Ontogeny of Squid Mantle Function: Changes in the Mechanics of Escape-Jet Locomotion in the Oval Squid, Sepioteuthis lessoniana Lesson, 1830

13 Pages·2002·6.1 MB·

by  Joseph T Thompson

Reference: Binl. Bull. 203: 14-26. (August 2(1(12)
Ontogeny of Squid Mantle Function: Changes in the
Mechanics of Escape-Jet Locomotion in the Oval
Squid, Sepioteuthis lessoniana Lesson, 1830
JOSEPH T. THOMPSON* AND WILLIAM M. KIER
Department ofBiology. CB&3280 Coker Hall, University ofNorth Carolina.
Chapel Hill. North Carolina 27599-j280
Abstract. In Sepioteuthis lessoniana, the oval squid, on- propulsion efficiency of the exhalant phase of the jet is
togenetic changes in the kinematics of the mantle during highest in hatchlings. and (2) that the mechanics of the
escape-jet locomotion imply a decline in the relative mass circumferential muscles of the mantle change during
flux of the escape jet and may affect the peak weight- growth.
specific thrust ofthe escapejet. Toexamine the relationship
between ontogenetic changes in the kinematics ofthe man-
Introduction
tle and the thrust generated during the escape jet, we simul-
taneously measured the peak thrust and the kinematics of The kinematics and mechanics of locomotion change
We
the mantle ofsquid tethered to a force transducer. tested significantly during the growth of many aquatic animals,
an ontogeneticmsmeries of S. lessoniana that ranged in size especially in those that are small at hatching and grow to
from 5 to 40 dorsal mantle length (DML). In newly large body size. The ontogeny of the kinematics and me-
ehsactacpheedjetsqauniddsr,etahcrhuesdtapemaakxeidm4u0mnoifsbaefttwereetnhe0.s1ta0rtmNofatnhde cdhyannaimciscofenlvoicroomnomteinotn omfaytherefalneictmaalcahsanigtegrionwtshe(Wheyidhrso,-
0.80 mN. In the largest animals, thrust peaked 70 ms after 1974: Batty. 1984; Webb and Weihs. 1986; Osse. 1990;
the start of the escape jet and reached a maximum of Williams, 1994; Miiller and Videler, 1996). The size ofthe
between 18 mN and 1 10mN. Peak thrust was normalizedby animal or its propulsor, the speed of the animal or its
the wet weight ofthe squid and also by the cross-sectional propulsor. and the kinematic viscosity ofwaterall affect the
area of the circumferential muscle that provides power for forces experienced by an aquatic animal (e.g., Lighthill.
the escapejet. The weight-specific peak thrust ofthe escape 1975). The Reynolds number(Re) describes therelationship
jet averaged 0.36 in newly hatched squid and increased amons these three factors.
significantly to an average of 1.5 in the largest squids
measured (P < 0.01 ). The thrust per unit area of circum- Re = IU/i' (Lishthill, 1975) (Eq. 1)
ferential muscle averaged 0.25 mN/mnr in hatchlings and
increased significantly to an average of 1.4 mN/mrrr in the where / is a characteristic length ofthe organism (e.g., body
largest animals tested (P < 0.01). The impulse of the or propulsor length). U is the velocity ofthe organism (or a
escapejet was also lowest in newly hatched individuals ( 1.3 portion of the organism) relative to the surrounding water.
mN s) and increased significantly to 1000 mN s in the and i'is the kinematic viscosity ofthe fluid (i.e.. the ratio of
largest squids measured (P < 0.01). These ontogenetic water viscosity to water density). The Reynolds number
changes in the mechanics of the escapejet suggest ( 1 ) that characterizes the relative importance of inertial forces and
viscous forces on the swimming animal (Vogel. 1994).
Thus, when the Reynolds number is high (>10\ inertial
emaR*ielTc.oeuinvwce.hdeod2mu5cJourlryes2p00o1n:deanccceepstheodul3d0hMeaaydd2r0e(s1s2.ed. E-mail: joethomp floorwce(s<d1o)m,ivniasctoesiltoycdoommoitniaotne;s.whIennanthientReerymneodliadtsenRuemybneorldiss
Ahhivvuilinn: DML. dorsal mantle length. number fluid regime (1 < Re < 103). inertial forces dom-
14
ONTOGENY OF SQUID MANTLE FUNCTION 15
inate. hut the effects ofviscosity are considerable (Daniel ct during the escape jet is highest in hatchling squids and
mm
cil.. 1992). decreases exponentially until the animals reach 15
The hydrodynamic environment influences the mechanics DML. In S. lessoniana specimens larger than 15 mm DML.
of locomotion during ontogeny. For example, newly the maximum rate of contraction of the mantle does not
hatched larval brine shrimp and fish swim in a fluid regime change significantly (Thompson and Kier, 200la).
