1
Pattern of Nipple Use by Puppies: A Comparison of the Dingo (Canis
2
dingo) and the Domestic Dog (Canis familiaris)
3
Robyn Hudson1, Heiko G. Rödel2, Marise Trejo Elizalde1, Lourdes Arteaga3, Gerard
4
Kennedy4, Bradley Smith5
5 6
1
7
Distrito Federal
8
2
9
13, Sorbonne Paris Cité, France
Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México,
Laboratoire d’Ethologie Expérimentale et Comparée E.A. 4443 (LEEC), Université Paris
10
3
11
Mexico
12
4
13
Australia
14
5
Centro Tlaxcala de Biología de la Conducta, Universidad Autónoma de Tlaxcala, Tlaxcala,
School of Psychology, Counselling and Psychotherapy. Cairnmillar Institute, Victoria,
Central Queensland University, School of Human Health and Social Sciences, Australia
15 16
We thank Lyn Watson and Lyn Whitworth from the Dingo Discovery and Research
17
Sanctuary, and Carolina Rojas for technical and bibliographical assistance. Financial support was
18
provided by Central Queensland University, and the Australian Dingo Foundation. M.T.E received
19
a post-graduate fellowship (294591) from the Mexican national funding agency CONACYT. A
20
travel grant was provided to R.H. by the Université Paris 13, Sorbonne Paris Cité, France.
21 22
Correspondence:
23
Robyn Hudson, Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de
24
México, A.P. 70228, Ciudad Universitaria C.P. 04510, Distrito Federal, Mexico
25
Ph. +52 55 5622 3828; Fax. +52 55 50044
26
Email:
[email protected] 1
27 28
Bradley Smith, Central Queensland University, School of Human Health and Social Sciences,
29
North Bruce Highway, Rockhampton, Queensland, Australia, 4702. Ph. + 61 7 4923 2813;
30
Fax. +61 7 4930 6460.
31
Email:
[email protected]
32 33
Email addresses of further authors:
34
Heiko G. Rödel:
[email protected]
35
Marise Trejo Elizalde:
[email protected]
36
Lourdes Arteaga:
[email protected]
37
Gerard Kennedy:
[email protected]
38 39 40 41
2
42
Abstract
43
Surprisingly little information is available about the behaviour of newborn mammals in the
44
functionally vital context of suckling. We have previously reported notable differences in the
45
pattern of nipple use by kittens of the domestic cat and puppies of the domestic dog. Whereas
46
kittens rapidly develop a “teat order”, with each individual using principally one or two
47
particular nipples, puppies show no such pattern. We asked whether the more “chaotic”
48
behaviour seen in puppies of the domestic dog (Canis familiaris) could be the result of
49
relaxed selection due to domestication. In a first test of this hypothesis we studied suckling
50
behaviour in four litters of wild-type captive dingoes (Canis dingo), a canid species that has
51
inhabited the Australian mainland in substantial numbers for at least 5,000 years with
52
minimal human influence. On all measures of individual puppies’ behaviour – time spent
53
attached to nipples, lack of individual use of particular nipples and consequent absence of a
54
“teat order”, lack of synchronized suckling with other littermates, lack of agonistic behaviour
55
– we found no differences between the two species. In conclusion, we suggest that the
56
difference between the pattern of suckling behaviour of kittens of the domestic cat (and other
57
felids) and the domestic dog is not an artifact of domestication, but rather reflects
58
phylogenetic differences between felids and canids as a consequence of their different
59
lifestyles. These findings emphasize the need for comparative studies to avoid simplistic
60
generalizations from one or two species across broad taxonomic groups.
61 62
Key words: Suckling behaviour, “teat order”, sibling competition, Canis dingo, Canis
63
familiaris
64
3
65
All mammals depend for their early growth and survival on the mother’s milk and
66
obtaining this in sufficient quantities represents one of life’s earliest challenges. In particular,
67
most altricial young, blind at birth and often with limited motor abilities, must locate, attach
68
to and suck nipples without direct maternal assistance. In addition, as milk is energetically
69
costly for mothers to produce (Loudon & Racey, 1987; Speakman, 2008), it usually
70
represents a limiting resource for which littermates must compete (Drummond, Vázquez,
71
Sánchez-Cólon, Martínez-Gómez, & Hudson, 2000; Drummond, 2006). Despite the
72
importance of suckling, little information is available and then only for a few species on the
73
behaviour of altricial young in this vital context (Drummond, 2006; Hudson & Trillmich,
74
2008; Mock & Parker, 1997). Studying this behaviour, even in captivity, is notoriously
75
difficult; altricial young are usually hidden in dens, nests or pouches, human interference may
76
lead to them being abandoned, and they are often vigorously defended by well-armed
77
mothers or other caretakers.
78
For some years we have been studying mother-offspring interactions in the domestic
79
cat (Felis silvestris catus) and domestic dog (Canis familiaris). Consistent with other reports
80
in the literature (domestic cat: Ewer, 1959; Rosenblatt, 1971; puma Puma concolor: Pfeifer,
81
1980; snow leopard Panthera uncia: McVittie, 1978; Eurasian lynx Lynx lynx: Glukhova &
82
Naidenko, 2014), we found that newborn kittens rapidly develop a clear order in the
83
individual use of particular nipples, which they vigorously defend against encroaching
84
littermates (Hudson, Raihani, González, Bautista, & Distel, 2009; Hudson & Distel, 2013;
85
Hudson, 2014; Raihani, González, Arteaga, & Hudson, 2009). In contrast, and also in accord
86
with a previous report (Rheingold, 1963), we found that puppies of the domestic dog do not
87
do this, and that their pattern of suckling differs in several respects to the cat. Puppies of
88
various breeds and under various conditions of maintenance, showed no clear order in
89
individual nipple use, they showed no agonistic behaviour during suckling, and in contrast to 4
90
kittens which usually all attach to nipples in coordinated suckling bouts, puppies often
91
attached in a seemingly haphazard manner, with some suckling while others continued
92
searching for nipples, moved around the nest area, slept or played (Arteaga, Rödel, Trejo
93
Elizalde, González, & Hudson, 2013).
