A facilitated release mechanism for quills of the North American porcupine (Erethizon dorsatum)

Abstract

Quill-withdrawal experiments confirm a hypothesis proposing that North American porcupines (Erethizon dorsatum) give up a quill more easily if the quill's root is first thrust back sharply into the porcupine's skin. The hypothesis was
tested in 8 porcupines. When quills are impacted in a way that mimics contact with a predator, the tension required to remove
an impacted quill from the porcupine's skin was reduced from 90.3 g ± 16.8 SD by 38.1%. The phenomenon was observed only in quills that had been erected by porcupines. This tension-reducing mechanism
allows a porcupine to separate more readily from an attacking predator and adds to the effectiveness of porcupine quills as
smart weapons.

Full text

381
Journal of Mammalogy, 83(2):381–385, 2002
A FACILITATED RELEASE MECHANISM FOR QUILLS OF THE
NORTH AMERICAN PORCUPINE (ERETHIZON DORSATUM)
U
LDIS
R
OZE
*
Department of Biology, Queens College, Flushing, NY 11367
Quill-withdrawal experiments confirm a hypothesis proposing that North American por-
cupines (Erethizon dorsatum) give up a quill more easily if the quill’s root is first thrust
back sharply into the porcupine’s skin. The hypothesis was tested in 8 porcupines. When
quills are impacted in a way that mimics contact with a predator, the tension required to
remove an impacted quill from the porcupine’s skin was reduced from 90.3 g
6
16.8 SD
by 38.1%. The phenomenon was observed only in quills that had been erected by porcu-
pines. This tension-reducing mechanism allows a porcupine to separate more readily from
an attacking predator and adds to the effectiveness of porcupine quills as smart weapons.
Key words: Erethizon dorsatum, porcupine, quill
On contact with a predator, the North
American porcupine, Erethizon dorsatum,
may leave tens to hundreds of quills in its
opponent. These quills form a transitory
bridge attaching the porcupine to its antag-
onist. It is in the porcupine’s interest to
break this attachment as quickly as possi-
ble. Studies by Po-Chedley and Shadle
(1955) suggest that the tension required to
extract a single quill from the pelage of the
porcupine exceeds 100 g. The value varies
with season, with highest tensions observed
in winter and lowest in midsummer (Po-
Chedley and Shadle 1955). Increase in re-
lease tension in winter months correlates
with the greater need for body insulation at
this time. The quills contribute importantly
to this insulation function (DeMatteo and
Harlow 1997; Fournier and Thomas 1999).
As modified hairs, quills share the proper-
ties of piloerectability and airtrapping
shown by the fur and guard hairs (Chapman
and Roze 1997; Roze 1989).
The number of quills lost in a single quill
strike may be considerable. Shadle and Po-
Chedley (1949) report 79 quills embedded
in 1 author’s leg after a porcupine encoun-
* Correspondent: uldis
p
roze@qc.edu
ter. This presents a potential problem to the
porcupine: the force required to remove all
the quills of a routine quill strike may ex-
ceed the weight of the porcupine, putting it
in danger of becoming attached to its an-
tagonist.
This study examines a hypothesis pro-
posed by Chapman (Chapman and Roze
1997), that a porcupine can lose a quill
more easily after the quill has been first
thrust back into its own skin, as occurs dur-
ing normal porcupine–predator contact. The
hypothesis is based on the histological ob-
servation that the slender base of the quill
runs through a collagenous spool embedded
in the dermis (Chapman and Roze 1997). A
short distance distal to the spool, the quill
diameter widens rapidly, forming a shoul-
der. The shoulder prevents self-stabbing by
limiting the depth of a quill’s back-travel
into the dermis of the porcupine after a
strike against a predator. At the same time,
the back-travel, in some unknown manner,
weakens the contact between the quill fol-
licle and surrounding tissues, allowing for
easier removal of the quill (Figs. 1A and
1B). In anesthetized or calm porcupines,
with quills not erected, the contact between

