Capillarity and Fibre Types in Locomotory Muscles of Wild Yellow‐Legged Gulls (Larus cachinnans)

This study analyzes the Capillarity and fibre‐type distribution of six locomotory muscles of gulls. The morphological basis and the oxygen supply characteristics of the skeletal muscle of a species with a marked pattern of gliding flight are established, thus contributing to a better understanding of the physiology of a kind of flight with low energitic requirements. The four with muscles studied (scapulotriceps, pectoralis, scapulohu meralis, and extensor metacrpi) exhibited higher percentages of fast oxidative glycolytic fibres (>70%) and lower percent‐ages of slow oxidative fibres (>16%) than the muscles involved in nonflight locomotion (gastrocnemius and iliotibialis). Capil‐lary densities ranged from 816 to 1,233 capillaries mm‐2, having the highest value in the pectoralis. In this muscle, the fast oxidative glycolytic fibres had moderate staining for succingate dehydrogenase and relatively large fibre sizes, as deduced from the low fibre densities (589‐665 fibres mm‐2). All these findings are seen as an adaptive response for gliding, when the wing is held outstretched by isometric contractions. The leg muscles studied a considerable population of slow oxidative fibres (>14% in many regions),which suggests thatthey are adapted to postural activities. Regional variations in the relative distributions of fibre types in muscle gastrocnemius may reflect different functional demands placed on this muscle during terrestrial and aquatic locomotion. The predominance of oxidative fibres and capillary densities under 1,000 capillaries mm02 in leg muscles is probably a consequence of an adapta‐tion for slow swimming and maintenance of the posture on land rather than for other locomotory capabilities, such as endurance or sprint activities.

The whole muscles were completely excised from each gull, except for the samples from pectoralis, which were selected from the midbelly of the muscle, taking special care to dissect them out entirely from the superficial to the deep part. Muscles were marked before excising in order to determine sample orientation when processing. After their removal, muscles were cleaned of excess connective tissues and left in a resting position on a flat surface, frozen in 2-methylbutane cooled to 0160ЊC, and stored in liquid nitrogen until sectioning.

Histochemical Analysis
Transverse serial sections from the muscle equatorial zone and longitudinal sections of 14 -20 mm thick were cut in a cryostat (Reichart, Jung) at 020ЊC and mounted on 2% gelatinized slides. Sections were subsequently incubated for 5 min in a buffered fixative (Viscor et al. 1992) in order to prevent shrinkage or wrinkling. This fixation procedure does not influence the fibre typing, because it does not alter the activities of the  Viscor et al. 1992). Thereafter, the following histochemifields used for fibre-type determination and capillarization meacal assays were performed according to the methods referenced: surements. Numbers in each circle are the means for the six animals of the percentages of fibre types by number (upper semicircles) succinate dehydrogenase (Nachlas et al. 1957), a-glycerophosand the percentages of fibre types by area (lower semicircles). Anaphate dehydrogenase (Wattenberg and Leong 1960), myofitomical location: A, anterior; D, dorsal; P, posterior; V, ventral. brillar adenosine triphosphatase (ATPase; Brooke and Kaiser Sector code colour: grey, fast glycolytic; black, fast oxidative glyco-1970), muscle capillary identification by the ATPase method lytic; white, slow oxidative. (Fouces et al. 1993), Sudan black B (Chiffelle and Putt 1951), and nerve-ending identification by the combined myofibrillar ATPase and acetylcholinesterase technique (Torrella et al. 1993), developed in both longitudinal and transverse sections.

Sampling Procedure and Measurements
We designed a sampling protocol for each muscle to allow regional description of fibre-type composition and capillarization. Since this protocol has been extensively explained elsewhere (Torrella et al. 1996), only a brief description follows. First, we determined the major axis (i.e., the longest diameter) of the transverse section from each muscle and divided it into several regular intervals. Thereafter, secondary orthogonal axes that transected the divisions were drawn as lines, the total length of which was also divided into several regular intervals. This procedure yielded a grid on each sample from which we Figure 1. Anatomical location of the muscles studied. Pectoralis, selected the zones or ''fields'' for obtaining data. Thus, deextensor metacarpi radialis, scapulotriceps, iliotibialis cranialis, pending on the muscle studied, from two to seven fields were scapulohumeralis, and gastrocnemius lateralis from a lateral view; scapulohumeralis also from a dorsal view.