94
According to the literature, free ranging domestic dogs differ notably from wild
95
canids in relation to the quality and frequency of reproductive activity (seasonality, age of
96
first reproduction, frequency of oestrus, pair bonding) and parental feeding behaviours
97
(nursing, provisioning the mother and young by the male or other pack members) (Boitani,
98
Ciucci, & Andreoli, 1995; review by Lord, Feinstein, Smith, & Coppinger, 2013). The
99
authors argue that these differences reflect the domestic dog’s adaptation to the human niche
100
and access to readily available resources, which has subsequently relaxed the selection
101
pressure for extended and bi-parental care. We speculated (Arteaga et al., 2013) that the
102
seemingly less coordinated suckling behaviour in the dog might have been the result of
103
relaxed selection during this species’ much longer and closer association with humans than
104
has been the case for the domestic cat. Whereas archaeological evidence suggests that the
105
presumed wolf (Canis lupus) ancestor of the domestic dog started its association with human
106
hunter-gatherer societies 10,000 or more years ago (Clutton-Brock, 1995; Miklósi, 2007;
107
Shannon et al., 2015; Wang et al., 2015), the presumed ancestor of the domestic cat (Felis
108
lybica) is thought to have done so about 4,000 years ago with the advent of agriculture and
109
the proliferation of sedentary human communities (Serpell, 2014). Further evidence for the
110
longer history of domestication of the dog than of the cat is the fact that dogs are no longer
111
able to form reproductively self-sustaining populations independent of human communities
112
(Boitani et al., 1995), whereas domestic cats readily do so, sometimes to the point of posing a
113
threat to native wildlife, particularly of island ecologies (Fitzgerald & Turner, 2000). In order
114
to decide whether observed differences in the pattern of suckling behaviour between kittens 5
115
and puppies are due to different degrees of domestication, or rather represent phylogenetic
116
differences in the organization of this vital aspect of mother-young behaviour, it would be
117
necessary to investigate the pattern of suckling in non-domesticated, taxonomically close
118
relatives of the dog.
119
In a first step, it was our aim in the present study to investigate suckling behaviour in
120
puppies of the Australian wild dog or dingo Canis dingo (International Commission on
121
Zoological Nomenclature, 1957; Crowther, Fillios, Colman, & Letnic, 2014) during the early
122
den period, and to compare this with our previous findings in the domestic dog.
123
Although the precise origin of the dingo is contentious, it is likely descended from an
124
ancient population of a south-east Asian canid, and arrived on the Australian continent
125
around 5,000+ years ago (Smith & Savolainen, 2015). Since this time, the dingo has lived
126
independent of human influence, and from other dog or wolf populations, making it unique
127
among the world’s canids. The dingo inhabits most of the Australian mainland in appreciable
128
numbers, and as a top order predator has a profound influence on the structure and function
129
of Australian ecosystems (Glen, Dickman, Soulé, & Mackey, 2007; Newsome, Ballard,
130
Crowther, Dellinger, Fleming et al., 2015). Although the social organisation of dingoes is
131
highly variable (according to habitat, prey, climate and human activity), in general, they are a
132
highly social species that live in a family or social group, maintain a distinct territory, and
133
hunt cooperatively (Smith, 2015). They are seasonal breeders, giving birth to one litter of
134
typically 3 to 6 puppies a year during the winter months (May–August). The puppies are born
135
into a den where the mother remains for extended periods during the first postnatal weeks,
136
and where she is provisioned by her mate (Lord et al., 2013; Smith, 2015). Thus, and
137
importantly for the purpose of our study, the dingo’s pattern of reproduction and parental care
138
is more similar to that of its presumed ancestor the wolf, than is that of the domestic dog
6
139
(Boitani et al., 1995; Corbett, 2001; Mech & Boitani, 2003; Miklósi, 2007; Thompson, 1992;
140
Thompson, Rose, & Kok, 1992).
7
141 142 143
Methods
Animals, Housing Conditions and Sample Sizes
144
Dingo: Twelve dingo puppies from four litters (born in May of 2013) participated in
145
this study (Table 1). All parents were captive born and raised at the Dingo Discovery Centre
146
and Research Sanctuary (DDC), a privately run facility located in Victoria, Australia, and
147
passed a genetic analysis of purity conducted by the University of New South Wales,
148
Australia (Wilton, 2001). Breeding pairs were housed in enclosures (9 m x 1.5 m or 13.5 m2)
149
comprising 7 m of flooring made of quarry rubble (gravel) and a 2 m section of concrete
150
where a wooden whelping box (940 mm L x 635 mm W x 860 mm H) was positioned.
151
Shredded paper was placed inside the whelping box and was replaced daily or as required. A
152
corrugated iron roof covered 2 m of the enclosure above the concrete section, and the
153
remaining roof was covered with wire mesh to prevent escape. Enclosure walls were 2 m
154
high. All enclosure walls were constructed with wire mesh, with the addition of 1.2 m high
155
sheets of corrugated iron between enclosures to act as a windbreak and to prevent fighting
156
between neighbours. Access to one of seven large grassed exercise yards that included natural
157
dens, large hollow logs, trees, rocky areas and enrichment objects was provided daily. After
158
whelping, access to these yards was limited, and occurred on an ad-hoc basis.