between Re = 1 and Re = 20 (Batty. 19S4: Williams. 1994: Specificproblem. The ontogenetic changes in the kinemat-
Miiller el til.. 2000). where inertia! and viscous forces are ics ofthe mantle in S. lessoniana may affect the performance
nearly equal. In such a hydrodynamic environment, inertia! and the mechanical efficiency of escape-jet locomotion. Spe-
modes of locomotion, such as burst-and-coast swimming, cifically, changes in the kinematics of the mantle result in
=
are hypothesized to be energetically costly compared with higher relative mass flux ( mass ofwater exiting the mantle
drag-based modes of locomotion, such as steady unguilli- cavity per unit time, relative to body mass) in newly hatched
form swimming (Weihs, 1974; Miiller ct til., 2000). Thus, squids than in larger animals (Thompson and Kier, 200la).
hydrodynamic considerations may explain the observed Because the thrust produced during the escape jet is propor-
change from drag-based to inertial locomotion during the tional to the product of mass flux and the velocity of water
growth of larval brine shrimp and fish (Batty. 1984; Wil- exiting the funnel aperture (averaged over the funnel aperture;
liams, 1994; Miiller et at. 2000). Vogel, 1994), achange in the relative mass flux may affectthe
Locomotion in some animals may be fixed, despite shifts thrust generated during the escapejet.
in their hydrodynamic regime during growth. For example. To explore how changes in the kinematics of the mantle
thejet propulsion used by squids is analogous to the inertia- impact escape-jet locomotion, we measured the peak thrust
dependent burst-and-coast swimming employed by some generatedduring the escapejet in an ontogenetic seriesofS.
fish. A squid produces a burst of thrust as the mantle lessoniana. Weexamined the implicationsofanontogenetic
contracts and expels waterfrom the mantle cavity, followed change in jet thrust on the propulsion efficiency of escape-
by a coasting phase as the mantle expands and the mantle jet locomotion, and on the mechanics ofthe circumferential
cavity refills. Because squids are limited to inertia-depen- muscles that power the jet. We also discuss the possibility
dentjet propulsion for rapid swimming, the mechanics and that the evolution ofontogenetic changes in the morphology
energetics oflocomotion for small hatchling squids may be and kinematics of the mantle and funnel might reflect the
compromised given the near parity of inertia and viscosity scale effects of jet locomotion.
oftheir intermediate Re fluid environment (1 < Re < 100,
unpub. obs. ofSepioteuthis lessoniana during fast and slow Materials and Methods
jetting). For instance, unlike adults, newly hatched individ- Animals
uals of S. lessoniana (unpub. obs.) and Loli^o vulgaris
(Packard. 1969) coast only a short distance after a burst of We obtained an ontogenetic series ofSepioteuthis lesso-
thrust. Because inertia-based modes of locomotion are un- niana Lesson, 1830, the oval squid. Wild embryos collected
favorable in the hydrodynamic environment of the hatch- from Izo. Japan, in September 1999 and August 2000 were
lings (Miiller ct <;/., 2000). we might expect compensatory raised by the National Resource Center for Cephalopods
changes in the kinematics ofthe mantle and the mechanics (NRCC) at the University of Texas Medical Branch,
of jet locomotion as squid grow. Galveston, Texas. The cohort consisted of thousands of
Indeed, the amplitude of mantle movement during jet embryos from six to eight different egg mops. Thus, it is
locomotion changes as a squid grows. In Lolifto vtilgaris likely that the sample populations were not the offspring of
(Packard, 1969). L. opalescens (Gilly et til., 1991; Preuss et a few closely related individuals, but were representative of
a/.. 1997), and S. lessoniana (Thompson and Kier, 2001a), the natural population at the collection site.
the maximum amplitude of mantle movement during the The embryos were transported from the field to the
escape jet decreases dramatically during ontogeny. In S. NRCC and reared (Lee ettil., 1994). Commencing at hatch-
lessoniana, the maximum contraction and hyperinflation of ing and at regular intervals thereafter, live squid were sent
the mantle (both measured as the percent change from the via overnight express shipping from the NRCC to the Uni-
resting mantle diameter) are significantly greater during an versity ofNorth Carolina. The squids ranged in size from 5
escape jet in small hatchlings than in the larger animals to 40 mm dorsal mantle length (DML) and in age from
(Thompson and Kier. 2001a). The maximum contraction newly hatched to 6 weeks after hatching.
and hyperinflation of the mantle decrease exponentially Prior to the start of the experiments, the animals were
until the squid reach adorsal mantle length (DML) ofabout allowed at least 30 min to equilibrate in an 80-1 circular
15 mm. In squids larger than 15 mm DML. the maximum holding tank. The temperature (23 C) and salinity (35 ppt)
contraction and hyperinflation of the mantle during the of the water matched the temperature and salinity of the
escapejet remain unchanged (Thompson and Kier, 200la). waterin which the squid were raised. Circularwater flow in
In addition, the maximum rate of contraction ot the mantle the tank helped keep the squid swimming parallel to the
16 J. T. THOMPSON AND W. M. KIER
sides of the tank to prevent injury. No animal remained in the tether and returned to the holding tank, squid swam
the holding tank for longer than 2 h. normally, and their survival was similar to that of undis-
Every animal used for the experiments was in excellent turbed specimens. All experiments met the animal care
condition. The coloration and activity of the squids were guidelines of the University of North Carolina at Chapel
qualitatively similar to those ofanimals in the large culture Hill.
