Lisa Pfannes-Varrow 1!! Ontogenetic and inter-species scaling of Crocodile Jaw Musculature

Transcription

1 scaling of Crocodile Jaw Musculature Ontogenetic and Inter-Species Scaling of Crocodile Jaw Musculature The Royal Veterinary College Zoological Society of London MSc Wild Animal Biology Scientific paper submitted in part fulfilment of the requirements for the degree of Masters in Science in Wild Animal Biology, University of London

2 scaling of Crocodile Jaw Musculature Scientific Paper Table of Contents Introduction Methods and Materials Results Discussion Conclusions..15 References Appendix

3 scaling of Crocodile Jaw Musculature 3

4 Abstract Throughout ontogeny crocodiles undergo dramatic morphological transformations altering the size and shape of the jaw and therefore altering their biomechanical performances. As muscles are producing the bite-force it suggests that there might be scaling in the jaw musculature themselves. The relative masses, lengths and internal architecture (fibre lengths and physiological cross-sectional areas) of jaw musculature were recorded for 16 Nile crocodiles (Crocodylus niloticus) and four other species of crocodiles at various ontogenetic stages. The data was analysed using regression analysis to determine the scaling relationships with body mass and skull surface area between Nile crocodiles (Crocodylus niloticus) and inter-species. The ontogenetic study amongst Nile crocodiles revealed that there was a decrease in physiological cross sectional-areas (force) of the Supratemporalis (MST) and Externus Profundus (MAMEP). Therefore force production of the fast muscles decreased as body mass increased. The pterygoideous muscles remain important at all ages, as they scaled isometrically with body mass. Overall juvenile Nile crocodiles exhibit narrow and long snouts, whereas adult crocodiles tend to have relatively wider and shorter snouts. Inter-species study revealed that the morphological changes in muscle properties are not unique to Nile crocodiles as they too followed a similar scaling pattern. Specimens decreased in muscle force as the skull surface area increased this was not in keeping with previous studies on Alligators (Alligatoridae) suggesting Nile crocodiles show variation from these species. Key words: Crocodiles; muscle architecture; scaling; jaw musculature; biomechanical performance

5 Introduction Modern crocodiles (or crocodilians) can be separated into three families including the Crocodylidae ( true crocodiles), Alligatoridae (alligators and caimans) and Gavialidae ( Indian gharial) (D. Alderton, 2009). There are 23 living species of crocodile found across the world (D. Alderton, 2009). Nile Crocodiles (crocodylus niloticus) are found throughout sub-saharan Africa, extending to Egypt in the east, and are also present in western Madagascar (D. Alderton, 2009). Alligatoridae (alligators and caimans) are found in the southeastern USA, concentrated largely in the states of Florida and Louisiana (D.Alderton, 2009). One of the most obvious morphological features that distinguish these groups and the species that comprise them are the proportions of the skull (i.e. snout length and width) (Pierce et al 2008). These changes in skull shape have shown to have dramatic effects on biomechanical performance, such that short and wide skulls are mechanically very strong and long and narrow skulls are mechanically very weak (McHenry et al 2006; Pierce et al 2008). This association between skull proportions and overall strength has shown strong correlation with a crocodile's feeding behavior, such that those crocodiles with relatively 'weak' skulls consume small prey (e.g. Gavialis gangeticus) and those with relatively 'strong' skulls comes large prey (e.g. Crocodilius niloticus) (McHenery et al 2006; Pierce et al 2008). Beyond interspecies differences, the skulls of extant crocodilians also undergo dramatic morphological transformations during development that mimic those between species (Grigg et al. 1993). Crocodilian skulls vary in snout length and width according to their stages in ontogenetic development and across species. These changes in skull properties can alter the mechanical performance over ontogeny as seen in American alligators (Alligator mississippiensis) (Erickson et al. 2003). Mechanical performance was measured via the bite-force throughout ontogeny and showed that it scaled with positive allometry with respect to body mass, head length, jaw length, snout-vent length and total length (Erickson et al. 2003). This positive allometry was due to changes in skeletal elements throughout ontogeny however it is unknown how the jaw muscles change throughout ontogeny. As muscles are producing the bite-force, this suggests that there might also be scaling in the muscles themselves. If not then other features of the skull, such as shape, may be impacting bite-force. Scaling Changes in size or scale can have consequences that alter structure or function and so scaling can account for this among otherwise similar organisms (K. Schmidt-Nielsen, 1984). Scaling therefore can be used as an informative tool when examining morphological changes, making it possible to compare two physical parameters. For example scaling muscles mass with body mass during ontogeny and between species. Carrying out a scaling study between Nile crocodiles at different morphological stages will determine if the properties of jaw muscles change in proportion to body mass throughout morphology or not. Jaw muscles that scale allometrically with body mass/skull surface area reject the null hypothesis and raise further questions as to why this might be. Allometric scaling suggests there maybe ontogenetic changes of the crocodilian skull. Erickson et al. suggests that if results are geometrically similar there must be other reasons for changes in bite force (Erickson et al. 2003). Bite performance in alligators did 5