159 160
--- Insert Table 1 about here ---
161 162
The mothers were accustomed to human handling and presence, and appeared well
163
adjusted to the study conditions. Socialization of the puppies began at 1–3 days of age, with
164
puppies handled gently for brief periods by two of the staff members (including weighing),
165
until they were about 4–5 weeks of age. Adult dingoes were fed a combination of dry dog 8
166
food and fresh chicken (carcass or ground carcass mince). The diet of the mother was
167
changed 14 days prior to the estimated whelping date to a higher energy food, and mothers
168
were fed twice a day (morning and evening). This continued until the puppies started to be
169
weaned to semi-solid food when their teeth had fully erupted at approximately 21 days of age
170
(at the discretion of the sanctuary staff). The food consisted of dry puppy food and chicken
171
mince soaked in warm water. Puppies were fed twice a day (morning and evening) separately
172
from the adults.
173
Dog: Data from 42 puppies from nine mothers, reanalysed from a previous study
174
(details of mothers, puppies and methods in Arteaga et al., 2013) contributed to the present
175
report. Briefly, the animals were pets kept in different private homes, they were of various
176
breeds, ages and parity, and were fed a variety of diets and maintained under a variety of
177
conditions.
178 179 180
Procedures Dingo: Den behaviours were recorded continuously using a small surveillance camera
181
(Swann Pro 530) and a digital video recorder (Swann 4 channel DVR, model DVR4-4000).
182
Video recordings were made at D1 resolution, at a frame rate of 20 fps, and max bit rate 1024
183
kbps. Cameras were installed in the top corner of four dens at least 24 hours prior to birth.
184
Recording began at least 24 hours prior to birth, and ceased when the puppies started
185
spending more time outside the den box (or at 21 days, whatever occurred first). This resulted
186
in continuous post partum footage totalling between 17 and 21 days except for a total of 5
187
days of lost footage due to power outages and destruction of camera equipment by dingoes.
188
Individual puppies could be distinguished by distinctive coat markings. As in our previous
189
report on suckling behaviour in the domestic dog (Arteaga et al., 2013), the main behaviours 9
190
scored were the frequency and time each puppy spent attached to which nipple. Like most
191
dogs, dingoes typically have four pairs of nipples, which we labelled from anterior to
192
posterior as rows 1–4, and whether the nipples were on the left or on the right side of the
193
mother’s ventrum. We also noted any incidents of apparent contest behaviour such as
194
unattached puppies pushing with their head and muzzle against the muzzle of attached
195
littermates apparently trying to dislodge them (cf. Hudson et al., 2009; Hudson & Distel,
196
2013 for kittens). However, as no such behaviours were recorded (see also results of the
197
inter-observer reliability check below) this is not further considered.
198
Dog: A detailed report of procedures is given in Arteaga et al. (2013). Briefly, so as
199
not to disturb mothers and at the request of some owners, we did not attempt to alter or
200
standardize rearing conditions in any way. Soon after birth, puppies were fitted with a
201
coloured neck ribbon for individual identification, and where possible litters were filmed
202
every second or third day during a single nursing session. For this, mothers were separated
203
from their litters an hour before recording to increase their motivation to nurse and of the
204
puppies to suckle. This was generally effective, and nursing was then filmed for 30 min or
205
until the mother left the litter for more than 3 min, which ever was first. This resulted in the
206
recording of an average of 8.1 sessions per litter. The behaviours scored were the same as
207
those described for the dingo above.
208 209 210
Ethics Note Dingo mothers gave birth with no natal deaths, raised their young to weaning without
211
apparent difficulty, and allowed observation and manipulation of their puppies by sanctuary
212
staff without protest. Puppies were weighed approximately every third day from birth to
213
check weight gain as an indicator of general health. All puppies survived to the end of the 10
214
study (with the exception of the two smallest puppies in Opal’s litter which were euthanized
215
by the owner on postnatal day 1, and a puppy in Freckle’s litter which disappeared from
216
unknown causes on postnatal day 3 but was possibly eaten by the mother), and were sold as
217
pets or to zoos or fauna parks. All methods were approved by the Central Queensland
218
University Animal Ethics Committee, project number A12/11-293. For the dogs, details of
219
treatment of the animals are given in Arteaga et al. (2013). Use of them for this study
220
complied with the guidelines for the care and use of animals in research of the American
221
Psychological Association, and according to the National Guide for the Production, Care and
222
Use of Laboratory Animals, Mexico (Norma Oficial Mexicana NOM-062-200-1999).
223 224
Data Analysis and Sample Sizes
225
Statistical analyses were done with the programme R, version 3.1.1 (R Core Team,
226
2014). The sample sizes available for analysis were 12 dingo puppies from the four litters that
227
could be reliably distinguished by distinctive fur markings, and 42 puppies of the domestic
228
dog from nine litters for which we had sufficient data for comparison with the dingo sample.
229
To enable a direct comparison of findings for the two species we reanalysed the data from
230
Arteaga et al. (2013) so as to correspond to our analysis here for the dingo, and we have
231
modified the previous graphical representations accordingly. However, as reported in Results
232
and Discussion, this did not change the findings or conclusions of our previous study in the
233
dog.
234
First, we aimed to compare different parameters observed during approximately
235
equivalent 30-minute observations between puppies of the dingo and the dog. For this, we
236
calculated average values of the total time spent attached to nipples and the time spent
237
attached to a particular nipple, averaged for each individual puppy over all available nursing 11
238
sessions (32 sessions, always 8 per litter in the dingo; 68 sessions, on average 7.6 in the dog)
239
during the first three postnatal weeks. Comparisons were done with linear mixed-effects
240
models (LMM), based on restricted maximum likelihood estimates using the package lme4
241
(Bates, Maechler, & Bolker, 2014). The identity of the litter was used as a random factor.
