0.893; a = 0.05
vegetatedbeachedgewas2.71 m (SD = 1.87 m; range =
1.0-4.94 m; n = 6), and at emergencethe meandistance
betweennests and the nearestwater was 9.43 m (SD =
4.48 m; range = 4.8-15.2 m; n = 6). Of the six beachnests
sampledin 1995,threefacedsoutheast(azimuth1010, 187°,
and 172°) and three faced southwest(azimuth222°, 228°,
and 276°); the meandistancebetweenbeachnestsand the
forestedgewas2.88 m (SD = 1.47m; range = 1.7 -5.57 m;
n = 6), and at emergencethe meandistancebetweenbeach
nestsand the nearestwater was 16.5 m (SD = 9.37 m;
range = 11.0- 34.95m; n = 6). The oneroadsidenestsampled in eachyear was approximately7 m from the forested
incline and between100and 150 m from the nearestwater.
In 1994,42 turtles from sevennestswere powder-marked
and trailed for a maximumof 4 days in 1994. Thirty-six
hatchlings from sevennestswere powder-markedin 1995
and tracked for a maximumof 11 days. In both 1994and
1995,hatchling orientationon the first day after emergence
from the nestwas randomwith respectto the nearestwater
(a = 0.05) (Figs. 1 and 2).
Nestmatesfrom three nests sampledin 1994 displayed
directed movementon the first day (Table 1). In hatchlings
from only one nestwas movementconsideredto be directed
towardthe nearestwater, althoughneitherhatchlingactually
enteredthe water. Hatchlings from two other nestsshowed
directedmovementon the first day, but not toward water. In
the first case,hatchlingsfrom an inland nestinitially moved
down the sloped road shoulder. On the following days,
hatchlingsclimbed back up the slopeto cross the road or
climbed into the vegetatedincline. In the secondcasehatchlings movedalmostparallelto the water line, a direction that
did not follow the topographyor lead toward vegetation.
Most hatchlingsfrom this nestwere severelybitten by ants
upon emergenceand it is possible that they were moving
away from the anthill.
Resultswere similar in 1995. For five nests,no directed
movementwas shownamongnestmateson the first day of
travel (Table 1); however,in two nestsnestmatesdisplayed
directedmovementtoward water, although,as in 1994,most
hatchlingsthat movedtoward water did not enter it on the
first day.
Of 27 turtles for which multiday dataare available,only
@ 1997 NRC Canada
Standinget al.
1391
Fig. 2. Movementof 28 E. blandingii hatchlingsin relation to the nearestwater to the nest in 1995. Each point on the unit circle
representsthe angle (v) (azimuth from north) of the resultantvector (R) of eachhatchling's first day trajectory, which has been
convertedto the new polar angle (V') (azimuth in relation to water)so that the direction to the nearestwater is standardizedamong
nests.The arrow is the mean vector (m') of the sample;its length (r') is 0.168 and angle «f>')is 2.50 clockwise from water. At a
significancelevel (a) of 0.05, we acceptthe null hypothesisthat the initial movementis randomwith respectto the nearestwater.
WATER
t
~
t
n = 28 hatchlings
r' = 0.168;<1>'
= 2.50
u = 1.258;a = 0.05
6 followed a consistentbearingbetweenthe first and second
daysof travel(Table2). Of 9 hatchlingsfor whichmore than
2 days of travel are available,only 1 was consistentamong
days (4 days; A = 5.8). This hatchling maintaineda route
almostparallelto the water line and ultimately its trail was
lost at the forest edge.
Trail description
While the interpretationof trendswithin andamongclutches
is important,someof the mostcuriousresultswere revealed
throughclose inspectionof individual trails. For instance,
not only did mosthatchlingsnot maintaina straightcourse
among days, but manytrails within days were convoluted.
Hatchlingsfrequentlychangeddirection, looping and backtracking as well as crisscrossingtheir own trail repeatedly.
Convolutedtrails were restrictedto openareassuchas the
beachand road, althoughnumerousturtles in these areas
navigated along fairly straight routes. Interestingly, all
hatchlings,regardlessof the characterof their trail in the
open, navigatedalong straightcourseswhen under vegetation cover.