6 not scale to body or head size in the proportions predicted from isometric growth, instead they scaled with positive allometry. There were no changes in bite force as trophic ecology shifted (Erickson et al. 2003) If all species are the same, then the only thing that can change bite force is the shape of the skull as this could change the angle of muscle attachment and force trajectories. Reasons for allometric scaling may be due to ecological specialisations (Erickson et al. 2004), including shifts in diet during development or ontogenetic changes in skull shape. Skull shape however will not necessarily change the intrinsic characteristics of the muscles, but it might change orientation of muscle attachment, which has a big impact on producing force behind the bite. Jaw muscles power the feeding apparatus via complex, coordinated movements (Holliday et al. 2007) and may adapt to changes in diet as the crocodilian develops. A study on Chinese alligators (Alligator sinensiis) showed allometry between head and body size (Wu et al. 2006) giving evidence that these changes were as a result of changes in fights for prey. Alterations in feeding habits could have a large impact on the muscles themselves and so could results in allometry in the muscles properties when scaled with body mass. Alligator Hatchlings initially eat insects (Morurgo et al. 1991) and tiny fish, in later life their diet shifts and they consume crustaceans and small to medium-sized vertebrates and then as they reach adulthood their diet shifts again to consumption of large to medium-sized mammals and crushing of large turtles (Erickson et al. 2003, W.K. Michael, 2006). Nile crocodiles have similar diet shifts (Harold et al. 2003, W.K. Michael, 2006). A study examining the feeding habits of juvenile Saltwater crocodiles (Crocodylus porosus) showed that a large number of small prey such as crustaceans were eaten and a small number of large prey such as rats (Grahame et al. 1991). Nutrition intake levels reflect relative changes in diet (more large prey in larger crocodiles) (Grahame et al. 1991) these changes in diet could alter the frequency of jaw openings and the bite force behind them therefore altering the muscle properties. There is no statistical difference between the bite-force of captive and wild alligators, and so doesn t affect this study because they are all captive (Erickson et al. 2004). Bite force was not altered between different sexes (Erickson et al, 2003). This therefore does not propose an issue for discussion in this study. There was allometric scaling seen with head size against body mass in Caiman crocodilians which was a result of changes in skull shape throughout ontogeny (Verdade et al. 2000) These changes in skull shape throughout ontogeny are similar to what Erickson et al. found and it could be due to changes in the muscles themselves that alter the bite-force. These studies provide good evidence that a change in skull shape and size can alter the bite force during ontogeny as a result of allometric growth of individual skeletal elements and associated musculature (Erickson et al. 2003). Interestingly this study also suggested that the overall pattern of bite force increased throughout ontogeny did not vary in association with major diet shifts (Erickson et al. 2003) unlike the study Wu. et al carried out. Scaling studies to date have not explored quantitative changes of crocodilian musculature during ontogeny and between species in relation to body mass. 6

7 Muscle anatomy There are seven jaw muscles (Holliday et al 2007) that can be split into different functional groups consisting of 1 abduction and 6 adductors. The MandibulMDM (depressor mandibularis) is the only abductor muscle and is required to open the jaw (B. Busbey et al, 2005) and is positioned externally of the skull. The anterior and posterior pterygoideous (MPtA+MPtP) are the strong muscles and its line of action is most distant from the jaw joint, causing it to have a larger mechanical advantage and so a greater force is produced (Pierce et al. 2009). The MPT is therefore responsible for creating a slower, stronger bite ideal for catching tougher and larger prey. These muscles do not act during prey holding (B. Busbey et al, 2005). Supratemporalis (MST) and Externus Profundus (MAMEP) are the fast muscles, known as the adroit muscles because its line of action passes closer to the jaw joint causing it to produce more rapid but relatively weak movements (Pierce et al. 2009). MAMEP is thus responsible for the fast snapping bites produces by a crocodile; this bite is ideal for catching small fast prey. Mandibulae Externus Superficials (MAMES) and Mandibulae posterior (MAMP) are active during holding of the prey and so are known as the holding muscles (B. Busbey et al, 2005). All jaw muscles of the crocodile are internal of the skull except MandibulMDM (depressor mandibularis) (Huchzermeyer et al. 2003). The powerful jaw muscles are all on the median aspect of the mandible, thus broadening the posterior of the skull. Study Objectives Scaling the muscle properties of crocodile jaws will help determine the underlying principles of muscle biomechanics. There may be patterns, which become evident and fit in with the previous studies done by Erickson et al This study will help to validate models of crocodiles and make them more realistic enabling reconstruct of extinct fossil musculature; this will ensure they are more accurate. These reconstructions could be used to predict bite forces without having to measure them directly from the animal. An ontogenetic study will be carried out between Nile crocodiles (Crocodylus niloticus) varying at different morphological stages. It is traditional to do scaling with body mass but in this study scaling was also done with skull surface area of the crocodilians. This information allows muscle architectural properties to be extracted from museum specimens, extant and fossil crocodiles for reconstruction. Then an inter-species study will be carried out to compare the Niles to other species of crocodilians by scaling muscles properties of other species with body mass and skull surface area and comparing. Scaling Niles at different morphological stages will answer the question as to how the jaw musculature changes as body mass increases. The inter-species study will see if the relationship between jaw musculature properties and body mass remains constant between Niles and inter-species. If so, this will suggest that the results are not species dependent. 7

