Functional morphology of the cranio-mandibular complex of the Guira cuckoo (Aves)

2.1 Dissected specimens and description

Osteological descriptions follow Baumel and Witmer (1993), except where noted. A more detailed description and figures can be found in Supporting Information. The myological descriptions follow Holliday and Witmer (2007) for the adductor muscles and Vanden Berge and Zweers (1993) for the rest of the muscles.

Three adult specimens of Guira guira Gmelin, 1788 (Cuculidae, Crotophaginae), were dissected (CIT-O 278, CIT-O 279, and CIT-O 280). After dissection, the material was skeletonized and deposited in the Avian Osteological Collection of the Centro de Investigaciones en Ciencias de la Tierra CICTERRA (CIT-O). The animals were collected from the wild in Deán Funes (Córdoba), Argentina, following the protocols approved by the animal care committee and adhering to the legal requirements of Argentina, and preserved at −13.4°C. The muscles of the cranio-mandibular complex were identified and removed from their origin and insertion sites. The description of the muscles proceeds in order of appearance from superficial to deep muscles. Photographs of the muscles (taken with a Nikon Coolpix P100 digital camera) and illustrations of origin and insertion sites accompany the description.

Computed tomography (CT) scans were acquired using noncommercial equipment developed by researchers from Grupo de Espectroscopía Atómica y Nuclear (GEAN) at Facultad de Matemática, Astronomía y Física (FaMAF). In this equipment, the sample is mounted on a goniometric table with an angular resolution of 4 mrad, in order to obtain the different projections (two dimensional [2D]-images in which the X-ray beam have a certain direction), while the X-ray source and the detector remain fixed. This equipment allows obtaining conventional 2D-images of objects with an equivalent thickness of approximately 7 cm²/g and a field of view of 200 mm × 200 mm. In this article, the images were acquired using a conventional X-ray source with an anode of W, accelerating voltage of 30 kV, filters of Zr and Al to modulate the beam energy distribution. The creation of three-dimensional models is obtained from multiple 2D-images, from different angular positions, and various mathematical methods of reconstruction (Buzug, 2008). CT-data shown in this work, with a voxel size of 194 μm³, were obtained from 1,600 projections equally distributed in 360° and using the usual Filtered Backprojection (FBP) reconstruction algorithm, based on the Radon inverse transformation (Buzug, 2008), with a Shepp-Logan filter and spline interpolation. These algorithms were implemented in MATLAB software.

2.2 Biomechanical modeling

Following Sustaita (2008), we used a static two-dimensional biomechanical modeling approach (i.e., no cranial kinesis) to estimate the potential forces the jaw could exert. The jaw was modeled as a simple third-class lever (i.e., with the input force situated between the fulcrum and the output force). In this system, the adductor musculature provides the input force to close the jaw and hold the prey in the bill where the output force is exerted (Figure 1). The quadrato-mandibular joint acts as the pivot of the lever. Since most birds are isognathous (i.e., bite at the same time on both sides; Witmer & Rose, 1991), the analysis can be applied to each side. The cranio-mandibular joint was considered frictionless and the actions of the ligaments were dismissed (Sustaita, 2008).

The mechanical advantage (MA) of each muscle was calculated as the ratio between in-lever moment arm (m_i) and out-lever moment arm (m_o) (Hildebrand & Goslow, 2001). Through MA values, it is possible to establish the relative contributions of force or velocity to the cranio-mandibular complex when a muscle contracts. The in-lever moment arm is measured as the perpendicular distance between the fulcrum and the lines of action of each jaw muscle, whereas the out-lever moment arm is the straight length from the pivot to the place where the jaw contacts the trophic item. Because the actual lines of action of jaw muscles are often difficult to estimate as they have a large area of origin and/or insertion, we followed the geometric method proposed by Vizcaíno, de Iuliis, and Bargo (1998) which allows us to obtain a moment arm measurement that takes into account the range of muscle origin and insertion sites. We schematized and digitized origin and insertion sites of each jaw muscle. The center of the origin was calculated by establishing two peripheral points in the muscle origin (the most antero-dorsal and postero-ventral sites) and estimating the midpoint between them (Vizcaíno et al. 1998). One line was drawn between the most anterior part of the insertion and the center of the origin and one line was drawn between the posterior part of the insertion and the center of the origin. The angle between these two lines was subdivided into several lines of action (five in the present study). The moment arms of each line of action were measured and a mean moment arm was calculated. This procedure was repeated with the point of origin changed from the center to the most posterior and most anterior point (Figure 1). From these three mean moment arms (posterior, middle, and anterior), a new mean moment arm is calculated, which represents the moment arm for the studied muscle's line of action. The procedure is then repeated for the rest of the jaw muscles.