NRCC
tanks at the (pers. obs.). Many ofthe animals in this
study captured and consumed prey (feederguppies ormysid
shrimp) while in the holding tank. None ofthe squids inked Critique ofthe method
during shipping or during the anesthesia prior to the exper-
iments. Previously published methods of estimating the thrust
produced by jetting squids require measuring the funnel
The tether aperture during jet locomotion (see Johnson et al., 1972;
O'Dor, 1988a; Anderson and DeMont, 2000). Because a
We measured the thrust produced during escape-jet squid can adjust both the orientation of the'funnel and the
locomotion in squid tethered to a force transducer. Squid size and shape of the funnel aperture during the jet (Zuev,
were tethered (see Thompson and Kier, 2001a) as follows. 1966; O'Dor, 1988a), accurate measurement of funnel di-
Individual squid were removed from the holding tank and ameteris difficult. These problems areexacerbatedin newly
anesthetized lightly in a 1:1 solution of 7.5% MgCl2 : arti- hatched squid by their inability to maintain position in a
ficial seawater (Messengeret ai. 1985). A needle (0.3-mm- flow tank, by their small size, and by the small funnel
diameter insect pin for small animals or 0.7-mm-diameter aperture. In addition, the method employed by Johnson et
hypodermic needle for large animals) was inserted through al. (1972) and O'Dor (1988a) is not applicable to hatchling
the brachial web ofthe squid, anterior to the brain cartilage squid because it is not currently possible to monitor mantle
and posteriorto the buccal mass. The needle was positioned cavity pressure in these small animals without substantially
between these two rigid structures to prevent it from tearing altering normal mantle kinematics.
the soft tissue of the squid. The needle was inserted into a The tethering approach adopted here was chosen to by-
hollow stainless steel post (hypodermic tubing) attached to pass these difficulties. Tethering to a force transducer is
a force transducer. Flat, polyethylene washers on the post particularly effective forsmall animals (see Svetlichnyy and
and needle were positioned above and below the head to Svetlichnyy, 1986; Lenz and Hartline, 1999), and the
prevent vertical movement. method has the considerable advantage ofnot requiring the
Insertion ofthe needle through the anesthetized squid was small and dynamic funnel aperture ofhatchling andjuvenile
rapid and required minimal handling ofthe animal. Individ- squids to be measured to calculate thrust. Nevertheless, a
uals of S. lessoniana become nearly transparent under an- potential disadvantage of this approach is its interference
esthesia, making the buccal mass and the brain cartilage with the movement of the mantle during the escape jet.
readily visible. Needle placement was verified after the Because the tether did not touch the mantle and the kine-
experiment by examination of the location of the needle matics of the mantle of tethered squids were qualitatively
entrance and exit wounds. similar to those of untethered squids swimming in the
The tethered squid and force transducer were transferred holding tank (see Thompson and Kier, 2001a), it is unlikely
to an aquarium (0.4 m by 0.2 m by 0.2 m) filled with 20 to that this was a source of significant error. Another possible
23 C artificial seawater and were allowed to recover. Re- disadvantage is that the flow field around a tethered squid is
covery, which was judged by the return of normal chro- stationary. During an escape jet, an untethered squid has a
matophore patterning and fin activity, normally occurred velocity relative to the water, and this velocity affects the
within 10 min. Tethered squid remained alive and in appar- magnitude ofthe thrust required to keep a squidjetting at a
ent good health for up to several hours, though most squid given velocity (Vogel, 1994). Because we were investigat-
were tethered for fewer than 20 min. ing the maximum thrust that a squid can generate and not
Although tethering is an invasive technique, it did not the variations in thrust produced duringjetting, the station-
appear to be unduly traumatic to the squid. First, tethered ary flow field is not an issue. The final potential disadvan-
squid behaved similarly to the animals in the holding tank. tage is that the post ofthe tether, which was located down-
Both the tethered and free-swimming squid spent most of stream ofthe funnel aperture (Fig. 1A), could alterthrustby
the time hovering, using the fins and low-amplitude jets. interfering with the fluid exiting the funnel aperture. Sig-
Second, unlike squid that are in distress or startled, more nificant alterations in thrust caused by the post are consid-
than 90% ofthe tethered squid did not eject ink. Third, the ered unlikely because ofits small size relative to the funnel
chromatophore patterns of tethered squid did not differ aperture and the distance of the post from the funnel aper-
qualitatively from the patterns exhibited by freely swim- ture. Additionally, there were no significant differences
ming squid in the holding tank. Finally, when removed from between the thrust measured using eitherofour force trans-
ONTOGENY OF SQUID MANTLE FUNCTION 17
measure the small thrust forces ofthe hutchlings. Thus, the
resonant frequency of the force transducer was fairly low
(about 55 Hz when unloaded in air).
We also built a second force transducer (transducer 2)
with a higher resonant frequency (about 110 Hz when
unloaded in air). It consisted of a single aluminum blade
element with semiconductor strain gauges wired as a half
bridge (Fig. IB, see Appendix). The higher gauge factorof
the semiconductor strain gauges allowed the stiffness ofthe
blade elementtobe increased without sacrificing sensitivity.