8 It is unknown whether there are changes in the properties of jaw musculature (length, physiological cross sectional area and mass) of the Nile crocodiles during ontogeny and/ or between different crocodilian species. Materials and Methods There were 24 crocodilians held at Hawkshead campus, varying in species and ontogenetic stages. The species being dissected were Nile crocodile (Crocodylus niloticus) (x16), Dwarf crocodile (Osteolaemus tetraspis) (x3), Mexican crocodile (species) (x3), Spectacled Caiman (Caiman crocodilus) (x1) and Black Caiman (Melanosuchus niger) (x1). These captive crocodiles were from Le Ferme aux Crocodiles in the South of France. Prior to dissections the crocodilians jaws underwent CT scans so as not to lose any vital information and the body mass (kg) of each crocodilian were measured to the nearest gram (g) and recorded. Dissection The crocodilians, once thawed, were decapitated and their bodies were then repackaged and labeled appropriately before returning them to the freezer. The skull length was measured (from the tip of the snout to the back of the skull table) and skull width (the posterior edge of the skull table between quadrates) and recorded in millimeters (mm). Photographs were taken of the skull against a scale prior to dissection and during the dissection of each labeled muscle. The Skull surface area was calculated using photographs that were taken of each crocodile skull via the program ImageJ. For each photograph the program set the scale for both width and length and calculated the skull surface area. For each specimen seven jaw muscles were dissected, labeled with the abbreviated muscle name and its orientation in the skull, these seven muscles can be separated into different functional groups (Table1) Table 1: Overview of the seven jaw muscles found in crocodiles used in this study Abbreviation (muscle name) MDM (depressor mandibularis); MPta (anterior pterygoideous) MPtp (posterior pterygoideous) MST (supratemporalis) MAMEP (externus profundus) MAMES (mandibulae externus superficialis) MAMP (mandibulae posterior) Muscle origin (D) Squamosal surface and quadrate Dorsal surface of palatine, pterygoid ectopterygoid, verntral surface of interorbital septum Caudomedial and caudolateral edge of pterygoid Ventrolateral surface of laterosphenoid lateral bridge c Ventrolateral surface parietal Rostrolateral surface of quadrate and quadratojugal Rostral surface of quadrate Note: traditional abbreviations for the muscle names are used unlike the terminology used in Holliday et al. (2007) Muscle Function Abductor- jaw opening muscle Adductor- Strong muscles Adductor- Strong muscles Adductor- Fast muscles Adductor- Fast muscles Adductor- Holding muscles Adductor- Holding muscles 8