242
We then analysed whether and how the cumulative number of nipples used by
243
individual puppies (dependent variable) increased over the first four consecutive nursing
244
sessions (factor with 4 levels) observed during the first two postnatal weeks when suckling
245
was most frequent and sessions that were comparable with those of the previous dog study
246
could be most readily obtained. Generalised linear mixed effects models (GLMM) for
247
Poisson distributed were used, including litter identity and puppy identity as random factors.
248
Furthermore, we included individual-level random effects to account for potential effects of
249
overdispersion (Browne, Subramanian, Jones, & Goldstein, 2005).
250
Third, we explored the distribution of the time and the frequency per nursing session
251
that individual puppies of the two species were attached to mothers’ different nipples. For
252
this, these two variables were plotted versus the ranked attachment time to nipples as well as
253
their frequency of use. Purported preferences in using particular nipples would be evident
254
from a sharp decrease in the duration or in the frequency of attachment to nipples other than
255
to the most used nipple (rank 1, see Fig. 2). Furthermore, we checked whether the (negative)
256
slope of this association differed between puppies of the dingo and dog by testing the
257
interaction between the ranked time or frequency of use (covariates) and species (factor with
258
2 levels). Linear mixed-effects models were used, including litter and puppy identity as
259
random factors.
260
Finally, we tested for correlations across time in puppies’ proportional duration and
261
frequency of nipple use between two time intervals, the first postnatal week and the
262
second/third postnatal weeks. Again, LMM models were used, including litter and puppy 12
263
identity as random factors. For both time intervals and in the dingo as well as in the dog, the
264
duration and the frequency of attachment to mothers’ nipples were adjusted to a normal
265
distribution by log[x+1] transformation. For all associations and comparisons tested, we
266
calculated Nagelkerke’s Pseudo-R2, which can be interpreted as a measure of the proportion
267
of variance explained (Magee, 1990; Nagelkerke, 1991).
268
We checked all models for homogeneity of variances and goodness of fit by plotting
269
residuals versus fitted values. In addition, normality of the model residuals of linear mixed-
270
effects models was assured by checking normal probability plots (Faraway, 2006). P-values
271
were calculated by type III Wald Chi square tests.
272
In an inter-observer reliability check a second observer, experienced in the evaluation
273
of suckling behaviour in kittens and domestic dog puppies but blind to the results obtained
274
for the dingo puppies, re-evaluated a sample of the video material. Because we were
275
interested in possible agonistic behaviour among littermates during suckling as well as in
276
nipple use, we chose to re-evaluate the behaviour in one of the larger litters (mother Petal)
277
where all puppies survived. The behaviour of four puppies (33% of total number of subjects),
278
during 8 sessions (25% of total number of sessions) was re-analyzed. Because of lack of
279
familiarity with the dingo puppies and caution in reliably distinguishing among such
280
similarly-coloured individuals, the second observer only felt confident in registering details
281
for approximately half as many nipple attachments as the first, highly experienced observer.
282
Nevertheless, good correspondence was found between these observations and the
283
corresponding records for the first observer. Thus, neither observer registered any agonistic
284
interactions although explicitly instructed to look for these. The two observers were also in
285
100% agreement on the number of different nipples (4 to 5) each of the puppies suckled. For
286
each nipple attachment they showed 88% agreement on which puppy was attached to which
287
nipple. In all cases in which the two observers differed, the puppy was recorded as being on a 13
288
neighboring nipple, either beside or opposite. Furthermore, an intra-class correlation based on
289
1,000 permutations (library rptR, Schielzeth & Nakagawa, 2013) revealed a high
290
concordance in the attachment times quantified by the 2 observers for each of these events
291
(R2 = 0.939, CI95%[0.806, 0.980], N = 17, P < 0.001).
292
14
293 294 295 296
Results
Comparison of Suckling Behaviour in Dingo and Dog Puppies The total time that individual puppies were attached to nipples during a nursing
297
session, the average number of nipples they used per session as well as the mean time they
298
were attached to a particular nipple did not differ significantly between puppies of the dingo
299
and the domestic dog (Table 2).
300 301
--- Insert Table 2 about here ---
302 303
Moreover, in a high percentage of nursing sessions in both dingoes (15 out of 32;
304
46.9%) and dogs (30 out of 68; 44.1%), not all puppies were attached to nipples but were
305
either apparently searching for a nipple on the mother’s ventrum, moving around the den or
306
nest area, playing with littermates or apparently sleeping. And finally, in dingoes as in dogs,
307
during suckling we did not observe any directly agonistic behaviour or behaviour that could
308
be interpreted as competition for particular nipples.
309 310 311
Cumulative Use of Different Nipples In the dingo (GLMM: 32 = 12.10, PseudoR2 = 0.249, P = 0.007) as well as in the
312
domestic dog ( 32 = 59.85, PseudoR2 = 0.342, P < 0.001), the cumulative number of nipples
313
used by individual puppies increased significantly across the first four nursing sessions
314
recorded during the first and second postnatal week (post hoc comparisons in Fig. 1). By the
315
end of the sampling period all puppies had used several nipples, and some individuals had
316
used all. The slope of this increase did not differ between puppies of the dingo and the dog, as
15
317
indicated by the non-significant interaction of nursing session × species ( 32 = 1.30, PseudoR2 =
318
0.006, P = 0.73).