Using analysis ofcovariance (Glantz and Slinker, 2001 ),
we compared the data from the two force transducers. The
adjusted slopes and means ofthe twodatasets did not differ
significantly (ANCOVA, P = 0.82). Therefore, we pooled
the data from both transducers.
B
Each force transducer was calibrated in air after comple-
tion of the day's experiments.
Thrust measurement and mantle kinematics
Escape-jet behavior ofsquids tethered to the force trans-
ducers was videotaped using an S-VHS video camera (Pa-
nasonic AG-450 Professional). A light-emitting diode
(LED) flashing at 15 Hz was placed in the field of view of
the camera to aid in aligning the video fields with the
transducer output. The video output, the pre-amplified out-
put voltage from the force transducer, and the 15-Hz signal
from the function generator connected to the LED were
collected simultaneously on a magnetic tape data recorder
(A. R. Vetter Co.. Rebersburg, PA).
Escape-jet data were digitized from the magnetic tape at
500 Hz using MacScope MAC-600 (Thornton Associates,
Figure 1. The tethering apparatus and the force transducers. The Waltham, MA) and DATAQ DI-700 (DATAQ Instruments
dashed lines indicate the water level during the experiments. (Al Trans- Inc. Akron, OH) analog-to-digital converters, and analyzed
ducer1.Theinsetattoprightisaplanviewoftheschematicwiththesquid using MacScope and WinDAQ waveform analysis soft-
and tether apparatus removed. The bracket above the squid indicates the ware.
daolrusmalinmuamntblaesel;enBgtBhE.(DbMrLa)ss,balnaddetheelwehmietnet;arBrNo,wbproaisnstsntuot;1/N3.DnMeeLd.le;ABP,, Prior to digitization, the raw thrust-time data from trans-
stainless steel post; SG, thin foil strain gauge; W. plastic washer. (B) ducer 1 were filtered with a 45-Hz low-pass filter (4 pole
Transducer 2. The inset at left is a front view ofthe transducer with the Butterworth; Ithaco, Inc., Ithaca, NY). Because (1) the
squid removed. Abbreviations same as in A except as follows: ABE, fundamental frequency of the escape jets varied from 5 to
aluminum blade element; BB, box beam; S. screws; SSG, semiconductor 10 Hz, and (2) the contribution to the raw thrust-time data
strain gauge. by harmonics above the 4th order was negligible for all the
escape jets measured (indeed, the raw thrust-time data of
dfuucnenresl:aonndeownheicwhhihcahddtihdenpootst(sleoecaFtige.d d1o).wnstream of the opmneolasykttfehoserccfaeiprssetretjcewtoosrdhweader.rmeoTnhciecosmr)pa,owsthdeeadtfaiolfftrertoihmnegtdfriuadnndsndaoumtceeanrftfa2elctweatrnhede
not filtered because the signal-to-noise ratio was very high.
The force transducers Video sequences of escape-jet behavior were viewed
We built two force transducers to measure the thrust using a Panasonic AG-1980Pprofessional S-VHS videocas-
produced by the squid. The firstforce transducer(transducer sette recorder. Individual video fields were digitized with a
1) consisted ofthin foil strain gauges, wired as a full bridge frame grabber card (Imagenation, Beaverton. OR) and
(see Biewenerand Full, 1992), and glued to two brass blade changes in mantle diameter measured using morphometrics
elements (Fig. 1A; see Appendix fordetails). We used long, software (SPSS Science, Chicago, IL). During anescapejet,
DML
thin blade elements to obtain the sensitivity necessary to the outer diameter of the mantle at 1/3 (from the
18 J. T. THOMPSON AND W. M. K1ER
anteriormargin ofthe mantle; see Fig. 1A) was measured in 1983; Bartol, 2001). Cross-sectional area of the "white"
each video field from the S-VHS videotapes and plotted circumferential muscles was estimated as follows. The
against time. The diameter at 1/3 DML was selected be- thickness of the ventral midline of the mantle at 1/2 DML
cause this location experiences the largest amplitude change was measured in unfixed squids. The squids were anesthe-
during the escape jet (Thompson and Kier, 200la). Time tized in 7.5% MgCl2 and killed by decapitation. Blocks of
was estimated from the video camera capture rate of 60 tissue fromthe midline ofthe ventral mantle were then fixed
fields per second. and embedded in paraffin. Sections parallel to the sagittal
The LED and its voltage spike were used to align the plane (i.e., transverse tothe long axes ofthe circumferential
mantle kinematics and thrust data. The plot ofthrust versus muscle fibers) werecut at 5 ju,m and stained. The ratioofthe
time (i.e., the calibrated output of the force transducer) for thickness ofthe band ofwhite circumferential muscle fibers
the same escapejet was compared with the mantle kinemat- to the thickness of the entire mantle wall was measured
ics. from stained sections. Using the initial measurement of
mantle thickness in unfixed animals, the width ofthe white
Data
analysis circumferential muscle band could then be determined. The
The peak thrust produced during an escapejet was mea- cross-sectionDalMaLrea of the muscles was estimated by mul-
sured as the difference between the baseline and the highest tiplying the by the width of the white muscle band.