Statistics
Data from all variables are expressed as sample means with 95% confidence limits. The capillary and fibre densities, capillary-tofibre ratio, and percentage of oxidative fibres by number and by area (considering slow and fast oxidative fibres together) were analyzed with a two-way ANOVA for each muscle, taking ''field'' (regional variability) and ''animal'' (individual variability) as factors. A multiple comparison test using Scheffé's procedure was performed in order to determine differences in sample means between all pairs of fields from the same muscle.

Fibre Types
Table 1 and Figure 5 show the fibre types found in the six muscles studied and their descriptive characteristics. It is inter- Figure 3. Transverse sections of muscle scapulotriceps (above) and muscle scapulohumeralis caudalis (below). Each section displays the fields used for fibre-type determination and capillarization measurements. Numbers in each circle are the means for the six animals of the percentages of fibre types by number (upper semicircles) and the percentages of fibre types by area (lower semicircles). Anatomical location: A, anterior; D, dorsal; P, posterior; V, ventral. Sector code colour: grey, fast glycolytic; black, fast oxidative glycolytic; white, slow oxidative.
Fibre types were classified according to the basic scheme of Peter et al. (1972), with the histochemical assays mentioned above as descriptive criteria. Measurements were made from photomicrographs taken at a magnification of 801 and 2001 with a light microscope (Dialux, Leitz, Wetzlar, Germany) equipped with a camera (Wild, MPS51, Heerburg, Switzerland). Fibre-type frequencies were obtained by counting all muscle fibres (100 -200) of the field and were expressed as percentages (number of fibres of a particular type/number of all fibres in field). The contribution of each fibre type to the total cross-sectional area of the muscle field (percentage of fibre by area) was calculated by multiplying the percentage of a fibre of a particular type by the mean cross-sectional area of that fibre type and dividing the result by the total area of all the fibres of the field. All fibre measurements were carried out Figure 4. Transverse sections of the muscles iliotibialis cranialis by means of a digitizer tablet (Calcomp 23180-4, Anaheim, (above) and muscle gastrocnemius lateralis (below). Each section displays the fields used for fibre-type determination and capillari-Calif.) connected to a personal computer using suitable softzation measurements. Numbers in each circle are the means for ware (Sigma-Scan version 1.20, Jandel Scientific, Erkrath, Gerthe six animals of the percentages of fibre types by number (upper many). Capillary density, fibre density, and capillary-to-fibre semicircles) and the percentages of fibre types by area (lower semiratio were determined from 2 1 10 5 mm 2 areas of tissue in circles). Anatomical localization: A, anterior; E, external; I, internal; each field and corrected to express them as capillaries and P, posterior. Sector code colour: grey, fast glycolytic; black, fast oxidative glycolytic; white, slow oxidative. fibres per square millimeter.
9g14$$jy08 05-28-98 11:51:50 pzal UC: PHYS ZOO esting to note that in most pectoralis areas sampled (fields numerical proportion of almost 15%. The extensor metacarpi showed similar amounts of slow oxidative fibres, 6% and 8%, labeled P.1, P.2, and P.3 in Fig. 2), two myofibrillar ATPasestaining intensities were found in fast oxidative glycolytic fibres in both of its bellies (Fig. 2). after alkali preincubation (Table 1; Fig. 5H). The predominant subtype of fast oxidative glycolytic fibres had moderate alkali Tissue Morphological Parameters myofibrillar ATPase stability, a dark staining pattern of Sudan B reaction, and a distribution throughout pectoralis muscle Table 2 shows the tissue morphological parameters studied for from 77.0% in field P.1 to 80.2% in field P.2 and 85.6% in each field. In all cases in which fast glycolytic fibres were present, field P.3 (Fig. 2). The rest of the muscle was composed of a the values of the percentage of oxidative fibres by area were fast oxidative glycolytic subtype that showed dark alkali myofrom 1% to 9% lower than the percentages by number, as a fibrillar ATPase and light Sudan B staining. No differences in consequence of the greater relative size of fast glycolytic versus the other histochemical assays were noted between them. oxidative fibres. Table 2 also shows the capillary density values, Figures 2 -4 show transverse sections of each muscle with which had a low range of variation (from 816 to 1,233 capillaries the proportion of the different fibre types for each field sammm 02 ) among the different muscle regions. Fields from iliotibpled. Fast oxidative glycolytic fibres were the dominant fibre ialis and pectoralis had the highest fibre densities (over 600 fibres type in all muscle fields. Their proportion by number was over mm 02 ). Capillary and fibre densities in the pectoralis muscle of 70% of all fibres in 19 of the 26 fields studied, which was different bird species are compiled in Table 3. especially evident for the wing muscles (Figs. 2, 3). It is also noteworthy that in the pectoralis muscle, the succinate dehydrogenase staining of fast oxidative glycolytic fibres was moder-Regional