382 Vol. 83, No. 2JOURNAL OF MAMMALOGY
F
IG
. 1.—Proposed mechanism for facilitated
quill release, showing differences between erect
and relaxed quills. Abbreviations: e
5
epider-
mis; p
5
piloerector muscle; q
5
quill base; r
5
retinaculum; s
5
spool; t
5
transverse muscle.
A) Erect quill in aroused porcupine, with trans-
verse and piloerector muscles contracted. B)
Erect quill has traveled deeper into the skin after
striking an object and moving through immobi-
lized spool, with quill root shearing attachment
to surrounding tissue. C) Relaxed quill in anes-
thetized porcupine, with transverse and piloerec-
tor muscles relaxed. D) Relaxed quill has struck
an object. Spool and surrounding tissues have
traveled with quill root, preventing shear of root
attachment. Arrow indicates direction of impact
on quill.
quill follicle and surrounding tissues is not
disturbed by a similar backward thrust, pre-
sumably because the spool is free to move
with the quill shaft (Figs. 1C and 1D). The
nature of the attachment of quill follicle to
surrounding tissues remains to be clarified.
This study tests the tension-reduction hy-
pothesis by simulating porcupine–predator
contact via a styrofoam block struck against
selected quills, then comparing the with-
drawal tensions of struck and undisturbed
quills.
M
ATERIALS AND
M
ETHODS
Eight porcupines were tested in a 2-by-2 de-
sign. For each animal, quill removal tension was
measured for both impacted quills and those not
impacted. Quill impaction was done first in
aroused porcupines, with erect quills, then in the
same porcupines under anesthesia, with relaxed
quills. Removal tensions for unimpacted control
quills were measured in both cases. All quill re-
movals were done under anesthesia.
Porcupines were captured at a salt source in
the Catskill Mountains of New York (Roze
1984). They included 4 adults (body weights
5.2–7.2 kg), 3 yearlings 1–2 years old (weights
2.8–3.0 kg), and 1 juvenile (weight 1.9 kg). Age
determinations were made according to Earle
and Kramm (1980). Seven of the 8 porcupines
were females.
Porcupines were captured in closable plastic
coolers (22 by 32 by 23 cm; Igloo, Rubbermaid
Corp., Wooster, Ohio) and transported to an ex-
amining room within minutes. For examination,
the top of the cooler was partially opened under
bright light. Porcupines responded by lowering
their heads and presenting the erected quill sur-
face of the posterior lumbar region (known as
the rosette). A cube of Styrofoam, 2 by 2 by 2
cm and held in a hemostat, was struck against
the erect quill surface and the Styrofoam cube
left in place. The strike left 5–23 quills embed-
ded in the Styrofoam (X
¯
5
9.5 quills). The an-
imal was then anesthetized with 10 mg/kg ke-
tamine–HCl by intramuscular injection (Ketaset,
Fort Dodge Laboratories, Fort Dodge, Iowa).
Once anesthetized, the porcupine was removed
from the cooler and the Styrofoam block re-
moved by cutting all inserted quills at their distal
ends, approximately 0.5 cm from the Styrofoam
cube. Afterward, quills that had been embedded
in Styrofoam could be identified by their trun-
cated white stumps. To measure tension needed
to remove the quills, each truncated quill in the
skin was clamped with an alligator electric clip
mounted on a 300-g spring scale (Homs Corp.,
Belmont, California). The scale was retracted at
a rate of approximately 1 cm/s and the peak ten-
sion recorded. For each cut quill, a control (un-
manipulated) quill of similar size and location
was tested in the same way in the anesthetized
porcupine.
The same porcupines, while still anesthetized
and with quills relaxed, were again struck with
Styrofoam blocks and quill-withdrawal tensions
measured as above. In this case, because the
quills were lying flat against the body, a small
area of quills was elevated by means of a stiff