thrust achieved during the escapejet. The average ofthe 25 We also measured the maximum funnel aperture in a
digitized data points (50 ms) immediately preceding the subset of the sample population. The diameter was mea-
start ofthe escapejet was used as the baseline. The average sured with an ocular micrometer in squids that had been
of five digitized data points (10 ms) at the highest thrust anesthetized following tethering.
produced was considered the peak thrust. The peak thrust
was measured in a minimum of eight escape jets for each
squid and only the largest thrust reported. Statistics
We also determined the average rate of thrust increase
during each escape jet. This rate was calculated as the The sample population used in this study was subdivided
average slope ofthe thrust-time trace over the interval from into the life-history stages described by Segawa(1987). The
the start of the jet to the peak thrust. Only the highest life-history stages were selected as an independent organi-
average rate foreach squid was reported. In most instances, zation scheme upon which to base many of the statistical
the escape jet that yielded the highest peak thrust also had analyses. Segawa (1987) studied the life cycle of S. lesso-
the highest average slope. niana from embryo to adult and divided the life cycle into
We calculated impulse over the interval from the start of seven stages based on morphological and ecological char-
the escape jet to the point at which the thrust-time trace acters. These stages include hatchling (up to 10 mm DML),
returned to the baseline (see arrows in Fig. 2B). Impulse is juvenile 1 (11-25 mm DML), juvenile 2 (26-40 mm
the integral of the product of force and time, and is a DML). young 1 (41-60 mm DML), young 2 (61-100 mm
measure of the change in momentum of a squid during an DML), subadult (100-150 mm DML), andadult (>150mm
escape jet. The impulse was calculated for a minimum of DML). In this classification, external adult morphology is
eightescapejets foreach squid, and only the largest impulse achieved around 40 mm DML (the end of the juvenile 2
was mm
reported. stage), and onset of sexual maturity occurs at 150
DML. The sample population of S. lessoniuna used in the
Morphometrics ofmantle ami funnel current investigation included the hatchling,juvenile 1, and
We normalizedthe thrustdataforeach squidin twoways. juvenile 2 stages.
First, we divided the peak thrust ofthe escapejet by the wet Nonparametric statistics were used for most of the anal-
weight of the squid to yield the weight-specific thrust. To yses. For comparisons among the life-history stages,
determine weight, squids were anesthetized, lifted from the Kruskal-Wallis one-way analysis of variance on ranks was
anesthetic by the posteriortip ofthe mantle (to permit water used with Dunn's method ofpairwise multiple comparisons
todrain from the mantlecavity), blotted dry, and weighed to (Zar, 1996). Comparisons among all squid in the ontoge-
the nearest 5 mg on an electronic balance. netic series were made using the Spearman rank order
Second, we divided the peak thrust by the cross-sectional correlation (Zar, 1996).
area of the mitochondria-poor circumferential muscles of All log-transformed (base 10) data passed normality and
the mantle. These muscles, analogous to the white muscle homoscedasticity tests. Regression lines were fitted to the
fibers of mammals (Bone et ai, 1981; Mommsen et ai, data using an ordinary least squares approximation. All
1981), are a relevant measure because it is likely that they statistical analyses were completed using SigmaStat 1.01
provide the power forescape-jet locomotion (Gosline etai.. (SPSS Science, Chicago, IL) and JMP (SAS, Cary, NC).
ONTOGENY OF SQUID MANTLE FUNCTION 19
DML = mm DML = mm 4>
5.0 35 OcJD
03
UJS 0.05
4U1 0.00 (cJ
4> -0.05 <u
-0.10
-> -0.15 (J
u U
-0.20
c -0.25 *s*
03 03
D
40
30
20
GS/J
10
r-
-10
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Time
(s)
Figure2. Plotofmantlecircumferencechangeandthrustversustimeduringthreeescapejetsinahatchling
(DML. 5 mm) and ajuvenile 2 (DML. 35 mm) stage ofthe squid Sepioteuthis lessnniana. (A and C) Mantle
circumferencechangewasnormalizedbytherestingmantlediameterat 1/3 DML(representedbythehorizontal
lines) in the anesthetized squid. Negative values indicate contraction of the mantle; positive values indicate
hyperinflation. Each datapoint represents the mantlecircumference measurement from a single video field. (B
andD)Plotsofthrustversustimeforthesamesequencesofescape-jettinginpanelsAandC,respectively.The
trace in panel B was from transducer 1. The raw data were filteredat45 Hz(lowpass).The arrowson the plot
indicate the interval over which the impulse was calculated. The black bar to the left of the first escapejet
indicates the 50-ms interval used tocalculate the average baseline thrust. The small barat the top ofthe first
escapejet indicatesthe 10-msinterval usedtocalculatethepeakthrust. Thetrace in panel Disfromtransducer
2. The raw data are presented.
Results mantle, thrust increased rapidly (Figs. 2B, D). In hatch-
ling stage squid, the thrust peaked within 40 ms 7.4 ms
Thrust-time traces
(mean standard deviation) ofthe start ofthe escape-jet,
There were no major qualitative differences in the shape whereas injuvenile 2 stage squid thrust peaked within 71
ofthe thrust-time traces during ontogeny. At the start ofan ms 14 ms.