Numeroushatchlings in this study avoided water. On
severaloccasions,hatchlingsfollowed straighttrails in the
directionof water, but veeredeither after having reachedit
or within a few metres of the shoreline. Frequentlythese
hatchlingsresumeda courseinto the woods. In 1994, 6 of
42 trails endedin water or < 10 cm from the water's edge;
it is suspectedthat theseturtles enteredwater. Two hatchlings whosetrails werelost> 2 m from waterwere observed
at the shoreline aquatic-basking3 days after their initial
release.Onehatchlingthatenteredwaterreemergedapproximately6 m westof the point of entry and proceededup the
beachtoward the forest. Of 36 turtles tracked in 1995,the
trails of 8 endedin water; of these turtles, 1 enteredwater
on the first day of travel, walked less than 10 m along the
shoreline,and emergedto spendits first night on shore.
@ 1997 NRC Canada
".,
",
1392
~
Can. J. Zool. Vol. 75, 1997
Table 2. The rangeas a measureof dispersionof daily orientation (v) of travel
routes of individual hatcWings.
Range (A)
HatcWing Days I and 2
No.
of travel
Consistentbearing
No. of days
(A < 22.5°):
Over all days
of travel
of travel
(n > 2)*
on first 2 days? overilll?
IB94
IC94
ID94
1E94
IF94
42.7
61.8
171.8
173.4
2. 7
-.;'"~~
IG94
2A94
2B94
2C94
2D94
2E94
2F94
2G94
2H94
8A94
2A95
2B95
2C95
2D95
2E95
2G95
2H95
6A95
I1A95
11C95
I1D95
IIG95
Total
8.3
99.8
27.4
170.1
163.1
99.6
179.6
93.2
90.0
115.2
1.2
31.8
85.6
1.2
140.5
104.8
110.7
152.3
8.5
4.8
61.4
101.6
--*
78.1
--193.2
-180.4
--5.8
35.6
136.2
--Yes
-289.3
233.0
182.4
--Yes
,tt""
:1
'-
~
'"'"
Yes
Yes
,:
3
3
3
4*
3
4
Yes
Yes
6/27
1/9
4*
4
3
-Yes
-*
Note: If the range is not greaterthan22.5°. the hatchlingis consideredto have followed a
consistentbearingamongdays.
*The numberof days of travel is not necessarilythe numberof days after release.since
someindividuals remainedstationaryfor at least I full day. For example,hatchling 2A95
movedon only 4 of the II days on which it wasobserved.
Two hatchlings in 1995 were observed in aquatic overnight forms (sleeping/resting sites) in <3 cm of water. In
addition to aquatic forms, hatchlings used overnight and daytime forms similar to those described by Butler and Graham
Discussion
Many selective advantages of postemergence water-seeking
behaviour in hatchling turtles are presumed; these include the
(1995), burying themselves in leaf litter and vegetation and
nestling in beach cobble, in sphagnum, and under roots and
logs. Fossorial movements were observed in both years.
Individuals moved from a terrestrial form along a route
approximately 10 cm beneath sphagnum and leaf litter. In
1995, several hatchlings spent the night in terrestrial forms,
beneath vegetation and leaf litter, or exposed on the beach,
when temperatures fell below freezing (-5.5°C).
Generally, hatchlings showed no propensity for seeking
cover or for following slope. Infrequently, trails overlapped,
but only for short distances « 2 m). Nestmates released on
the same day showed no tendency to move in the same direction as one another on the first day or subsequentdays. Trails
shown in Fig. 3 are representative of patterns observed in all
nests sampled in this study, though none of these hatchlings
followed a tortuous trail in the open; all were released on the
same afternoon.
avoidance of avian and terrestrial predators, access to suitable basking and feeding habitat, and decreased risk of thermal stress and desiccation. However, few supporting data
are available. For freshwater turtles at high latitudes, water
should be particularly important if hatchlings emerge from
the nest in autumn because it provides insulation against tissue freezing during winter.