319 320
--- Insert Fig. 1 about here ---
321 322
Distribution of Individual Nipple Use
323
The percentage of the total time that puppies of the two species spent attached to
324
different nipples as well as their proportional frequency of the use of different nipples,
325
showed right-skewed distributions in ranking, evenly decreasing in all cases (Fig. 2a–c). The
326
negative slopes of these right-skewed distributions did not differ significantly between
327
puppies of the dingo and dog with respect to proportional time that individual puppies were
328
attached to the different nipples (LMM: % time × species: 12 = 0.46, PseudoR2 = 0.003, P =
329
0.50) or to the proportional frequency of nipple use (% frequency × species: 12 = 0.24,
330
PseudoR
331
puppies show greater use of one or two particular nipples, as would be indicated by a sharp
332
right-skewed drop in the staircase function. Indirectly, this also indicates that the puppies
333
showed no particular preference in the use of nipples with respect to location on the female’s
334
ventrum.
2
= 0.001, P = 0.62). Together, these data indicate that in neither species did the
335 336
--- Fig. 2 about here ---
337 338 339
Lack of Correlations Across Time in Nipples Use There was no indication of a significant association across time in nipple use, i.e.
340
between the first postnatal week and the second and third postnatal week, with respect to the
341
percentage of the total time that the puppies of the two species spent attached to different 16
342
nipples (LMM: dingo: 12 = 0.45, PseudoR2 = 0.005, P = 0.50; dog: 12 = 0.41, PseudoR2 = 0.001,
343
P = 0.52), or to their proportional frequency in the use of different nipples (dingo: 12 = 1.29,
344
PseudoR
2
= 0.014, P = 0.26; dog: 12 = 1.94, PseudoR2 = 0.006, P = 0.16).
345
17
346 347 348
Discussion
For all of the behavioural measures considered in the present study the pattern of
349
nipple use by dingo puppies did not differ from the pattern we had previously found in
350
puppies of the domestic dog (Arteaga et al., 2013). During nursing sessions, dingo and dog
351
puppies spent similar amounts of time attached to nipples, they attached to a similar number
352
of different nipples, and they spent similar amounts of time attached to any one nipple before
353
releasing it. As in the dog, dingo puppies did not show a clear order in the use of particular
354
nipples but rather across nursing sessions each puppy used several different nipples.
355
Moreover, in both dingoes and dogs, in close to half of observed nursing session one or more
356
puppies did not attach to nipples, nor appeared motivated to do so as indicated by them
357
moving around the den, sleeping or playing. This is in contrast to the cat in which in our
358
experience the whole litter usually attaches to nipples during a nursing session, and
359
particularly during the first two postnatal weeks (Hudson et al., 2009; Raihani et al., 2009; R.
360
Hudson, own observations). In both dingo and dog puppies, overtly agonistic behaviour
361
during suckling or behaviour that could be interpreted as competition for nipples was never
362
seen. This contrasts with reports of agonistic behaviour among suckling kittens and piglets
363
(Hudson et al., 2009; review in Hudson & Distel, 2013) but is consistent with the apparent
364
lack of such behaviour among wolf pups (Packard, Mech, & Ream, 1992).
365
As for many other canid and felid species, the behaviour of dingoes is extremely
366
difficult to study in the wild. They are highly mobile, occupy large territories, they are wary
367
and difficult to observe in most environments, and the identification of individuals is often
368
difficult. Regarding captive studies, only few sanctuaries in Australia (or the world) breed
369
dingoes, limiting the possibility to obtain large samples (Smith & Watson, 2015). The DDC
370
where the present study was conducted is the largest breeding sanctuary presently in 18
371
existence, but because dingoes (like wolves but in contrast to domestic dogs) only bred once
372
a year, even here obtaining a closely monitored large sample would require a study of several
373
years. Nevertheless, returning to the question originally motivating this study, we may
374
tentatively conclude that the pattern of nipple use by the domestic dog, so similar to the dingo
375
and so different to the domestic cat and other felids, is unlikely to be an artefact of
376
domestication but rather represents a phylogenetic difference. Despite our small sample size,
377
the present findings for the dingo seem to be reliable. The parents were all pure dingoes, that
378
is, genetic analysis did not show any domestic dog alleles in their DNA (Wilton, 2001). They
379
were habituated to human presence and permitted handling of the puppies by familiar
380
caretakers without protest. Nevertheless, the effect of captive conditions such as the provision
381
of food and the presence of neighbouring breeding pairs is not known, and despite the
382
difficulties mentioned above, ultimately the validity of the findings needs to be checked by
383
studies of wild dingoes and their litters in natural dens.
384
However, in support of the validity of the main finding of similarity in the suckling
385
behaviour of puppies of the dingo and domestic dog, and differences compared to the
386
behaviour of kittens of the domestic cat, we note the consistency in reports on cats and other
387
felids tested under various conditions of captivity although also often based on only few
388
litters. For example, highly similar patterns of sucking behaviour in kittens have been
389
reported in studies of animals reared in confined laboratory cages (Rosenblatt, 1971) and in
390
unconfined house cats (Ewer, 1959; Hudson et al., 2009; Raihani et al., 2009), as well as in
391
other felid species kept under a range of captive conditions (Glukhova & Naidenko, 2014;
392
McVittie, 1978; Pfeifer, 1980). Such similarity among felid species, but so different to the
393
domestic dog and now apparently to the dingo, suggest that in general the young of these two
394
taxonomic groups, canids and felids, might indeed have evolved different patterns of
395
behaviour in the nursing context. Although this clearly needs to be confirmed by the 19
396
investigation of additional species from both taxonomic groups, and where possible in the
397
wild, we may already ask how such differences might be explained?