escape jet the mantle usually expands beyond its resting In the final event of the escape-jet sequence, the mantle
diameter. During this "hyperinflation," waterrills the mantle re-expands to its original shape and the mantle cavity is
cavity through large openings along the left and right ante- refilled. At the end of the mantle contraction and during
rior edges of the mantle. The plots of thrust versus time mantle refilling, the thrust decreased rapidly, though at a
often showed negative thrust that was correlated with hy- lower rate than the force increase during the early phase of
perinflation of the mantle as observed in the video records mantle contraction (Fig. 2B, D). In many ofthe thrust-time
(Fig. 2). traces, though most noticeably in thejuvenile 1 andjuvenile
Next in the escape-jet sequence, the mantle contracts 2 stage animals, the thrust decreased below the baseline
rapidly and ejects water from the mantle cavity through during refilling ofthe mantle cavity, implying reverse thrust
the funnel aperture. Concurrent with contraction of the (the end ofjet 1 in Fig. 2B; at the end ofjet 3 in Fig. 2D).
20 J. T. THOMPSON AND W. M. KIFR
2.5 ofeach squid. The weight-specific peak thrust ofthe escape
DML
jet increased significantly with (Spearman rank order
2.0 - correlation, r = 0.68, P < 0.001. n = 54), from 0.20 to
1.5 0.40 in hatchlings to between 1.0 and 2.5 in the largest
squids tested (Fig. 4A). Sorting the data by the life-history
1.0
stages of Segawa (1987) indicated that the mean of0.36 in
H 0.5 the hatchling stage squid was significantly lower than the
.O0<H3U 0.0 satvaegreagseqsuiodsf,0r.e5s5peacntdive1l.y5(iPn <the0.j0u1v;enTilaebleI a1)n.dInjuavdedniitlieon2,
0o0 -0.5 - thejuvenile 1 stage animals produced lower weight-specific
-1.0 - thrust duri<ng the escape-jet than did the juvenile 2 stage
squids (P 0.01; Table 1).
-1.5 The peak thrust data were also normalized by the cross-
0.6 0.8 1.0 1.2 1.4 1.6 1.8 sectional area of the "white" circumferential muscle fibers
Log Dorsal Mantle Length (mm) (i.e., the muscles that probably provide the power for es-
Figure 3. The log ofthe peak thrust generated during the escapejet cape-jet thrust; see Gosline et ai. 1983, and Bartol, 2001 ).
versus the log of dorsal mantle length. Each data point represents the Newly hatched squids generated much lower thrust per unit
highestpeakthrustproducedbyonesquid,andthustherearenoerrorbars. area of white circumferential muscle than did older, larger
Log peak thrust was regressed against the log of the DML, using an animals (Fig. 4B). Sorting the normalized thrust data by the
Jcourovdreinnneiarlreyof2letsahtseatg-pesl.qouta.reHs,mHeatthcohdl.inTghesteaqgueatsiqounidsis;iJnld.icJautveedniinleth1e sutpapgeer; lJe2f,t alsirtfaeega-ehoifsstqwourhiyidtssetapcgrieorsdcouufmcfeSederegsnaitwginaailf(i1mc9ua8ns7tcl)lyeinl(do0iw.ce2ar5tedmthtNrhu/asttmhmapte2crhtluhinaningt
)
either the juvenile 1 (0.47 mN/mm2) or juvenile 2 (1.4
Peak thrust mN/mm2) stage squids (P < 0.01; Table 1). In addition, the
jetAsincerxepaescetded,witthhe pdoeraskalthrmuasnttplerodluecngetdhdu(rSipnegarthmeanescraapnke jtuhvaennijlueven1ilseta2gestsaqgueidasnipmraoldsuc(ePd<lo0w.e0r1n;oTramballeiz1e)d. thrust
order correlation, r = 0.97, P < 0.001, n = 54). Newly
mN
hatched S. lessoniana produced between 0.10 and 0.80 Rate ofthrust increase
mN of thrust during the escape jet, whereas the largest
squids studied generated between 18 mN and 110 mN of The average rate of thrust increase of the escape jet
thrust (Fig. 3; see Table 1 for descriptive statistics based on increased significantly with dorsal mantle length (Spearman
life-history stage). Sorting the data by the life-history stages rank ordercorrelation, /- = 0.92. P < 0.001, n = 54). The
of Segawa (1987) indicated that hatchling stage squid gen- mean rate of thrust increase varied from 3 to 18 mN/s in
erated lower peak thrust than either the juvenile 1 orjuve- hatchlings and increased rapidly to between 300 and 1800
nile 2 stage squid (P < 0.01; Table 1). Juvenile 1 stage mN/s in the largest squids studied (Fig. 5). Sorting the data
squid produced lower peak thrust than juvenile 2 stage by the life-history stages of Segawa (1987) indicated that
animals during the escape jet (P < 0.01; Table 1). the average rate ofthrust increase in hatchling stage animals
The peak thrust data were normalized by the wet weight (13 mN/s) was significantly lowerthan in eitherthejuvenile
Table 1
A comparison ofthe mechanics ofthe escapejetamong squiddividedinto the life-historystages ofSegawa 11987)
ONTOGENY OF SQUID MANTLE FUNCTION 21
tfl
22 J. T. THOMPSON AND W. M. KIER
Discussion
M.inlL-wall(Y=1.29X-1.96 r=0<W>
0.0 O Whitemuscleonly(Y=1.34X-2.07 r=n.<W) The peak thrust, weight-specific peak thrust, rate of thrust
-0.2 - increase,and impulseoftheescapejet all increase significantly
during the growth of S. lessoniana. These changes in the
-0.4 - mechanics of the escape jet correlate strongly with striking
ontogenetic changes in (1) the morphology of the mantle
(Thompsonand Kier.2001b)andfunnel,and(2)the maximum
amplitudeandrateofmantlemovementduringthejet(Thomp-
sonandKier, 2001a). Furthermore, thechangesinthemechan-
ics of the escape jet have implications for circumferential
muscle mechanics and the propulsion efficiency ofthejet.