ROZE—FACILITATED QUILL RELEASEMay 2002 383
T
ABLE
1.—Tensions needed to remove quills from rosette areas of porcupines in aroused (quills
erect) and anesthetized (quills relaxed) states in quill-release experiments. Experimental quills were
struck with Styrofoam blocks; control quills were not manipulated before being removed.
Treatment n
Removal tensions (g)
Experimental
X
¯
SD
Control
X
¯
SD P
a
Aroused
b
Anesthetized
b
8
8
55.9
80.1
10.8
11.7
90.3
83.7
16.8
20.9
,
0.001
0.438
a
Probability, repeated measures ANOVA (SPSS, Inc. 1998).
b
For aroused porcupines, there were 5–23 paired tests (impacted versus unimpacted) per individual; for anesthetized porcupines,
4–11 paired tests (impacted versus unimpacted) per individual.
wire to serve as a target for the Styrofoam block.
Data were analyzed by repeated measures anal-
ysis of variance (ANOVA), with a follow-up
paired-sample t-test (SPSS, Inc. 1998).
The Styrofoam block technique probably un-
derestimates the true degree of facilitation of
quill separation. This is because erect rosette
quills point in all directions; some may catch the
Styrofoam so lightly that full shearing of quill–
follicle supports may not take place. For 1 year-
ling female, the quill tension experiment was run
on both lower-back quills and tail quills, with
the latter embedded in Styrofoam by a tail-slap.
The tight confines of the cooler typically pre-
vented use of tails by the larger adult porcu-
pines.
The protocol for quill withdrawal was inad-
vertently extended to human skin after 2 quill
strikes suffered by the author during the course
of the study. The strikes left 20 quills embedded
in a hand and thumb. The tension required to
extract these was measured with alligator clip
and spring scale as for the quills extracted from
porcupines.
After the quill tension experiments, porcu-
pines were weighed, sex and reproductive status
determined, and unmarked animals were given
a groin tattoo. After recovery from the anes-
thetic, they were released at the site of capture.
R
ESULTS
With erect quills, less tension was needed
to remove impacted quills than non–im-
pacted quills, whereas for relaxed quills,
there was no difference. In aroused (quill-
erected) animals, impacted quills were re-
moved with 38.1% less tension than were
unmanipulated quills from the same ani-
mals after anesthesia (Table 1). The reduc-
tion is significant (repeated measures AN-
OVA: F
5
60.08, d.f.
5
1, 6, P
,
0.0001).
There was a significant interaction between
arousal state and reduction of quill tension
(F
5
18.09, d.f.
5
1, 6, P
5
0.005). A
follow-up analysis by paired-samples t-test
showed tension reduction was significant
only in the aroused state (t
5
8.11, d.f.
5
7, P
,
0.001). There was no significant re-
duction of tension in the anesthetized (quill-
relaxed) state (t
5
0.82, d.f.
5
7, P
5
0.438). Reduction in quill tension was not
significantly related to body weight (F
5
0.872, d.f.
5
1, 6, P
5
0.386). Only one of
the animals tested was a male, so sexual
differences were not tested. A similar facil-
itation mechanism appears to be true for tail
quills. For the single individual in whom
tail quills were tested in the aroused state,
manipulation reduced removal tension from
88.2 g
6
34.0 SD to 43.7
6
24.4 g, a re-
duction of 50.4% (t
5
5.28, P
,
0.001,
paired t-test).
D
ISCUSSION
After a quill strike, it is in the porcu-
pine’s interest to separate quickly from its
adversary. Moreover, after the separation, it
is in the porcupine’s interest that the quills
stay with the adversary. Yet the tension re-
quired to remove quills from human skin
(128.0
6
67.6 g for 5 quills in side of hand
and 153.3
6
83.2 g for 15 quills in thumb
suffered during the course of this study) ap-

384 Vol. 83, No. 2JOURNAL OF MAMMALOGY
proaches the tension needed to remove un-
manipulated quills from the porcupine ro-
sette (90.3
6
16.8 g). These removal ten-
sions recorded from human skin probably
overestimate the human–porcupine differ-
ence because any quill more firmly attached
to the porcupine would have been unavail-
able for measurement. Any enhancement of
this difference as follows back-thrust of the
quill into the porcupine’s skin would allow
the porcupine to separate more quickly
from its adversary and ensure the quill
stayed with the adversary. Rapid separation
from an adversary after a quill strike should
increase the maneuvering ability of a por-
cupine and aid its escape.
Back-thrusting a relaxed quill into the
skin of the porcupine (as observed under
anesthesia) did not reduce removal tension.
Presumably, with relaxation of the piloerec-
tor and transverse muscles in the anesthe-
tized state, the attachments between quill
root and surrounding tissues are not sheared
because the quill and surrounding tissues
are free to move as a unit (Fig. 1). Such a
default mechanism would prevent unnec-
essary loss of quills when body surfaces are
pressed against den walls, tree trunks, and
other static surfaces of the environment.
The evolutionary homologies of the
spool require further study. A possible can-
didate is the network of coarse collagenous
fibers of the reticular layer of the dermis,
arranged in a plane roughly parallel to the
skin (Fawcett 1994). On visual inspection,
these fibers appear similar to the collage-
nous fibers making up the spool. Forces ac-
counting for their tight radius of curvature
in the spool remain to be explored. There
is neither a papillary layer nor a vitreous
layer around the telogen quill follicle’s dis-
tal region.
Vincent and Owers (1986) propose an al-
ternative predator–porcupine separation
mechanism in the Hystricidae and Erethi-
zontidae. The quills of porcupines in these
groups are physically adapted for keeping a
predator as far away as possible. On the ba-
sis of mechanical considerations, Vincent
and Owers propose that such quills are de-
signed to break at the tip after predator im-
pact. Broken quills in human skin from Er-
ethizon have been observed by many au-
thors, including Roze (1989), Shadle
(1947), and Taylor (1935). The breaking of
the quill would facilitate predator–porcu-
pine separation. However, in my observa-
tions, broken quills have always represent-
ed a small proportion of the total quill-
strike, and breaking typically occurs after
porcupine separation, when the predator at-
tempts to extract skin-embedded quills. Of
the 79 leg-embedded quills reported by
Shadle and Po-Chedley (1949), none had
broken. Likewise, of the 20 skin-embedded
quills suffered during this study, all were
intact. It therefore appears that quill break-
age in Erethizon dorsatum represents a mi-
nor mechanism for facilitated release.
This study adds to our understanding of
the porcupine quill as a smart weapon. Be-
cause their quills confer strong antipredator
protection, porcupines use them in apose-
matic signaling, thus reducing predator at-
tack. The 1st signal, olfactory in nature, in-
volves the release of a porcupine-specific
odorant, R-delta-decalactone (Li et al.
1997). The odorant is disseminated by the
osmetrichial adaptations of rosette quills. A
2nd signal, effective in the nocturnal envi-
ronment of the foraging porcupine, is based
on the black-and-white visual contrast gen-
erated by quills of the lower back and tail.
The contrast is enhanced by brighteners in
the white regions of the quills (U. Roze, in
litt.).
But the same quill adaptations that make
them so effective against predators may
generate dangers for the porcupines. Thus,
because porcupines routinely fall out of
trees, they risk self-impalement (Hale and
Fuller 1996; Roze 1989). Risks of infection
in such accidents are reduced by the quill
surface coating of free fatty acids, which
have antibiotic properties (Roze et al.
1990). By evolving quill adaptations for ol-
factory and visual aposematic signaling, ad-
aptations for reducing self-infection risk,