398
As we suggested previously, one possibility might relate to the generally different
399
lifestyles of these two taxonomic groups (Arteaga et al., 2013). As obligate carnivores,
400
usually lone hunters (Bradshaw, 2006; Fitzgerald & Turner, 2000; Sunquist & Sunquist,
401
2002), and dependent on their agility and climbing ability to escape danger, it should be
402
advantageous for feline mothers to maintain a minimum number of active, burdensome
403
mammary glands. This implies, however, that the kittens or cubs should ensure an adequate
404
milk supply by regularly sucking their “own” nipple, thereby preventing its involution (Ewer,
405
1959; R. Hudson, own observations, Kim, Easter & Hurley, 2001 in pigs Sus scrofa; see
406
Hudson & Distel, 2013; Hudson, 2014 for reviews of the possible functional significance of
407
the teat order in kittens and piglets). By contrast, more omnivorous and group- or pair-living
408
canid mothers may have been under less selection pressure to minimize the number of active
409
mammary glands. This may be particularly the case as mothers are able to remain in the den
410
with their litter for long periods due to provisioning with food by their mate or the pack
411
(wolf: Mech, Wolf, & Packard, 1999; Mech & Boitani, 2003; dingo: Corbett, 2001; Smith,
412
2015; Thomson, 1992; red fox Vulpes vulpes: Henry, 1996). Such provisioning, and with
413
apparently similar consequences for nursing/suckling behaviour, is not confined to canids. A
414
similar case is also provided by a rodent, the naked mole-rat Heterocephalus glaber. In this
415
species the breeding female of a colony is provisioned by other colony members, and the
416
young have continuous, non-synchronized and apparently non-competitive access to nipples
417
(Sherman et al., 1999). The young of canid and other mammalian mothers who receive social
418
support (with confirmation from further species certainly needed) are able to sample a range
419
of nipples at any moment and with little or no obstruction from littermates, who at any
420
particular time may not be similarly motivated to suckle (cf. Rheingold, 1963). This would 20
421
imply, however, that the young can extract milk from the often considerably enlarged
422
mammary glands at will. Although milk let-down in the dog, as in other mammals, is
423
stimulated by the release of the hormone oxytocin from the posterior pituitary in response to
424
sucking (Pickford, 1960), to our knowledge nothing is known about the operation of this
425
mechanism in response to the repeated, intermittent and apparently uncoordinated suckling
426
by individual puppies. By contrast, a litter of kittens obtains milk in a single let-down of only
427
a few seconds per nursing episode after several minutes of combined sucking (Hudson et al.,
428
2009), a situation possibly serving rapid milk transfer to the young by a mother who must
429
then soon leave them to resume hunting.
430
Given increasing interest in the ontogeny of individual differences in behavioural
431
phenotypes (Stamps & Groothuis, 2010; Trillmich & Hudson, 2011), including in the
432
potential role of littermates in shaping these (Drummond et al., 2000; Hudson & Trillmich,
433
2008; Hudson, Bautista, Reyes-Meza, Morales Montor, & Rödel, 2011; Bautista, Zepeda,
434
Reyes-Meza, Martínez-Gómez, Rödel & Hudson, 2015), it is important to have an
435
appreciation of the range of behavioural patterns and adaptive specializations in such a
436
functionally vital and ubiquitous context for mammals as nursing by mothers and suckling by
437
the young. There is a clear need for more information on the diversity of behavioural patterns
438
and their physiological underpinnings here.
439 440
21
441
References
442 443
Arteaga, L., Rödel, H. G., Trejo Elizalde, M., González, D. & Hudson, R. (2013). The pattern
444
of nipple use before weaning among littermates of the domestic dog. Ethology, 119,
445
12–19.
446
Bates, D., Maechler, M. & Bolker, B. (2014). lme4: Linear mixed-effects models using Eigen
447
and S4. http://CRAN.R-project.org/package=lme4: R package version 1.1–7.
448
Bautista, A., Zepeda, J. A., Reyes-Meza, V., Martínez-Gómez, M., Rödel, H. G. & Hudson,
449
R. (2015). Contribution of within-litter interactions to individual differences in early
450
postnatal growth in the domestic rabbit. Animal Behaviour, 108, 145–153.
451
Boitani, L., Ciucci, F. P. & Andreli, G. (1995). Population biology and ecology of feral dogs
452
in central Italy. In J. Serpell (Ed.), The domestic dog. (pp. 217–244). Cambridge, UK:
453
Cambridge University Press.
454 455
Bradshaw, J. W. S. (2006). The evolutionary basis for the feeding behavior of domestic dogs (Canis familiaris) and cats (Felis catus). Journal of Nutrition, 136, 1927S–1931S.
456
Browne, W. J., Subramanian, S. V., Jones, K. & Goldstein, H. (2005). Variance partitioning
457
in multilevel logistic models that exhibit overdispersion. Journal of the Royal
458
Statistical Society: Series A, 168, 599–613.
459 460 461 462 463 464
Clutton-Brock, J. (1995). Origin of the dog: domestication and early history. In J. Serpell (Ed.), The domestic dog. (pp. 7–20). Cambridge, UK: Cambridge University Press. Corbett, L. K. (2001). Dingoes in Australia and Asia. 2nd edition. Marlston, South Australia: JB Books. Crowther, M., Fillios, M., Colman, N. & Letnic, M. (2014). An updated description of the Australian dingo (Canis dingo Meyer, 1793). Journal of Zoology, 293, 192–203.
22
465 466
Drummond, H. (2006). Dominance in vertebrate broods and litters. Quarterly Review of Biology, 81, 3–32.
467
Drummond, H., Vázquez, E., Sánchez-Cólon, S., Martínez-Gómez, M. & Hudson, R. (2000).
468
Competition for milk in the domestic rabbit: survivors benefit from littermate deaths.
469
Ethology, 106, 511–526.
470
Ewer, R. F. (1959). Suckling behaviour in kittens. Behaviour, 15, 146–162.