-1.2
0.6 0.8 1.0 1.2 1.4 1.6 Ontogeny ofmuscle mechanics
Log Dorsal Mantle Length (mm)
The large increase in the relative thickness of the mus-
Figure7. Mantlemorphometricdata.Logthicknessofthemantlewall culature of the mantle wall (Fig. 7) may help explain the
(filled symbols) and log thickness of the layer of"white" circumferential ontogenetic increase in the weight-specific thrust of the
omfanbtoltehmtuhsecelnetsir(eopmeanntslyembwoallls)anvedrstuhselloagyedrorosfalwhmiatnetlcierlceunmgftehr.enTthiealthmiucskcnleesss escapejet (Fig. 4A). For a pressurized cylinder such as the
was measured at 'A dorsal mantle length along the ventral midline. See the mantle, the relationship between the stress in the wall, the
materials and methods section for details. Log mantle thickness and log pressure, and the radius is given by the following equation
thicknessofthe whitemuscleswereregressedagainstthe log DML. usingan (after Fung. 1994):
ordinary least-squares method. The regression equations are located in the
=
upper left corner ofthe plot. Note the positive allometry: relative to mantle <r r, pit (Eq. 2)
length,boththethicknessofthemantlewallandthelayerofwhitemusclesare
highest in the oldest, largest animals studied. where a is the average circumferential stress in the wall, p
is the internal pressure, is the radius of the inner edge of
/-,
the mantle wall, and t is the thickness of the mantle wall.
produced less impulse during the escape jet than the juve- Solving equation (2) for pressure (/>),
<
nile 2 stage animals (P 0.01; Table 1). =
/) (rt/r, (Eq. 3)
Morphometrics ofmantle and funnel and substituting measured values ofmantle radius (Thompson
The thickness ofthe mantle wall increased from between
mm mm mm
0.070 and 0.13 in hatchlings, up to 2.5 in the
largest animals studied (Fig. 7). The slope (1.29) of the 0.4
regression ofthe thickness ofthe mantle wall against DML 11
<wL)
and the slope ( 1.34) ofthe regression ofthe thickness ofthe 3
DML
white circumferential muscle layer against were sig- OQJ.
<
nificantly greater than 1.0 (P 0.01). This indicates
positive allometry of mantle thickness, relative to DML. v, o.o 4
during growth (Fig. 7). In addition, the slopes of the two
regressions were not significantly different (P > 0.05). -0.2 -
suggesting that the positive allometric increase in mantle
wall thickness (relative to DML) was due, in large part, to p -0.4 .
the ontogenetic allometric increase in the width ofthe white 503 Y=0.65X-0.86
cirTchuemfderieanmteitaelrmofustchleefulnanyeelra(pFeirg.tur7)e.in anesthetizedsquids -ooa -0.6 r=0.87
increased during growth. The slope (0.65) ofthe regression 1 0.6 0.8 1.0 1.2 1.4 1.6 1.8
of funnel aperture against DML was, however, less than Log Dorsal Mantle Length (mm)
unity, indicating that the aperture of the funnel relative to Figure8. Logdiameterofthefunnel apertureirr.vi/.vlogdorsalmantle
body length decreased with growth (Fig. 8). Indeed, the length. Maximum funnel aperture was measured in anesthetized squids.
average funnel aperture in anesthetized newly hatched The regression was done using an ordinary least squares method, and the
squids is about 1.8 times larger (relative to mantle length) eaqlulaomteitornyisoflisttheed rienlatthieonlsohwipe:r rtihgehtfucnonrenleraopefrtthuerepliost.laNrogtesetth(erelnaetgiavteivteo
than the average aperture in anesthetized juvenile 2 stage mantle length) in the smallest squids. H, Hatchling stage squids: Jl.
squids. Juvenile 1 stage; 12. Juvenile 2 stage.