9 After each muscle was isolated from the skull muscle length, muscle mass, fascicle length, and pinnation angle were measured. Muscle and fibre length were measured to the nearest millimeter (mm) and muscles were weighed to the nearest of a gram (g). Limited fiber pinnation is present in MDM, MAMP, MAMES, MAMEP and MST; therefore, to prevent overestimating PCSA, fibre pinnation was set to zero. Both MPtP and MPtA have fiber pinnation, so the angles of the fibers were measured using a protractor to the nearest degree. Subsequently, masses and lengths were converted to kilograms (Kg) and meters (m), respectively, for further analysis. Fibre length has a large effect on the distance over which the muscle may contract, known as the working range (Allen et al. 2010). The total area of muscle force (physiological cross-sectional area PCSA) is determined by the number of fibres present (Allen et al. 2010) and has a large effect on muscle forcegenerating capacity. The angle that the muscle fibres insert into internal tendons affects the packing of the fibres but can allow large PCSA in small volumes to exhibit large contractile forces (Allen et al. 2010). It is therefore necessary to calculate the PCSA. Using muscle mass, fiber length and pinnation angle, the physiological cross sectional area (PCSA) was determined using the following equations: Equation 1: V musc = M musc / ρ musc Equation 2: PCSA= (V musc / L fasc ) cosθ Equation 3: Fmax= PCSAσ max Note: although Fmax is the more functionally relevant muscle property, its estimation via this method is probably only qualitatively accurate. Individual muscle fibre type populations, the ability of fibers to rotate during contraction and other factors all have large effect on the force a muscle is able to apply, and are all unaccounted for in this study. For this reason we will mostly discuss muscle PCSA below. Muscle volume was calculated by dividing muscle mass by the given constant for muscle density (ρ musc = 1060kg m -3 ) (Mendez J. Keys et al., 1960) (Equation: 1). PCSA was estimated to be muscle volume divided by mean muscle fibre length multiplied by the cosine of mean muscle pinnation (θ) (Equation: 2). And muscle force was determined by multiplying PCSA by the estimated isometric stress of muscle (insert here) (Equation: 3). Once the measurements and calculations were made for each individual crocodilian dissected, scaling of each variable with respect to body mass and skull surface area was assessed using ordinary least squares regression analyses. Regression analyses Firstly the x variables (body mass or skull surface area) and the y variables (muscle mass, muscle length, fiber length and PCSA) were all logged using Log 10 in excel formulas. 9

10 Regression analysis determines the relationship of muscle architectural properties with body mass in an ontogenetic series of specimens or between species as they may vary vastly in body mass (Allen et al. 2010). In this investigation the main predictive value in linear regression will be body mass and then as a supplementary investigation, skull surface area. If the main predictive value is body mass and the muscular properties scale with geometric similarity it is expected that characteristic lengths (fascicle lengths, muscle lengths) will scale with body mass 0.33, characteristic area properties (PCSA) will scale with body mass 0.67 and characteristic masses (muscle mass) will scale with body mass 1.0. The same scaling exponents are used if the predictive value is skull surface area and the muscle properties scale with geometric similarity. To determine whether or not each variable scales with isometry or allometry, the 95% confidence intervals (CIs) around the slope (or scaling exponent) were calculated to statistically verify if the slope differs from geometric similarity. If the variable slopes - are higher than those for geometric similarity then it is said to be positively allometric (muscle properties are relatively larger than body mass), and those with slopes lower than geometric similarity are negatively allometric (muscle properties are relatively smaller than body mass). All regressions were carried out using the statistical programme Prism 5. Results Scaling regression analysis of muscle properties on body mass The results from the linear regression of muscle variables on body mass are listed in Table 3 of the Appendix. Scaling of Muscle length (m) Vs Body mass (Kg) Generally, muscle length in all seven jaw muscles shows a strong correlation with body mass mass, with R 2 values > 0.9. Within C. niloticus, all muscles scaled with geometric similarity (Cls encompassing 0.33), except the MPtA which scaled with positive allometry (Cls > 0.33). The same pattern hold true when All crocodilian species are examined, however, MAMES also appears to scale with positive allometry (Cls > 0.33). Scaling of fibre length (m) Vs. body mass (Kg) Regression of fibre length on body mass recovered R 2 values ranging between (MPtA ALL) and (MAMP Niles) Fibre lengths were found to scale mostly with strong positive allometry (Cls > 0.33) in all 7 muscles, except MPtA amongst the Niles, which scaled with geometric similarity (Cls encompassing 0.33). Scaling of muscle mass (Kg) Vs. body mass (Kg) In general, regression of muscle mass on body mass was found to correlate tightly, with R 2 values > for all muscles. Muscles that scale with geometric similarity (Cls encompassing 1.0) include MDM (All), MPtP (All), MAMES, MST and MAMP.Negative allometry of muscles mass (Cls < 1.0) was observed in MDM (Niles), MPtP (Niles) and MAMEP. Scaling of Physiological cross-sectional area (PCSA) Vs. Body mass (Kg) 10

