Society of Integrative and Comparative Biology 2009

Loggerhead Sea Turtle (Caretta caretta) Feeding on Mackerel-Baited Longline Hooks

Marshall, C.D.,* Moss, A.L., Guzman, A.
Texas A&M University at Galveston, Department of Marine Biology,

The large longline fishery bycatch of loggerhead & leatherback sea turtles is a serious concern for both managers & conservationists.  US longliners have switched from a “J” hook to a circle hook, & switched bait from squid to mackerel; both changes are thought to decrease bycatch.  However, few studies have investigated loggerhead feeding in detail.  Therefore, we conducted a kinematic study using 6 captive loggerhead turtles feeding on barb-less, mackerel-baited longline hooks (18-0 circle vs. J-25) to characterize loggerhead feeding on mackerel. Kinematic variables for loggerheads feeding on mackerel from 18-0 circle hooks were: max gape = 6.7 cm ± 1.22, time to max gape = 798 ms ± 379.4, max gape angle = 44.6° ± 9.84, max hyoid depression = 1.4 cm ± 0.60, time to max hyoid depression = 1002 ms ± 396.5, open gape angle velocity = 517°/s ± 156.7, closing gape angle velocity = 741°/s ± 203.7, & gape cycle = 1030 ms ± 412.7.  Kinematic variables for loggerheads feeding from J-25 hooks were: max gape = 7.2 cm ± 1.2, time to max gape = 896 ms ± 524.2, max gape angle = 47.4° ± 1.03, max hyoid depression = 1.53 cm ± 0.71, time to max hyoid depression = 927 ms ± 349.7, open gape angle velocity = 516°/s ± 190.6, closing gape angle velocity = 999°/s ± 295.5, & gape cycle = 937 ms ± 434.0. No significant differences were found between hook type, with the exception of max closing gape angle velocity, which was faster when feeding on the 18-0 circle hook (P<0.01).  These data support a recent study using squid baited circle & “J” hooks, which suggested that loggerhead feeding kinematics were conserved regardless of hook type.  However, in this study it was clear that turtles were feeding further from the hook compared to squid baited hooks.

Suction and ram feeding kinematics in two divergent odontocetes

Kane, E.; Marshall, C. D.
Department of Wildlife & Fisheries Science, Texas A and M University
Department of Marine Biology, Texas A and M University at Galveston

The feeding behavior and morphology of some species of odontocetes are convergent with other aquatic suction feeders; however, few cetacean kinematic feeding studies have been conducted.  To explore suction and ram feeding in odontocetes, controlled feeding trials using 7 beluga whales (Delphinapterus leucas) and 7 Pacific white-sided dolphins (PWD; Lagenorhynchus obliquidens) were conducted.  Kinematic analysis used anatomical landmarks to characterize feeding behavior.  Mean kinematic variables for belugas were: max gape = 7.18 cm ± 2.78, time to max gape = 250 ms ± 170, max gape angle = 16.9 deg ± 4.89, max gape angle opening velocity = 118 deg/s ± 50.0, max gape angle closing velocity = 120deg/s ± 48.7, max hyoid depression = 2.80 cm ± 1.87, time to max hyoid depression = 397 ms ± 339, max prey velocity = 227 cm/s ± 98.9, and gape cycle duration = 526 ms ± 310.  Lateral gape occlusion was observed for belugas in all trials; max percent of rostrum length occluded at initial prey movement = 82.0% ± 8.35; height-to-width ratio of the pursed aperture in belugas was nearly circular.  Beluga pursed gape variables were: max pursed gape angle = 76.4 deg ± 20.7, max pursed gape angle opening velocity = 872deg/s ± 427, max pursed gape angle closing velocity = 1092deg/s ± 509.  All PWD kinematics were significantly different than belugas except max gape, max gape angle, and hyoid depression.  Some pursing in PWD was observed in most trials; max percent length occluded = 40.5% ± 25.6 and pursed aperture ratio was non-circular.  Ram-suction index was -0.03 ± 0.37 for belugas and +0.32 ± 0.40 for PWD.  Belugas appear highly specialized for the generation of suction despite RSI values.  Ram feeding was dominant in PWD, but some suction ability was observed.