Throughout the range of E. blandingii, hatchlings emerge
from the nest in autumn (Butler and Graham 1995; Ernst and
Barbour 1989; Power 1989; Congdon et al. 1983; Bleakney
1963); it is generally assumed that neonates seek aquatic
hibernacula. However, females typically nest inland (Butler
and Graham 1995; Ross and Anderson 1990; Petokas 1986;
Congdon et al. 1983), where the view of water from the nest
site is often obscured by vegetation. Although few detailed
descriptions of nesting sites exist for this species (Butler and
Graham 1995; Power 1989; Petokas 1986), it appears that
@ 1997NRC Canada
;
Standinget a/.
Fig.
on
avoidance.
among
',
3.
the
Early
same
days;
1393
POstemergence
day.
These they
trails
They
patterns
did
showed
not
of
no
Were
orient
8
E
blafldingii
propensity
observed
themselves
to
in
mOst
with
hatchlings
Scent-trail
respect
nests
in
one
in
another
to this vegetation,
study.
a
Nova
SCotia
nor
water,
Population.
to
follow
or
any
slope
other
Nestmates
Or
a
Were
Consistent
beach
feature,
:iiiiiiiiiiiiiiiiiiiiiiiiiiiiii
released
compass
and
at
bearing
there
is
evidence
.Nest
0
CJ]Tree
OVernight
[onn
COVer
Water
5m
::::::::::::::::::::::;
--:::::::::::::::::::;
the nearestWaterbody to inland nestsis often marshor WetEhrenfeld 1968; Ehrenfeldand Carr 1967; Anderson1958)
land and not a large bOdyof openWater(Butler and Graham
upon emergence,since thesecues Would be ineffective in
1995; Power 1989; Petokas1986; Congdonet al. 1983).It
orienting from inland nests.
is unlikely, then, that orientation mechanismssimilar to
Our findings suPPOrt
thoseof Butler and Graham(1995),
thoseof neonatemarineturtlesand beach-nesting
freshwater
who concludedthat slope,openhorizon, and the silhouette
turtleshaveevolvedin E. blandingii. That is, it is unreasonof nearbyvegetationdo not appearto be critical cuesin the
able to assumethatE. blandingiihatchlingsusecuessuchas
orientationof E. blandingii hatchlings(Figs. 1, 2, and 3,
slope,photic gradients,openhorizon,and vegetationsilhouTable 2). Although Butler and Graham (1995) provided
ette (Lohmannand Lohmann 1996; Godfrey and Barreto
limited evidenceof an olfactory componentin orientation,
1995;Salmonet al. 1992; Witheringtonand Bjorndal1991;
that simUltaneouslyreleasednestmatestended to disperse
@ 1997 NRC Canada
the
nest
site
within
or
of
Water
1394
ICan.
randomly
fixate
(Table
2, Fig.
on an olfactory
structural
visual
feature.
cues
while
(personal
the orientation
open,
migration
vegetation
possibility
mechanism
ment efficiency
investigation.
Regardless
of the
actual
movement
random
with
respect
wandering
is adaptive,
newly
emerged
able to find
water
with
"seek"
water,
a way
that they
Although
finding
to
water"
our results
their
conclusion
in their
own
conclusion.
and
For
of
tracked
"necessarily
water"
standing
.in order
Butler
are also
water,"
the
from
used
Graham 1995).
In our study,
...ery
hat~hlmgs dIsplayed
respect
to water
entere?
water,
McNe~1
(1996,
hatchlmgs
dry
(FIgS.
others
pool
1 and 2);
overtly
see footnote
released
vernal
for
(Butler
several
and
wIth
turtles
3). Data
that newly
s edge are repelled
from
emerged
vanation
of e~ergent
to dIsperse
~atchlmgs,
m both
as well
as t~e tendency
the Nova.ScotIa
in
distribution
from
et al.
Understanding
in
and spring
of this
part
and
presents
because
in part
interesting
man-
of
its
small
because
of
its behavioural
populations
elsewhere
1995;
Power
et al.
hatchling
may be another
population.
behaviour
size
and
in the species' range
1994;
Power
1989).
in this threatened
popu1a-
tion is a conservation
priority.