11 In general, PCSA and body mass were strongly correlated for each jaw muscle with R 2 values > 0.9. Most muscles scaled with negative allometry (Cls < 0.67) except MPtA, which scaled with geometric similarity (Cls > 0.67). The R 2 values for MAMEP and MST were almost 0, this suggests there is little correlation between PCSA and body mass for these muscles. Scaling regression analysis of muscle properties on Skull surface area The results from the linear regression of muscle variables on skull surface area are listed in Table 4 of the Appendix. Muscle length (m) Vs. Skull surface area (m 2 ) There is strong correlation between muscle length and skull surface area, with R 2 values generall being > 0.7. (MAMES all, MAMEP all and MST all were all <0.7). Nonetheless, all muscle lengths scaled with negative allometry (CIs < 2.0). Fibre length (m) Vs. Skull surface area (m 2 ) The coefficient of determination for most muscles is > 0.7 (MAMEP all and MST all are <0.7) showing there is correlation between fibre length and skull surface area. All muscles showed negative allometry (Cls < 2.0). Muscle mass (Kg) Vs. Skull surface area (m 2 ) Generally, muscle mass when scaled with skull surface area showed a good correlated with with R 2 values > 0.7. All muscles were found to scale with positive allometry (CIs > 0.67) Physiological cross sectional area (PCSA) Vs. Skull surface area (m 2 ) Overall the coefficient of determination was a lot lower than other muscle variables scaling on skull surface area. There was little correlation between PCSA and skull surface area with R 2 values being as small as for MAMEP. Most muscles showed R 2 values around 0.7 suggesting there is some correlation; however they were in MAMEP and in MST. Within the Niles, all muscles were found to scale with negative allometry (CIs < 1.0), except for the MPtA, which scaled with geometric similarity. A similar pattern was seen for All crocodililians, except the MPtP was also found to scale with geometric similarity. Discussion Ontogenetic study of the Nile crocodile (Crocodylus niloticus) Average distribution of muscle properties Ecological specializations in terms of feeding and foraging strategies have been suggested in previous studies to be major controls on the morphological diversity of modern crocodile skulls (Pierce et al.2008; 2009, W.K. Michael, 2006). 11

12 Crocodilians with long and narrow snout have well developed MAMEP (externus profundus) and expanded supatemporal fenestra but M. pterygoideus muscles and pterygoid bone are relatively reduced (Pierce et al. 2009), H. Endo et al, 2002). Conversely, crocodilains with short and wide snouts have M. pterygoideus (MPT) and pterygoid bones that are extensively developed and MAMEP and supertemporal fenestrae that are comparatively small (Pierce et al. 2009, Iordansky, 1964). Thus there is a connection between snout length and the development of 'fast' muscles. This inverse between properties suggest that as Crocodilius niloticus increases in body mass though ontogeny the skulls become shorter relative to its width (Table 5.) In addition to this, the PCSA of the MAMEP (and MST) becomes relatively smaller as body mass increases (Table 3). As PCSA is proportional to force, these results imply a correlation between decreasing snout length and decreasing force production of the MAMEP (and MST), meaning that the 'fast' muscles are less heavily relied on in Crocodilus niloticusindividulas of ever increasing body mass. Interestingly, however, the strong pterygoideus muscles remain isometric with increasing body mass and are, therefore, seem important to C. niloticus regardless of age (Table 3). Fig. 1. Shows that MPtP has the largest muscle mass in the adult Nile and the juvenile (NNC3). MAMEP is on average has the smallest muscle mass but weighs relatively more in the juvenile than the adult Nile. The average PCSA (Fig. 2) for MAMEP was also relatively larger in the juvenile Nile (NNC3) than the adult and showed negative allometry with increasing body mass (Table 3). However from previous studies there was a positive relationship between bite force, skull dimensions and body mass (Erickson et al. 2003) so, if the juvenile C. niloticus has muscles that can produce relatively greater force that would imply that the bite force would decrease relative to increasing body size. Bite force is determined by the amount of muscle force and the distance of the bite from the back of the jaw this could suggest why juveniles have smaller bite forces because they have longer snouts and so their bite point is further away. These changes in skull dimensions between juvenile and adult Nile crocodiles could be the cause of changes in bite force throughout ontogeny or maybe this suggests that there is a difference between alligators (Erickson et al. 2003) and Nile crocodiles. Average fibre lengths normalized to body mass (Fig. 3) showed that the adult Nile had relatively larger fiber lengths than the juvenile Nile. Both MAMEP and MPT scaled with positive allometry (Table 3). Overall MAMEP had a higher average percentage contribution to total muscle force in Nile juveniles than in adult Niles (Fig.4). MPT force was higher in the adult Nile but seemed proportionate to the lesser force produced by the juvenile (Fig.4). Overall juvenile crocodiles tend to exhibit a narrow and long snout, whereas adult crocodiles tend to have relatively wider shorter snouts (Pierce et al. 2008). Muscle Architecture and hypotheses Muscle properties such as fibre length and PCSA have a larger affect on the mechanics of the crocodile jaw and may be altered by some of the ecological specializations mentioned above. 12