Age Analysis and Population Parameters of Bottlenose Dolphins (Tursiops truncatus) Along Coastal Texas: Preliminary Analyses

Rachel Neuenhoff and Christopher D. Marshall
Department of Marine Biology, Texas A&M University at Galveston

Terrestrial mammalian intrauterine life exhibits two distinct growth patterns, the non-linear relationship (embryonic) and the linear relationship (fetal).  In general, terrestrial mammalian growth curves are constructed on a continuum of pre- and postnatal life, which reflects mammalian growth. Early studies of terrestrial mammalian growth, invited marine mammal biologists to undertake similar studies of cetaceans to validate a unifying theme in mammalian growth patterns.  To date, cetacean growth studies have precluded the prenatal portion of the dataset, reducing the ability that growth curves are accurate.  Length from stranding records and age determined by growth layer group analyses were collected from Texas coast bottlenose dolphins (Tursiops truncatus) (N=84).  Length-at-age data was analyzed with a least squares Gompertz growth model (Adj. R² = 0.84).  Preliminary analysis indicates maximal age to be 38 years.  Mean length at birth for bottlenose dolphins along coastal Texas was determined to be 106 cm, which did not differ significantly from the previously reported value of 110 cm.  Preliminary, prenatal biparietal diameter data (N=3) were linearly regressed against known gestational age (Adj. R²= 0.95).  Discrepancies in variation among marine mammal growth curves could be reduced by combining the pre and postnatal data, as is commonly done in terrestrial mammalian growth modeling.  The resulting complete pre and postnatal growth curve in bottlenose dolphins may support the hypothesis that a generalized mammalian growth pattern exists, regardless of habitat and phylogeny.


Prenatal Bottlenose Dolphin (Tursiops truncatus) Data Impacts Population Parameters Estimated by Length-at-age Growth Curves

Rachel Neuenhoff1, Heidi Whitehead2 and Christopher D. Marshall3
1Department of Wildlife and Fisheries, Texas A&M University; 2 Texas Marine Mammal Stranding Network; 3Department of Marine Biology, Texas A&M University

Accurate demographic information from long-lived, apex predators, such as bottlenose dolphins (Tursiops truncatus), is vital to efficient conservation and management strategies.  Population parameter fluctuations over time can indicate ecosystem perturbation or density compensation mechanisms.  Consistent growth models allow direct demographic comparisons among geographic regions but often, only postnatal length and age data are considered resulting in lower r2 values, non-constant variance and non-normal residual distribution.  Since early life history strategies often impact the growth rate constant (k), it is critical to include prenatal data where possible.  The ages of postnatal individuals (N=295) were estimated using standard tooth aging methodology.  Radiographic and ultrasonic images were used to estimate the gestational age of all suspected fetuses (N=391) reported stranded along coastal Texas since 1981.  Two Gompertz growth curves were fit to length-at-age data (one postnatal and one pre- and postnatal).  When prenatal lengths and ages were incorporated into the length-at-age curve, r2 increased (0.8 to 0.92), growth rate roughly tripled (0.24 to 0.99) and asymptotic length was reached earlier in life (six years).  In terrestrial animals that produce precocial offspring, maximal growth rate (curve inflection) is reached in utero.  Therefore, terrestrial vertebrate growth modelers include prenatal data in their analyses.  Odontocetes share this life history strategy by giving birth to precocial young.  When prenatal data are excluded from analyses, length- or weight-at-age models may drastically underestimate growth rate.  This is the first study to show how life history strategies can influence cetacean, population parameter estimation and better facilitate direct comparisons across regions.