2.3 PCSA and estimation of bite force

The values of PCSA are proportional to the amount of tension a muscle can produce (Bock, 1974; Gans, 1982; Kaufman, An, Litchy, & Chao, 1991). The PCSA of each muscle was calculated following Sustaita (2008) and according to the equation (Lieber, 2002):

where m is muscle mass (in grams), Ɵ is the average angle of pinnation (in degrees), ρ is the density of muscle tissue (1,060 kg/m³; Pennycuick, 1996), and l is the average fiber length (in millimeters).

Pinnation angles were measured relative to the main axis of the muscle from photographs of isolated muscles and an average was calculated from 16 measures of each muscle. The dissected muscles were then immersed in 15% HNO₃ for 24 hr in order to dissolve the connective tissue that binds the fascicles, separate them, and measure their length (Sustaita, 2008). Average fascicle length was calculated from 10 fibers per muscle. All measurements were taken from digital photographs with a reference grid using ImageJ software (Schneider, Rasband, & Eliceiri, 2012). Because it is difficult to preserve the integrity of individual muscles during the process of muscle dissection, some measurements could only be taken from two specimens (Table 1). These measurements were used to calculate PCSA, which was then used to estimate bite force of G. guira. The output-force (F_o) of each muscle was calculated by the equation (Hildebrand & Goslow, 2001):

where F_i is the force of the muscle estimated from its PCSA, m_i is the in-lever moment arm, and m_o is the out-lever moment arm. Bite force was calculated as the sum of all the output-forces multiplied by 2, considering both sides of the jaws and assuming bilateral symmetry (Huber & Motta, 2004; Thomason, 1991). For comparisons with published data for other bird species, estimated bite forces were scaled to body mass. With the exception of the body mass measured for the specimens studied here, published body masses were used (Dunning, 2008).

3.1 Description of the jaw musculature

3.1.1 Depressor muscle

M. depressor mandibulae (dm)

This muscle has parallel fibers (Figure 2a) and a fleshy origin on the laterodorsal edge of the os exoccipitale, ventral to the crista nuchalis transversa (Figure 3a). The medial fibers are attached to a crest of the os exoccipitale not mentioned in previous osteological descriptions (Supporting Information Figure S3). The muscle spreads caudally to the cavum tympanicum and has a fleshy insertion on the fossa caudalis of the jaw (Figure 3h).

3.2 Biomechanical modelling, PCSA, and bite force

Mean muscle masses (M), averaged fiber lengths (FL), average angles of pinnation (A), input-forces (F_i), output-forces (F_o), PCSA, and MA of each muscle are summarized in Table 1.

Total average jaw muscle mass was 0.914 g, which represents 0.65% of mean body mass. PCSA values ranged from 2.676 mm² for the amp to 24.326 mm² for the ptd. The ptd (6.082 N) and ames (6.074 N) input-force were the highest, almost double that of the pss (the third strongest muscle). Estimated bite force for Guira guira was 6.676 N, while standardized bite force (BF/BM) was 0.047 N/g (Table 2).

The dm presented the lowest value of MA (0.05), indicating a relatively greater functional specialization for velocity, whereas the highest MA were for psp (0.284) and for ames (0.205), indicating a relatively greater capacity for force prior to velocity.