13 Fibre length has a large effect on the distance over which the muscle may contract, this is known as the working range. Increased working range increases the fibre length (positive allometry) this results in small bite forces and quick snapping bites. Decreased working range results in decreased fibre length (negative allometry) causing smaller bite forces and slower bites ideal for eating larger prey. As the body mass of the Nile crocodile increases fibre length becomes relatively longer (Table 3). If juveniles have relatively longer, narrower snouts than adult Niles then the working range should decrease however results in this study shows fibre lengths increase with body mass (Fig. 3.) This suggests that juveniles have shorter fibres than the adult Niles and so would expect to have larger bite forces (Erickson et al. 2003) however PCSA gets smaller relative to body mass so the bite force becomes relatively less as the crocodile gets bigger. The raw data from this study could indeed be flawed, R 2 values for MAMEP and MST were almost 0 (Table 3), this suggests there is little correlation between PCSA and body mass for these muscles. However if the data is true it goes against the hypothesis that crocodiles with longer snouts have an increased working range and so have fast snapping bites. This hypothesis was based on species with long snouts (e.g. gavialis) and the data could suggests that maybe this relationship does not hold true throughout ontogeny in on species. For future study it would be interesting to compare fibre lengths and PCSAs between crocodiles with long and short snouts to see if this hypothesis is correct. The maximum force a muscle can generate depends on the physiological cross sectional area (PCSA). It demonstrates the influence of muscle architecture on force production; if all factors are equal the pinnate muscles are capable of generating more contraction force than the muscles with parallel fibers. PCSA does not depend on muscle mass or fiber length alone. As the body mass increases amongst the Nile Crocodiles the muscle mass scaled isometric but the fiber lengths where relatively longer, the PCSA however was relatively smaller expressing negative allometry. This is because when calculating the PCSA we divide with the fibre length and if these are long compared to muscle mass it results in a smaller PCSA. When muscle mass increases due to physical development this may be due to an increase in fiber lengths, there were no relative increase in muscle mass thus PCSA is totally dependent on fiber length. Bite force is determined the PCSA and how close to those muscles the bite occurs. Bites that occur close to the muscle will be larger than those further away (Erickson et al. 2003). The Nile skull gets shorter with size so the bite will occur closer to the muscle and could therefore have the potential to increase with body size. Comparing juvenile with adult Nile crocodile using a Function space plot A muscle function space plot compares the function of each jaw muscle and can be used to compare the differences between the Adult Nile Crocodile and the smallest juvenile Nile crocodile (Fig.5). Fiber length and PCSA normalized to appropriate exponents of body mass were plotted for each of the seven jaw muscles for the Adult and the juvenile Nile crocodiles. Muscles that express relatively large PCSAs and small fiber lengths are located at the upper left of the function space plot (Fig. 5). Small fiber lengths suggest a small working range and therefore a smaller bite 13

14 force however these muscles have large PCSAs. This suggests that these muscles are able to produce large forces over a small working range and so are suitable for generating a strong bite force. Muscles with large PCSAs can be fit into small areas by arranging the fibers into a pinnate pattern; this is seen with the anterior and posterior pterygoideous (MPtA and MPtP) muscles. Short fibres however limit the excursion capacity of the muscle. This fibre arrangement suggests that penate muscles are specialized for force production but have limited ability to produce larger excursions. Muscles that express relatively longer fibre lengths and smaller PCSAs are located at the lower right of the function space plot (Fig. 5). Longer fibres suggest an increased working range and therefore an increased bite force, however the PCSAs are smaller so the bite force is smaller. Comparing the adult Nile crocodile with the juvenile crocodile it is evident that they poses different locations on the function space plot (Fig. 5). The juvenile Niles has relatively smaller fiber lengths and larger PCSAs and so the plots are located on the upper left (Fig.5). The adult Nile crocodile has relatively longer fibre length and smaller PCSAs and so the plots are located on the lower right (Fig. 5). The fast muscles MST (supratemporalis) and MAMEP (externus profundus) in the adult nile occupy relatively longer fibre lengths and smaller PCSA area of the function space than the juvenile (towards the lower right, Fig.1). This function space is interpreted as representing displacement/working rangespecialised muscles, able to produce relatively smaller forces but to contract over longer distances, and so well-suited to snapping the jaw shut. This suggests that the fast muscles in the juvenile are made for crushing and the fast muscles in the adults are made for snapping this is a discrepancy that goes against the hypothesis that longer snouted juveniles require fast snapping movements to catch their small fast prey. Inter-species study If there are differences seen amongst different species it will become evident that these changes in jaw musculature are not specific to Niles themselves. The average muscle mass, fibre length and PCSA were all normalized to body mass using the appropriate exponents and plotted for each specie of crocodile (Fig.1-3). The average percentage of muscle contribution to total muscle force was also plotted for each species (Fig. 4) The scaling study between all species of crocodilians on a whole scales very similar with body mass as scaling amongst just the Niles. Fibre length increases respectively to an increase in body mass (positive allometry) for most muscles as seen amongst the Niles (Table 3). Muscle length generally scales isometrically with increasing body mass amongst all species (Table 3). Muscle mass shows negative allometry for most muscles as body mass increases as seen amongst the Niles (Table 3). PCSA becomes smaller relative to body mass amongst all species (Table 3). In keeping with the hypothesis drawn from the ontogenetic study between Niles we would expect crocodilians with broader shorter snouts to express large strong muscles and smaller fast muscles and 14