471
Faraway, J. J. (2006). Extending the linear model with R. Generalized linear, mixed effects
472
and nonparametric regression models. Boca Raton, USA: Chapman & Hall.
473
Fitzgerald, B. M. & Turner, D. C. (2000). Hunting behaviour of domestic cats and their
474
impact on prey populations. In D. C. Turner & P. Bateson (Eds.), The domestic cat.
475
The biology of its behaviour. 2nd edition. (pp. 151–175). Cambridge, UK: Cambridge
476
University Press.
477
Glen, A., Dickman, C., Soulé, M. & Mackey, B. (2007). Evaluating the role of the dingo as a
478
trophic regulator in Australian ecosystems. Austral Ecology, 32, 492–501.
479
Glukhova, A. & Naidenko, S. (2014). Suckling behavior in Eurasian lynx (Lynx lynx L.)
480
cubs: characteristics and correlation with competitive interactions. Zoo Biology, 33,
481
388–393.
482 483 484 485 486
Henry, J. D. (1996). Red fox. The catlike canine. Washington D.C., USA: Smithsonian Institution Press. Hudson, R. (2014). Behavioural epiphenomena revisited: reply to Skok and Škorjanc. Ethology, 120, 739–741. Hudson, R. & Trillmich, F. (2008). Sibling competition and cooperation in mammals:
487
challenges, developments and prospects. Behavioral Ecology and Sociobiology, 62,
488
299–307.
23
489
Hudson, R., Raihani, G., González, D., Bautista, A. & Distel, H. (2009). Nipple preference
490
and contests in suckling kittens of the domestic cat are unrelated to presumed nipple
491
quality. Developmental Psychobiology, 51, 322–332.
492
Hudson, R., Bautista, A., Reyes-Meza, V., Morales Montor, J. & Rödel, H. G. (2011). The
493
effect of siblings on early development: a potential contributor to personality
494
differences in mammals. Developmental Psychobiology, 53, 564–574.
495 496 497
Hudson, R. & Distel, H. (2013). Fighting by kittens and piglets during suckling: what does it mean? Ethology, 119, 353–359. International Commission on Zoological Nomencalture (1957). Opinion 451. Use of the
498
plenary powers to secure that the specific name “dingo” Meyer, 1793, as published in
499
the contribution “Canis dingo”, still be the oldest available name of the dingo of
500
Australia (class Mammalia). In F. Hemming (Ed.), Opinions and declarations
501
rendered by the International Commission on Zoological Nomenclature (pp. 329–
502
338). London, UK: International Trust for Zoological Nomenclature.
503
Kim, S. W., Easter, R. A. & Hurley, W. L. (2001). The regression of unsuckled mammary
504
glands during lactation in sows: The influence of lactation stage, dietary nutrients, and
505
litter size. Journal of Animal Science, 79, 2659–2668.
506
Lord, K., Feinstein, M., Smith, B. & Coppinger, R. (2013) Variation in reproductive traits of
507
members of the genus Canis with special attention to the domestic dog (Canis
508
familiaris). Behavioural Processes, 92, 131–142.
509 510 511 512
Loudon, A. S. & Racey, P. A. (1987). Reproductive energetics in mammals. Oxford, UK: Clarendon Press. Magee, L. (1990). R^2 measures based on Wald and likelihood ratio joint significance tests. American Statistician, 44, 250–253.
24
513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533
McVittie, R. (1978). Nursing behavior of snow leopard cubs. Applied Animal Ethology, 4, 159–168. Mech, L. D., Wolf, P. C. & Packard, J. M. (1999). Regurgitative food transfer among wild wolves. Canadian Journal of Zoology, 77, 1192–1195. Mech, L. D. & Boitani, L. (2003). Wolves. Behavior, ecology and conservation. Chicago, USA: University of Chicago Press. Miklósi, Á. (2007). Dog behaviour, evolution, and cognition. Oxford, UK: Oxford University Press. Mock, D. W. & Parker, G. A. (1997). The evolution of sibling rivalry. Oxford, UK: Oxford University Press. Nagelkerke, N. J. D. (1991). A note on a general definition of the coefficient of determination. Biometrika, 78, 691–692. Newsome, T., Ballard, G., Crowther, M., Dellinger, J., Fleming, P., et al. (2015) Resolving the value of the dingo in ecological restoration. Restoration Ecology, 25, 201–208. Packard, J. M., Mech, L. D. & Ream, R. R. (1992). Weaning in an arctic wolf pack: behavioral mechanisms. Canadian Journal of Zoology, 70, 1269–1275. Pfeifer, S. (1980). Role of the nursing order in social development of mountain lion kittens. Developmental Psychobiology, 13, 47–53. R Core Team. (2014). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, www.R-project.org. Raihani, G., González, D., Arteaga, L. & Hudson, R. (2009). Olfactory guidance of nipple
534
attachment and suckling in kittens of the domestic cat: inborn and learned responses.
535
Developmental Psychobiology, 51, 662–671.
536
Rheingold, H. L. (1963). Maternal behavior in the dog. In H. L. Rheingold (Ed.), Maternal
537
behavior in mammals (pp. 169–207). New York, USA: John Wiley & Sons. 25
538
Rosenblatt, J. S. (1971). Suckling and home orientation in the kitten: a comparative
539
developmental study. In E. Tobach, L. Aronson & E. Shaw (Eds.), The biopsychology
540
of development (pp. 345–410). New York, USA: Academic Press.
541
Schielzeth, H. & Nakagawa, S. (2013). rptR: Reproducibility for Gaussian and non-Gaussian
542
data. R package version 0.6.405/r52. http:/R-Forge.R-project.org/projects/rptr/
543
Serpell, J. A. (2014). Domestication and history of the cat. In D. C. Turner & P. Bateson
544
(Eds.), The domestic cat. The biology of its behaviour. 3rd edition. (pp. 83–100).