It is hoped that the protection
of younger
age-classes,
particularly
hatchlings,
will bolster
recruitment
vival
into the adult population.
and associated
E. blandingii
Scotia
through
this habitat
is important
Scotia.
terrestrial
overwintering
greater
detail
because
survivorship,
habitat
of
in Nova
surrounding
Not only do hatchlings
nesting
need to move
in search of hi berna cuI a, but the integrity
may affect the efficiency
identify
of nestto the sur-
Additionally,
to hatchling
that future
work
on this
of their travel.
population
sites used by hatchlings,
the behaviour
of young
our understanding
conservation
The protection
waterways
in Nova
importance
and elsewhere,
Scotia
to
and study
age-classes.
of the species
needs of the Nova
We
attempt
This
in
will
and the particular
population.
Acknowledgements
We
gratefully
World
of
an~ Mas~a-
acknowledge
Wildlife
Fund
Fund;
Acadia
24.55 Fund; Nova
Department
Program;
support
by water.
Our results i~ply
not o.nly that wa~er may not be the go.al but
als~ t~at no sImple habItat fe~ture IS sought. The behavioural
nestmates
hatch-
terrestrial
be accomplished
emergence
of E. blandingii
problems,
improve
the
and dis-
and. whI!e
It (FIg.
population
patchy
of withstanding
hatchlings
movement
4),suggest
at the water
with
course
rando?I
avoIded
strategy
recommend
in
~abitat,
habItat
Delayed
This
has not been
blandingii
since ice scouring
and November.
standing
"repeat-
a straight
pool
Scotia,
facet of the bet-hedging
agement
but this
if E.
destroy overwintering
nests. Nevertheless,
are uncovered
in Nova Scotia nests in late
of the vegetation
some of the
of
flooding
would
live hatchlings
Vol. 75, 1997
that this could
in Nova
they fail
habitat]
majority
vernal
proportIonately
(1995)
hatchlings
[wetland
it is unlikely
the nest cavity
October
nest,
Even
capable
areas must be protected.
not in concordance
deviated
to encounter
in such
'seek
In fact,
are physiologically
its potential
as possible.
and Graham's
although
sought
be
in the
in the literature.
ing habitat
if neonates
open horizon,
the nest.
instance,
non-randomly
absence
be organized
as directly
that hatchlings'
report
should
should
toward
from
behaviour
Thus,
goal
or the
so as to avoid
E. blandingii
lings
divergence
(Herman
selection
to have
to organize
their
efficiency.
support
upon emergence
results
edly
their
further
mechanism,
If water-seeking
movements
reach
the move-
warrants
natural
animals
neonate
of a lack of orientation
support
this
their
The
environment
relative
that
in the
orientation
1991).
suggests
impaired
hatchlings
to their
(Able
is more
cover and
influences
overwintering
hibernation,
its effectiveness.
structure
cues, one would
expect
mechanisms
that allow
use
was also noted
This
hatchlings
substantiated
do not seek or
is not totally
of E. blandingii
primarily
individuals
vegetation
may improve
not
the nest.
Such behaviour
that the vegetation
relevant
favoured
from
communication).
cover
do
on a particular
that hatchlings
in the open.
Butler
than
the movement
of
when they are under
convoluted
by B.O.
initial
that hatchlings
more
that hatchlings
our data suggest
avoid
vegetation,
directed
(straight)
more
any
We postulate
in their
Although
3) implies
cue
J. Zool.
the
ford,
Canada.
S.
individuals
following:
T.
Recov-
Canada
for
Species
E.
Whidden,
Muntz,
and
data.
assistance
R.
D.
T-shirt
also supplied
historical
their
Article
Scotia
Nicholas,
Whynot.
logistic
We
thank
in gathering
E.
Alcorn,
R.
Charlton,
C.