15 crocodilians with longer narrower snouts to have more developed fast muscles than the strong ones. The Black Caiman (Melanosuchus niger) is known to have the shortest snout out of all the species analysed (Pierce et al 2008). Interestingly the strong muscles of the Black caiman are on average less in muscle mass than the Dwarf crocodile (Osteolaemus tetraspis), Mexican crocodile (species) and Spectacled Caiman (Caiman crocodilus) (Fig. 1). The strong muscles do however contribute the most force to the total force amongst all species of crocodilians (Fig. 4) meaning the bite of these broad-snouted animals will be powerful but not necessarily fast. The Dwarf crocodile (Osteolaemus tetraspis), Mexican crocodile (morelet) and Spectacled Caiman (Caiman crocodilus) in general express similar average musculature properties (fibre length, muscle mass, PCSA). However the PCSA of the MPtP muscle in Morelet is exceptionally large compared to other species (Fig. 2). Morelets have broad snouts compared to other species like the spectacled caiman which has a short snout, this could be the reason why the strong muscles are much more forceful in the morelets. Interestingly the spectacled caiman and the juvenile Nile have similar PCSAs (Fig. 2) and fibre lengths (Fig. 3) for some muscles. This could be because these two species are both juveniles whereas the others are all older and larger. The strong muscles (MPtP and MPtA) produce more force (Fig. 2) in adult Niles, Osteolaemus, Morelet and black caiman suggesting that the juveniles (Nile and spectacled caiman) do not rely on these muscles at a young age. This study suggests that all species are following a similar scaling pattern and that this trend is seen amongst all species, this would need to be supported with further study. Scaling of muscle properties with skull surface area Each species of crocodilians have different skull dimensions that could alter the jaw musculature and therefore overall mechanics. Scaling muscle properties with skull surface area takes into account these changes across ontogeny and between different species. Muscle length became relatively shorter as the skull surface area increased (negative allometry), fibre length and PCSA also became relatively less as skull surface area increased (Table 4.) This negative allometry could be because the skull gets sorter and broader as the crocodile gets older. As the skull surface area increases the muscle mass however became relatively increased (positive allometry). The muscles may become broader as the skull surface area increases, therefore increasing the muscle mass. The specimens are decreasing in muscle force as the surface area (and body mass gets bigger) this goes against Erickson et al who showed there was a positive relationship between skull proportion and bite force. This suggests that Nile crocodiles are different from alligators or that the results from this study are flawed. 15

16 Conclusions Juvenile Nile crocodiles (Crocodylus niloticus) have relatively narrower and longer snouts than Adult Niles altering the types of muscles used. Juveniles have slower and stronger MAMEP (externus profundus) and MST (supratemporalis) for catching prey with a quick snapping action; this suggests they have relatively more muscle force than adults. Similar trends were seen inter-species suggesting these morphological changes in muscle properties are not unique to Nile crocodiles. The muscle properties scaled with negative and positive allometry with skull surface area suggesting changes in skull dimensions during morphology also impact packing of muscles and their properties. Specimens decreased in muscle force as the skull surface area increased this was not in keeping with previous studies on Alligators (Alligatoridae) suggesting Nile crocodiles show variation from these species. Acknowledgements Thank you, to Stephanie Pierce for letting me carry out this project, and for inspiring and encouraging me throughout. 16