4.1 Jaw muscle anatomy, functional morphology, and force-generating capacity

This study represents the first detailed myological description published for Guira guira and the order Cuculiformes. Marceliano (1996) in her unpublished thesis describes the jaw muscle anatomy of this and another cuculiform, Coccyzus americanus Linnaeus 1758, but her description is in the context of a comparative analysis of the jaw musculature of Opisthocomus hoazin Müller 1776 and is not specific to Cuculiformes. Our study of the jaw muscles of G. guira indicates that this taxon's musculature does not present obvious differences from the typical bauplan of Neognathae (see Holliday & Witmer, 2007; Vanden Berge & Zweers, 1993): there are no losses, gains, or new configurations of jaw muscles. Marceliano (1996) states that the m. protractor pterigoidei et quadrati (pars pterigoidei) (prt(pt)) muscle is absent in this species, but in the present study it was possible to distinguish this muscle in all the specimens dissected. There are also differences in some origin and insertion features described by Marceliano (1996); for example, she affirms that the m. depressor mandibulae (dm) has two bellies but we could identify only one in all the specimens dissected.

Despite the apparently unspecialized cranio-mandibular complex of G. guira, certain features of the individual muscles can be related to its trophic habit. Cuculiformes feed using their beak tip to pluck prey which is then processed mechanically at the base of the bill (Korzun et al., 2003). Catching living prey requires quick movements. In birds, the opening of the beak is performed by two muscles (Bhattacharyya, 2013; Bock, 1964): the dm and the m. protractor pterygoidei et quadrati (prt). In the G. guira jaw muscle system, the dm has one of the highest values of muscle mass, while prt is a small muscle. Nevertheless, the PCSA value for the dm is among the lowest values of the system (Table 1). The parallel fibers of the dm and its low MA values suggest its purpose could be to produce a rapid opening of the jaw. The static morphofunctional analysis presented by Korzun et al. (2003), which was restricted to Cuculus canorus, does not include PCSA or MA values, so there are no available data to compare with our results.

Closing of the beak is carried out by the adductor muscle complex (Bhattacharyya, 2013; Holliday & Witmer, 2007; Van der Meij & Bout, 2004). Reilly, McBrayer and White (2001) discriminated the main function of each adductor group in amniotes according to their position within the complex. The internal group usually runs from the anterior palate to the back of the lower jaw. This means that, when the jaw is depressed, this muscle group is positioned perpendicular to it, producing both rapid and powerful jaw closing. The external adductor muscles extend from the side of the skull down to the back of the lower jaw and are at their most perpendicular orientation to the jaw when it is closed, so they are more effective at applying static pressure on a food item already placed in the jaw. Because the main muscles of these two groups in birds are the pterygoids and the adductor mandibulae externus, they can be placed within the context of the functional scheme of Reilly et al. (2001). In agreement with this previous work, in G. guira, both pterygoid muscles present lower MA values indicating that they prioritize velocity over force, contributing to the capture of agile prey, while they also retract the kinetic system (e.g., Bhattacharyya, 2013). Conversely, the group of external adductor muscles (ames and amep) exhibit high values of MA indicating that these muscles prioritize force. The psp was the muscle with the highest MA as a result of its distal insertion relative to the fulcrum within the adductor chamber, which increases its in-lever arm. Furthermore, it is a vertically oriented muscle with a parallel fiber arrangement. All these features indicate a greater force-generating capacity, which could be related to diminution of protraction of the upper jaw while food is in the bill. Compared to the external adductor muscles, psp does not contract isometrically, and this requires an excessive use of energy (Bock, 1999). Therefore, its highest MA value could be more related to this energy budget, rather than the maximum force capacity of the muscle.

4.2 Bite force

Bite force has a strong influence on feeding strategies in that a zoophagous species should be able to discern which prey items it is capable of killing by biting (Herrel et al., 2005; Van der Meij & Bout, 2004, 2006) and which ones will require a different strategy (e.g., the use of its hind limbs).