ONTOGENY OF SQUID MANTLE FUNCTION 23
and Kier. 2()01a).ofmantlewall thickness(Fig. 7A).andofthe lengths per second (L/s) in halchling S. lessonianu to a low
peak isometric stress generated by the circumferential mantle of4 L/s in mature squids. The shortening velocity ofmuscle
musclesofsquids (Milliganetai, 1997). the maximum mantle \aries inversely with the load on the muscle and the lengths
cavity pressure that can he generated during the escape jet can of the thick filaments and sarcomeres. and in direct propor-
be calculated. The peak isometric stress produced by the tion to the rate of cross-bridge cycling (Schmidt-Nielsen,
"white" circumferential mantle muscles of adult Ultimo \'iil- 1997). The force generated by a muscle varies in proportion
garis was measured to be 262 mN/mnr, but it may range as with its cross-sectional area and the length of its thick
high as 4(K) mN/mnr (Milligan et <//.. 1997). Using those filaments (Schmidt-Nielsen. 1997). Thus, shorter thick fil-
values and assuming that the peak isometric stress of the aments in hatchling S. lessoniana than in older, larger
circumferential muscles does not change during ontogeny, we squids would be consistent with the observations that (1 ) the
calculate that a newly hatched squid (DML = 5 mm) should shortening velocity ofthe circumferential muscles is highest
generate mantle cavity pressures between 2 and 3 kPa during in hatchlings, and (2) the peak thrust generated perunit area
the escape jet, whereas a juvenile squid (DML = 40 mm) circumferential muscle is lowest in hatchlings. We plan to
should produce pressures between 6 and 9 kPa. This threefold measure the contractile properties and myofilament dimen-
difference in the maximum mantle cavity pressure during the sions of the circumferential muscles from an ontogenetic
escapejet results from asubstantial ontogenetic increase in the seriesofS. lessoniana toexamine the possibility ofachange
relative thickness of the mantle wall musculature and may in performance of the muscle during ontogeny.
explain part ofthe ontogenetic increase in the weight-specific
peak thrust ofthe escapejet that we observed. Propulsion efficiencv
The ontogenetic difference in the relative thickness ofthe The propulsion efficiency of jet locomotion may be
mantle wall alone is insufficient to explain the growth- higher in newly hatched individuals ofS. lessoniana than in
related changes in the mechanics ofthe escapejet. Ourdata older, larger animals. Propulsion efficiency is the ratio of
also suggest that the contractile properties ofthe circumfer- the power output by the animal (the product of thrust and
ential mantle muscles of S. lessoniaiui change during the velocity ofthe animal) to the power input by the animal
mN
growth. Newly hatched squid generated 0.25 thrust per (the kinetic energy input per unit time). The hydrodynamic
square millimeter of circumferential muscle during the es- propulsion efficiency (r\) of the exhalant phase of the jet
cape jet. whereas the largest squids studied were able to stroke can be calculated using the following equation of
produce nearly 6 times as much thrust per unit area of rocket-motor propulsion efficiency:
lmiunsgcslecon(tFiagi.n4aB;slTigahbtlley l1)o.weArltphrooupgohrttihoenmoafnttlheeswohfitheatccihr-- 7] = (2W,)/(V: + Vf). (AndersonandDeMont.2000) (Eq.4)
ecusmcfaepreenjtetiatlhamnusthcelemafnitbleerss tohfatadpurltosvi(deThtohmepspoonwearndforKietrh,e vwehleoiceityVoifs tthheejveetlorceiltaytivoeftfolotwhepassqtuitdh.eAscqcuoirddainndg tVoy- eisqutah-e
2001b). ontogenetic differences in the proportions of "red" tion (4). the highest propulsion efficiency is achieved when
eaxnpdla"iwnhitthee"sciixrfcouldmfdeirfefnetrieanlcemuisnctlherufsitbepresraurneitinasruefaficthiaetntwteo sV,quaipdprtooxaicmcaelteersatVe.,BaercealuastieveVl,ymluoswterbejetgrveealtoecrittyhainncVrefaosresa
observed. Indeed, normalizing the peak thrust data by the
propulsion efficiency.
entire thickness of the mantle wall (i.e., both the red and Our results support the hypothesis that V is relatively
white circumferential muscles) does not change the result. lower in hatchling than in larger S. lessonianfa individuals.
The thrust produced perunit area ofcircumferential mus- First, the tunnel is proportionately larger (about 1.8 times,
cle depends not only on the contractile properties of the relative to body size) in anesthetized newly hatched squids
circumferential muscles (e.g.. thick myolilament length), than in older, larger squids (Fig. 8; also see Clarke, 1966:
but also on the funnel aperture. Control of the funnel aper- Packard, 1969; Boletzky, 1974). By the principle of conti-
ture affects the peak thrust of the escape jet and. therefore, nuity, fora given volume ofwater exiting the mantle cavity
the measurement of the "unit thrust output" of the circum- per unit time, a larger funnel aperture will decrease the
ferential musculature. Thus, the ontogenetic differences in average water velocity, while a narrow nozzle will increase
the magnitude of thrust per unit area of the circumferential the average water velocity.
muscles may depend on several variables. Second, steady-statejet thrust is proportional to the prod-
An additional observation, however, supports the pro- uctofmass flux (= massofwaterperunittimethatexits the
posal that the contractile properties of the mantle muscle mantle cavity) and the difference between the velocity of
change with growth. Thompson and Kier (200la) reported water exiting the funnel aperture and the velocity of the
that the shortening velocity of the circumferential muscles animal (Vogel. 1994). In previous work on 5. lessoniuna
(measured indirectly from the contraction velocity of the (Thompson and Kier. 200la), we showed that significant
entire mantle during an escapejet) ranged from a high of 13 differences in the relative volume of water in the mantle