P. Hope, R. MacFarlane,
J. McNeil,
Mockford,
Turliuk,
the
Species
Endangered
Parks
and access to unpublished
following
from
University
Faculty
Association
Scotia Liquor
Commission/Nova
of the Environment
Parks
field
data:
T. Gilbertson,
T.
funding
Canada-Endangered
Drysdale,
L. MockT.
We
Ramey,
also
thank
chusetts populatIons,
~uggest that .hatchl~ngs .of .thiS specIes
may have a bet-hedgmg
or adaptIve
com-fllppmg
strategy
S. Bondrup-Nielsen,
J .S. Boates, P. Taylor,
O. Kukal,
and
the anonymous
reviewers
for their suggestions
and criticism
upon emergence.
(Cooper a~d Kaplan 1982; Cooper
In an unpredIctable
e~vIro~ent
such as Nova
of the manuscript;
D. Ryan for his statistical
design and analysis;
and B.O.
Butler and G.
where
the onset and seventy
hatchling
October,
emergence
a mixed
of wmter
is delayed
strategy
would
vary
until
among
198~).
ScotIa,
years and
mid-September
ensure
their
insight
expertise
Packard
in
for
and communications.
and
that some offspring
References
survive.
(1996, seeThat
footenote
numerous
4) studies
hatchlings
survived in overnight
our and atMcNeil's
temper-
Abl~, K.P. 1991. Common themes and VarIations m arumal onenta-
atures
limited
tlon systems. Am. Zool. 31: !57-l67.
Anderso~, P.K. 1958: The photIC responses and water-approach
behavIor of hatchlmg turtles. CopeIa, 1958: 211-215.
Batschelet, E. 1965. Statistical methods for the analysis of problems
in animal orientation and certain biological rhythms. The Ameri-
below freezing
(-5.5°C)
suggests that, at least to a
extent
hatchlings
in the Nova Scotia population
of
...'
E. blan~mg~l
are ab~e to supercoo~ (G. Packard,
personal
co~mumcatIon).
ThIS; coupled
wIth the obse.rve.d. water
avoIdance,
and
suggest that while
hibernacula,
Congdon
fossonal
movements
some hatchlings
terrestrial
et al.
hibernation
(1983)
provide
of
some
may overwinter
mdividuals,
can Institute
in aquatic
may also be possible.
anecdotal
evidence
65-26076,
of Biological
Washington,
Batschelet, E. 1981. Circular
of
Sciences,
Library
of
Congress
D.C.
statistics in biology.
Academic
Press,
New York.
@ 1997 NRC Canada
i
.,.,
Standing et al.
1395
Bleakney,J.S. 1958. A zoogeographicstudyof the amphibiansand
reptiles of easternCanadaBull. No. 155. Departmentof Northern Affairs and National Resources,Ottawa.
Bleakney,J.S. 1963.Notes on the distributionand life histories of
turtles in Nova Scotia. Can. Field-Nat. 77: 67-76.
Brown, J.R. 1927. A Blanding's turtle lays its eggs. Can. FieldNat. 41: 185.
Butler, B.O., and Graham,T.E. 1993. Tracking hatchling Blanding's turtles with fluorescentpigments. Herpetol. Rev. 24:
21-22.
Butler, B.O., and Graham,T.E. 1995. Early post-emergent
behavior and habitatselectionin hatchlingBlanding'sturtles,Emydoidea
blandingii,inMassachusetts.ChelonianConserv.BioI. 1: 187196.
Congdon,J.D., Tinkle, D.W., Breitenbach,G.L., and van Loben
Lohmann, K.J., and Lohmann, C.M.F. 1996. Orientation and
open-seanavigationin seaturtles. J. Exp. BioI. 199: 73-81.
MacCulloch, R.D., and Weller, W.F. 1988. Some aspects of
reproductionin a Lake Erie population of Blanding's turtle,
Emydoideablandingii. Can. J. Zool. 66: 2317-2319.
Pappas,M.J., and Brecke,B.J. 1992. Habitat selectionof juvenile
Blanding's turtles, Emydoidea blandingii. J. Herpetol. 26:
233-234.
Petokas,P.J. 1986. Patterns of reproductionand growth in the
freshwater turtle Emydoideablandingii. Ph.D. thesis, State
University of New York (SUNY), Binghamton.