Muscle Variability ate compared with the other muscles, such as the gastrocne- Table 4 shows the significance of the differences between fields mius ( Fig. 5A and C), that had high activities. Fast glycolytic and animals, from a two-way ANOVA test for each parameter fibres were the second most common fibre type. Although in and muscle. Two findings are noteworthy. First, there is great much lower proportion than fast oxidative glycolytic fibres, individual variability among wild yellow-legged gulls in all the they were present in all fields sampled except in the pectoralis parameters studied. Second, with the exception of gastrocnemuscle (Fig. 2) and in fields G.2 and G.3 (Fig. 4) of the gastrocmius and scapulohumeralis, there are no significant differences nemius muscle. Slow oxidative fibres were not found either in between the sampled fields for most parameters when the same the pectoralis or in the scapulotriceps but were found in all muscle fields are compared. If a conservative multiple compariother muscles, generally occupying the parts of the muscle son test, such as Scheffé's procedure, is applied, no significant closest to the bone. In the iliotibialis, slow oxidative fibres differences are evident between fields, with the exception of made up about 15% in both posterior fields (fields I.3 and I.4 the percentage of oxidative fibres by area in gastrocnemius in Fig. 4) but only 5% in the anterior parts of this muscle (Table 5). A clear difference (P°0.001) for this parameter (fields I.1 and I.2 in Fig. 4). In contrast, in the gastrocnemius, was evident between the aerobically anterior (fields G.2 -G.3, slow oxidative fibres composed 17% -34% of all muscle fibres. Fig. 4) and anaerobically posterior parts (fields G.5 -G.7, Fig. In this muscle, fibres were present in anterior and medial fields 4) of the gastrocnemius. Smaller differences (0.05 ¢ P ú 0.01 but were completely lacking in posterior fields (Fig. 4). The or 0.01 ¢ P ú 0.001) between other pairs of fields indicate only wing muscle with a marked presence of slow oxidative the presence of a gradient of percentage of oxidative fibres by fibres was scapulohumeralis, which had in its anatomically deepest and most anterior part (fields S.3 and S.5 in Fig. 3) a area throughout the major axis of this muscle.
9g14$$jy08 05-28-98 11:51:50 pzal UC: PHYS ZOO is well adapted to postural activities. It is interesting to note that, in some regions of gull leg muscles, fast oxidative glycolytic together with slow oxidative fibres represent 100% of fibres (Table 2). This exclusive aerobic fibre-type presence contrasts with that found in leg muscles of ducks, which possess a large percentage of fast glycolytic fibres (Turner and Butler 1988; Torrella et al. 1996). Differences in fibre types between these species may be related to locomotor behaviour; whereas a wide range of swimming and terrestrial behaviours are observed in ducks, gulls prefer buoyant swimming and walk infrequently (Cramp and Simmons 1985).
Wing Muscles. Most wing-muscle fields sampled in the present study were devoid of slow oxidative fibres. This is in accordance with the results reported by Maier (1983), who found no slow oxidative fibres in most forearm muscles of the pigeon. The distribution of slow oxidative fibres in the present study suggests that, of the wing muscles, the deep scapulohumeralis (fields S.3 and S.5, Fig. 3) is the most adapted to a postural role, perhaps assisting in wing folding by pulling the humerus to the body. In extensor metacarpi, slow oxidative fibres, composing less than 10% of all fibres (Fig. 2), may play a role in maintaining extension of the wing and stabilizing it during gliding flight.
In the yellow-legged gull, the pectoralis was composed solely of fast oxidative glycolytic fibres, a finding in agreement with previous studies on the pectoralis of other gull species (Rosser and George 1986;Viscor et al. 1991;Caldow and Furness 1993). This contrasts with the pectoralis muscle of the pigeon or the duck, where two markedly different fibre-type populations (fast glycolytic and fast oxidative glycolytic) have been found (for reviews, see George and Berger [1966] and Butler [1991]). These two populations are proposed to support different func- the present study, two populations of fast oxidative glycolytic fibres differing in alkali stability and lipid densities were ob-Discussion served. This finding is similar to that reported by Caldow and Furness (1993) in the herring gull (Larus argentatus). Since Fibre Types and Functional Implications differences in shortening velocity between fast fibre types have been demonstrated in other muscles (see Pette and Staron Leg Muscles. The presence of slow oxidative fibres in leg muscles of birds is well documented, and it is widely accepted that the 1990), fast oxidative glycolytic fibres of dark myofibrillar ATPase and light Sudan B staining, in the gull pectoralis, might slow oxidative fibres of leg muscles are recruited mainly for postural activities (see Butler 1991). Since yellow-legged gulls be recruited during burst activity, whereas the lipid-rich fast oxidative glycolytic fibre population might be preferentially spend much time loafing and standing (Cramp and Simmons 1985), the presence of high amounts of slow oxidative fibres recruited during long glides.