It has been already stated that only few bite force have been published data for birds and those are highly biased toward Passeriformes (Carril et al. 2015; Corbin et al., 2014; Degrange, Tambussi, Moreno, Witmer, & Wroe, 2010; Herrel et al. 2002, 2005, 2009; Santana et al., 2010; Soons et al., 2015; Sustaita, 2008; Van der Meij & Bout, 2004, 2006). Although these data are useful for a general comparison, definitive inferences should not be drawn from this limited information, especially given that bite forces were obtained by different methods and in different orders. Nonetheless, estimating bite force by the PCSA method has been proven to be an appropriate proxy of the real bite force (Santana et al., 2010; Sustaita & Hertel, 2010).

As listed in Table 2, zoophagous birds have the lowest values of standardized bite force, while the granivorous have the highest values. This concurs with previous studies that demonstrate that larger bite forces are needed for the processing and husking of seeds (Herrel et al., 2005; Van der Meij & Bout, 2006, 2008). Among zoophagous birds, G. guira presents the highest standardized bite force. However, these results should be interpreted with caution: (a) bite force estimates of predatory birds such as Falconiformes and Accipitriformes were made only for the beak tip (Sustaita, 2008), but the estimation for G. guira was calculated for the base of the beak (i.e., the decrease of the out-lever moment arm increases the force exerted on the food item), and (b) the diet composition and mechanism of food processing of G. guira are different than Falconiformes, Accipitriformes, and Gruiformes which prey mostly on vertebrates and process them by tearing off pieces with the tip of the beak before swallowing (Sustaita, 2008; Sustaita & Hertel, 2010). Guira guira has a diet mainly composed of insects, which it processes in a more caudal location in the beak (Korzun et al., 2003; Payne, 1997). Omnivorous birds have a less defined position within the table, which could be related to the wider range of trophic items, varying in size and hardness, in their diets. Using a functional ecological classification of organisms that takes into account the material properties of their prey items could be a better approach to analyzing the relationship between cranio-mandibular complex morphology, bite force, and feeding ecology than focusing on traditional dietary categories (Santana et al., 2010). For example, in the current classification, it might be expected that Zenaida macroura would have a higher standardized BF because it is a granivore. However, this does not take into account the fact that this species feeds on smaller and probably softer seeds than finches which have the highest BF/BM values.

4.3 Dietary implications

Guira guira is reported as the southern-most ranging cuculid with the broadest diet in terms of size and variety of food items ingested (Beltzer, 1995; Repenning et al. 2009). The literature classifies G. guira as a zoophagous bird (Beltzer, 1995; Bernal de Pereyra, 1927, 1935; Coutinho et al. 2014; Daguerre, 1922; Koski & Merçon, 2015; Morais et al. 2013; Payne, 1997; Repenning et al. 2009) among which it is occasionally characterized as an exclusively insectivorous species (Goijman & Zaccagnini, 2008; Legrand, 1934; Leveau & Leveau, 2004). Occasional consumption of large vertebrate prey by the species is likely mediated by food processing prior to swallowing, widening the range of potential prey size and type (Herrel, Damme, Vanhooydonck, & Vree, 2001; Hespenheide, 1966; Korzun, Erard, & Gasc, 2004; Korzun et al., 2003; Van der Meij & Bout, 2006).

With respect to the feeding strategy, this species is an active forager that usually feeds in a flock of a few individuals. Field observations and stomach content analyses indicate that this species prefers smaller and more abundant prey whenever available, likely because their capture demands a lower energy expenditure. But during seasons when standard food is limited, G. guira has been reported to ingest larger prey (Beltzer, 1995; Martins & Donatelli, 2001), an action probably facilitated by the structural and biomechanical features of its jaw musculature. This opportunistic diet shift of the Guira cuckoo underlines its dietary versatility and ability to effectively use available food resources.