Power, T.D., Herman, T.B., and Kerekes,J. 1994. Water colour
as a predictor of local distribution of Blanding's turtle,
Emydoideablandingii, in Nova Scotia. Can. Field-Nat. 108:
17-21.
Sels, R.C. 1983. Nesting ecology and hatching success in the
turtle Emydoidea blandingii. Herpetologica, 39: 417-429.
Power, T.D. 1989. Seasonal movements and nesting ecology of a
relict population of Blanding's turtle (Emydoidea blandingii
Cooper,W.S. 1981. Natural decisiontheory: a generalformalism
for the analysisof evolved characteristics.J. Theor. BioI. 92:
401-415.
Cooper,W.S., and Kaplan,R.H. 1982.Adaptive "coin-flipping":
(Holbrook» in Nova Scotia. M.Sc. thesis,Acadia University,
Wolfville, N.S.
Roland,A.E. 1945.The flora of Nova Scotia.Proc. N.S. Inst. Sci.
21(3).
a decision-theoretic examination of natural selection for random
individual variation. J. Theor. BioI. 94: 135 -151.
DePari, J.A., Linck, M.H., and Graham,T.E. 1987. Clutch size
of the Blanding's turtle, Emydoidea blandingii, in Massachusetts.Can. Field-Nat. 101: 440-442.
Ehrenfeld,D. W. 1968.The role of vision in the sea-findingorientation of the green turtle (Chelonia mydas). 2. Orientation
mechanismand range of spectralsensitivity. Anim. Behav. 16:
281-287.
Ehrenfeld, D.W. 1979. Behaviour associatedwith nesting. In
Turtles: perspectivesand research.John Wiley and Sons, New
York. pp. 417-434.
Ehrenfeld,D.W., and Carr, A. 1967.The role of vision in the sea
finding orientationof the greenturtle (Cheloniamydas).Anim.
Behav. 15: 25-36.
Ernst, C.H., and Barbour,R. W. 1989.Turtles of the world. Edited
by N. Dutro. SmithsonianInstitution Press,Washington,D.C.
Godfrey, M.H., and Barreto, R. 1995. Beachvegetationand sea
finding orientation of turtle hatchlings. BioI. Conserv. 74:
29-32.
Herman, T.B., Power, T.D., and Eaton, B.R. 1995. Status of
Blanding's turtles, Emydoidea blandingii, in Nova Scotia,
Canada.Can. Field-Nat. 109: 182-191.
Iverson,J.B. 1990.Patternsof survivorshipin turtles (order Testudines). Can. J. Zool. 69: 385-391.
Ross, D.A., and Anderson, R.K. 1990. Habitat use, movement,
and nesting of Emydoidea blandingii in central Wisconsin.
J. Herpetol. 24: 6-12.
Salmon,M., Wyneken,J., Fritz, E., and Lucas,M. 1992.Seafinding by hatchling sea turtles: role of brightness,silhouetteand
beachslopeas orientationcues. Behaviour,122: 56-77.
Snyder,L.L. 1921. Someobservationson Blanding's turtle. Can.
Field-Nat. 35: 17-18.
Standing,K.L., Morrison, I.P., and Herman,T.B. 1995.Notching
schemefor hatchlingBlanding's turtles in Kejimkujik National
Park. On file at the Centre for Wildlife and ConservationBiology, Acadia University, Wolfville, N.S.
Stapp, P., Young, J.K., VandeWoude,S., and Van Horne, B.
1994. An evaluationof the pathologicaleffects of fluorescent
powder on deer mice (Peromyscusmaniculatus).J. Mammal.
75: 704-709.
Witherington, B.E.,. and. Bjorndal, K:A. 1991. Influences .of
wavelengthand mtenslty on hatchlmg sea .turtle phototaxIs:
implications for sea-finding behavior. Copela, 1991: 10601069.
Zar, J.H. 1984. Biostatistical analysis. 2nd ed. Prentice Hall,
Englewood Cliffs, N.J.
@ 1997 NRC
Canada