545
Cambridge, UK: Cambridge University Press.
546
Shannon, L. M., Boyko, R. H., Castelhano, M., Cory, E., Haywood, J. J., McLean, C., …
547
Boyko, A. R. (2015). Genetic structure in village dogs reveals a Central Asian
548
domestication origin. Proceedings of the National Academy of Sciences, 112, 13639–
549
13644.
550
Sherman, P. W., Braude, S. & Jarves, U. M. (1999). Litter sizes and mammary numbers of
551
naked mole-rats: breaking the one-half rule. Journal of Mammalogy, 80, 720–733.
552
Smith, B. (2015). Biology and behaviour of the dingo. In B. Smith (Ed.), The dingo debate:
553
origins, behaviour and conservation (pp. 25–53). Clayton South, Victoria: CSIRO
554
Publishing.
555
Smith, B. & Savolainen, P. (2015). The origin and ancestry of the dingo. In B. Smith (Ed.),
556
The dingo debate: origins, behaviour and conservation (pp. 55–79). Clayton South,
557
Victoria: CSIRO Publishing.
558
Comment [RH1]: Can someone please check how JCP handles long co-author lists? Various journals do it this way.
Smith, B. & Watson, L. (2015). The role of private sanctuaries in dingo conservation and the
559
management of dingoes in captivity. In B. Smith (Ed.), The dingo debate: origins,
560
behaviour and conservation (pp. 277–299). Clayton South, Victoria: CSIRO
561
Publishing.
26
562
Speakman, J. R. (2008). The physiological costs of reproduction in small mammals.
563
Philosophical Transactions of the Royal Society of London, Series B: Biological
564
Sciences, 363, 375–398.
565 566 567 568
Stamps, J. & Groothuis, T. G. G. (2010). The development of animal personality: relevance, concepts and perspectves. Biological Reviews, 85, 301–325. Sunquist, M. & Sunquist, F. (2002). Wild cats of the world. Chicago, USA: University of Chicago Press.
569
Thomson, P. C. (1992). The behavioural ecology of dingoes in North-Western Australia: II.
570
Activity patterns, breeding season, and pup rearing. Wildlife Research, 19, 519–530.
571
Thomson, P. C., Rose, K. & Kok, N. E. (1992). The behavioural ecology of dingoes in North-
572
Western Australia: V. Population dynamics and variation in the social system.
573
Wildlife Research, 19, 565–584.
574 575
Trillmich, F. & Hudson, R. (2011). The emergence of personality in animals: the need for a developmental approach. Developmental Psychobiology, 53, 505–509.
576
Wang, G-D., Zhai, W., Yang, H-C., Wang, L., Zhong, L., Liu, Y-H., …. Zhang, Y-P. (2015).
577
Out of southern East Asia: the natural history of domestic dogs across the world. Cell
578
Research, 1–13, www.nature.com/cr
579
Comment [RH2]: As for Shannon et al. above.
Wilton, A. (2001). DNA methods of assessing dingo purity. In C. R. Dickman & D. Lunney
580
(Eds.), A symposium on the dingo (pp. 49–56). Sydney, New South Wales: Royal
581
Zoological Society of New South Wales, Australia.
582
27
583 584 585 586
Table 1
587
Details of the Dingo Mothers and Litters Dingo
Age (yrs)
Parity
Litter size
Sex M
F
Post nataldeaths
Final litter size
Amelia
2
0
5
2
3
0
5
Petal
5
0
6
4
2
0
6
Freckle
8
2
6
2
4
1
5
Opal
7
3
7
3
2
2
5
588 589
28
590 591 592 593
Table 2
594
Comparison of Different Behaviours Observed During Nursing Sessions in Puppies of the
595
Dingo and the Domestic Dog
12
P
602 s ± 53
1.52
0.22
2.1 ± 0.3
2.3 ± 0.2
0.57
0.45
255 s ± 39
286 s ± 33
0.38
0.54
Parameter
Dingo
Domestic dog
Attachment time / session
533 s ± 89
N nipples used /session Attachment time / nipple 596 597
Data are from 12 puppies of the dingo (4 litters) and 42 puppies of the domestic dog (9 litters)
598
recorded during 30-minute observation sessions. Means and 95% confidence intervals are
599
given, and statistical comparisons were done by linear mixed-effects models including litter
600
identity as a random factor. Data for the domestic dog are taken and reanalysed from Arteaga
601
et al., 2013.
602 603
29
604 605
Figure 1. Time course of the cumulative proportion of different nipples used by individual
606
puppies of (a) the dingo and (b) the domestic dog across four recorded nursing sessions
607
during the first 2 postnatal weeks. Solid horizontal lines give medians, boxes indicate 25th
608
and 75th percentiles, and vertical lines show the 10th and 90th percentiles. Sample sizes
609
(number of puppies) are given in the graph. See text for statistics. Data for the domestic dog
610
are taken and reanalysed from Arteaga et al., 2013.
611
30
612 613 614
Figure 2. Distribution of the (a, c) percentage time and (b, d) the percentage frequency that
615
puppies of the dingo (N = 12) and the domestic dog (N = 42) were attached to the different
616
nipples of their mother. Bars indicate means ± 95% confidence intervals, averaged over all
617
puppies. The ranking of the nipples indicates the absolute frequency and time of attachment
618
to the different nipples by each individual puppy, where 1 denotes the nipple per puppy with
619
the longest/most frequent attachment and 8 denotes the nipple with the shortest/least frequent
620
attachment. See text for statistics. Data for the domestic dog are taken and reanalysed from
621
Arteaga et al., 2013.
622 31