ROZE—FACILITATED QUILL RELEASEMay 2002 385
and adaptations for enhancing quill sepa-
ration after a predator strike, the porcupine
quill has maintained its effectiveness
against predators while reducing incidental
costs to its owner.
A
CKNOWLEDGMENTS
I thank D. M. Chapman for helpful discus-
sions of all aspects of the hypothesis, B. J. An-
thony for help with statistics, and two anony-
mous reviewers for valuable suggestions for im-
provement of the manuscript. This study was
supported in part by a grant from PSC-CUNY
(Professional Staff Congress–City University of
New York).
L
ITERATURE
C
ITED
C
HAPMAN
,D.M.,
AND
U. R
OZE
. 1997. Functional his-
tology of quill erection in the porcupine, Erethizon
dorsatum. Canadian Journal of Zoology 75:1–10.
D
E
M
ATTEO
, K. E.,
AND
H. J. H
ARLOW
. 1997. Ther-
moregulatory responses of the North American por-
cupine (Erethizon dorsatum bruneri) to decreasing
ambient temperature and increasing wind speed.
Comparative Biochemistry and Physiology, B. Com-
parative Biochemistry 116:339–346.
E
ARLE
, R. D.,
AND
K. R. K
RAMM
. 1980. Techniques for
age determination in the Canadian porcupine. Jour-
nal of Wildlife Management 44:413–419.
F
AWCETT
, D. W. 1994. Bloom and Fawcett, a textbook
of histology. 12th ed. Chapman and Hall, New York.
F
OURNIER
,F.,
AND
D. W. T
HOMAS
. 1999. Thermoregu-
lation and repeatability of oxygen-consumption
measurements in winter-acclimatized North Ameri-
can porcupines (Erethizon dorsatum). Canadian
Journal of Zoology 77:194–202.
H
ALE
, M. B.,
AND
T. K. F
ULLER
. 1996. Porcupine (Er-
ethizon dorsatum) demography in central Massachu-
setts. Canadian Journal of Zoology 74:480–484.
L
I
, G., U. R
OZE
,
AND
D. C. L
OCKE
. 1997. Warning odor
of the North American porcupine (Erethizon dor-
satum). Journal of Chemical Ecology 23:2737–2754.
P
O
-C
HEDLEY
,D.S.,
AND
A. R. S
HADLE
. 1955. Pelage
of the porcupine, Erethizon dorsatum dorsatum.
Journal of Mammalogy 36:84–95.
R
OZE
, U. 1984. Winter foraging by individual porcu-
pines. Canadian Journal of Zoology 62:2425–2428.
R
OZE
, U. 1989. The North American porcupine. Smith-
sonian Institution Press, Washington, D.C.
R
OZE
, U., D. C. L
OCKE
,
AND
N. V
ATAKIS
. 1990. Anti-
biotic properties of porcupine quills. Journal of
Chemical Ecology 16:725–734.
S
HADLE
, A. R. 1947. Porcupine spine penetration. Jour-
nal of Mammalogy 28:180–181.
S
HADLE
, A. R.,
AND
D. P
O
-C
HEDLEY
. 1949. Rate of
penetration of a porcupine spine. Journal of Mam-
malogy 30:172–173.
SPSS, I
NC
. 1998. SPSS for Windows. Version 9.0.0.
SPSS, Inc., Chicago, Illinois.
T
AYLOR
, W. P. 1935. Ecology and life history of the
porcupine (Erethizon epixanthum) as related to the
forests of Arizona and the southwestern United
States. University of Arizona Bulletin 6:1–177.
V
INCENT
,J.F.V.,
AND
P. O
WERS
. 1986. Mechanical
design of hedgehog spines and porcupine quills.
Journal of Zoology (London) 210:55–75.
Submitted 24 July 2000. Accepted 5 June 2001.
Associate Editor was Thomas J. O’Shea.