17 References 1. D. Alderton (2009) Animals. Amber books Ltd. Pp V. Allen, R. M. Elsey, N. Jones, J. Wright, J. R. Hutchinson. (2010) Functional specialization and ontogentic scaling of limb anatomy in Alligator mississippiensis. Journal of Anatomy. 216: B. Busbey (2005) Form and function of the feeding apparatus of Alligator Mississippians. Journal of Morphology, Vol 202, issue 1, pp G. M. Erickson, A. K. Lappin, K. A. Vliet. (2003). The ontogeny of bite-force performance in American alligator (Alligator mississippiensis). The Zoological Society of London. 260, G. M. Erickson, A. K. Lappin, T. Parker, K. A. Vliet. (2004) Comparison of bite-force performance between long-term captive and wild American alligators (Alligator mississippiensis). The Zoological Society of London. 262, H. Endo, R. Aoki, H. Taru, J. Kimura, M. Sasaki, M. Yamamoto, K. Arishima, Y. Hayashi. (2002) Comparative Functional Morphology of the Masticatory Apparatus in the Long-snouted crocodiles. Anatomia, Histologia, Embryologia. Vol31, issue 4, pp J. W. Grahame et al. (1991) Feeding, Growth and Food Conversion Rates of Wild Juvenile Saltwater Crocodiles (Crocodylus porosus) Journal of Herpetology, Vol.25, No, 4, pp Grigg, Gordon and Gans, Carl (1993) Morphology And Physiology Of The Crocodylia, in Fauna of Australia Vol 2A Amphibia and Reptilia, chapter 40, pp Australian Government Publishing Service, Canberra. 9. Dr. Harold, G. Cogger et al. (2003) Encyclopaedia of animals Reptiles, Amphibians and Fishes. Fog City Press. pp C. M. Holliday, L. M. Witmer. (2007). Archosar Adductor Chamber Evolution: Intergation of Musculoskeletal and Topological Criteria in Jaw Muscle Homology. Journal of Morphology. 268: F. W. Huchzermeyer (2003) Crocodiles: biology, husbandry and diseases. CAB international, Chapter 1 Crocodiles and alligators, pp Iordansky (1973) Biology of the Reptilia, Academic Press. Volume 4, pp C. R. Mchenry, P.D. Clausen, W. Daniel, M. B. Meers, A. Pendharkar. (2006) Biomechanics of the Rostrum in crocodilians: A comparative analysis using finite-element modelling. The Anatomical Record Part A. 288A: Mendez J. Keys A (1960) Density and composition if mammalian muscles. Metabolism 9, W. K. Michael (2006) The feeding ecology of yearling, Juvenile and Sub-adult Nile crocodiles, Crocodylus niloticus, in the Okavango Delta, Botswana. University of stellenbosch 16. B. Morurgo, G. Garyahu, B. Robinzon (1991) Food preference, fish attractabillity and behavior manifested toward new feed in young Nile crocodiles, Crocodylus niloticus. Physiology and Behavior, vol50, issue1, pp

18 17. S. E. Pierce, K.D. Angielczyk, E.J. Rayfield. (2008) Patterns of Morphospace Occupation and Mechanical Performance in Extant Crocodilian Skulls: A Combined Geometric Morphometric and Finite Element Modelling Approach. Journal of Morphology. 269: S. E. Pierce, K.D. Angielczky, E. J. Rayfield (2009) Shape and mechanics in thalattosuchian (Crocodylomorpha) skulls: implications for feeding behavior and niche partitioning. Journal of Anatomy. 215, pp K. Schmidt-Nielsen (1984) Scaling-why is animal size so important. Cambridge University Press. Chapter L. M. Verdade. (2000) Regression equations between body and head measurements in the broadsnouted caiman (Caiman latirostris). Rev. Brasil. Biol., 60(3): X. B. Wu, H. Xue, L. S. WU, J.L. Zhu, R. P. Wang. (2006) Regression analysis between body and head of Chinese alligators (Alligator sinensis)., Animal Biodiversity and Conservation

19 Appendix Table2:AveragemusclepropertiesintheskullbetweenNilecrocodiles(Crocodylusniloticus)atdifferentmorphologicalstagesandinter>species. 19

20 Table3:Resultsofregressionanalysisofmusclepropertiesonbodymassofdifferentspecimensofcrocodiles note:theseresultsusethelog10variables Table4:Resultsofregressionanalysisofmusclepropertiesonskullsurfaceareaofdifferentspecimensofcrocodiles note:theseresultsusethelog10variables Table5:Resultsofregressionanalysisofskullwidthandlengthonbodymass note:theseresultsusethelog10variables R 2 # y%intercept# slope## lower#95%#cl# upper#95%#cl# scaling#i/a# Skullwidthvsbodymass I Skulllengthvsbodymass A7 20

21 21 Figure1:AverageMusclemassnormalizedtobodymass 1 foreachspecimen

22 22 Figure2:AveragePCSApermusclenormalizedtobodymass 0.67 foreachspecimen

23 23 Figure3:Averagefiberlengthspermusclenormalizedtobodymass 0.33 foreachspecimen

24 24 Figure4:Percentagecontributionofeachjawmuscletototalmuscleforceforeachspecimen

25 Figure5:Functionspaceplot(fiberlengthvsPCSA,normalizedbyappropriateexponentsofbodymass)forjawmusculatureofNilecrocodiles(Crocodylus niloticus) 25