Yellow-legged gulls have a low capacity for manoeuvrability in this species is not surprising. In the present study, slow oxidative fibres were found to compose 20% -40% of some in tight spaces (Cramp and Simmons 1985), performing minimal phases of burst activity, and they glide even when preparing muscle fields (Fig. 4), which suggests that gull leg musculature 9g14$$jy08 05-28-98 11:51:50 pzal UC: PHYS ZOO to alight on the water (Pennycuick 1987). The differences in for oxidative enzymes have also been reported (Tobalske 1996).
These birds perform intermittent flight, periods of flapping the use of different modes of locomotion between birds such as pigeons or ducks and gulls may be reflected in differences alternated with periods of gliding, which also requires less energy than flapping (Rayner 1985). in the histochemical organization of the pectoralis of these species. Flapping flight represents a five-to eightfold increase In addition to pectoralis, the other wing and shoulder muscles studied stand out for their high proportion of fast oxidative in oxygen consumption from resting levels (Tucker 1972), and this high energy consumption is reflected in the high levels of glycolytic fibres; all 11 fields studied have percentages higher than 70%. The small muscle cross-sectional area occupied by succinate dehydrogenase activity found in flapping-flight birds (Suarez 1992;Leó n-Velarde et al. 1993). Gliding, in contrast, fast glycolytic fibres in wing and shoulder muscles of yellowlegged gulls may reflect the only occasional need for nonsteady requires only a twofold increase in oxygen consumption from resting levels (Baudinette and Schmidt-Nielsen 1974) and in-flapping flight in this species. Likewise, the presence of a large proportion of fast oxidative glycolytic fibres may reflect a need volves, in the gull, less activity in the pectoralis than does flapping flight (Goldspink et al. 1978). The lower energy de-for endurance fibres during the long periods of time that these birds remain in flight (see Carrera et al. 1981Carrera et al. , 1993. mands of this flight mode may be reflected in the moderate levels of succinate dehydrogenase staining found in the pecto-Capillarization and Behavioural Associations ralis in the present study (Fig. 5G). This suggestion derives some support from studies of several species of woodpeckers, Leg Muscles. There are two possible explanations for the low range of variation found in the capillary density values within where fast oxidative glycolytic fibres with moderate staining 9g14$$jy08 05-28-98 11:51:50 pzal UC: PHYS ZOO the gull gastrocnemius (816 -1,066 capillaries mm 02 ) and ilio-rate (Goldspink 1981). This may result in reduced oxygen and nutrient supply by capillaries to the working muscles, a task tibialis (1,032 -1,137 capillaries mm 02 ; Table 2). First, a high percentage of slow oxidative fibres is found in oxidative zones that could be affordable by having relatively low capillary density, that is, barely over 1,000 capillaries mm 02 . Second, though of both gull leg muscles (Fig. 4). Slow fibres have lower energy turnover than fast fibres, since they hydrolyze ATP at a slower the buoyant swimming performed by gulls is a high-endurance  Note. Since the comparisons were not significant in all pairs for capillary density, fibre density, and capillary-to-fibre ratio, here we show only the results of the test for the percentage of oxidative fibres by area. The results of the comparison test between all pairs of fields are given inside of the cells. Field locations are identified in Figure 4. NS, not significant; *, 0.05 ¢ P ú 0.01; **, 0.01 ¢ P ú 0.001; ***, P°0.001. activity, the recruitment of fast fibres during this type of swim-traction (Goldspink 1981). These values agree with some carming may not be as high-energy demanding as the sustained or diovascular adaptations reported in relation to flight activity short-burst swimming activities described in other bird species in birds. Viscor et al. (1985) found that gulls have lower blood such as ducks (Aigeldinger and Fish 1995). In fact, mallard volume per unit of body weight, hematocrit, and hemoglobin ducks have greater regionalization of leg muscles, with higher concentration than pigeons. Butler and Woakes (1980) showed capillary densities in the aerobic zones and lower densities in that, in the barnacle goose (Branta leucopsis), the heart rate the anaerobic parts (Turner and Butler 1988; Torrella et al. during soaring is 50% less than during flapping flight, indicat-1996). This regionalization might reflect a wider range of locoing a significant reduction in oxygen uptake. motory activities. In gulls, less variation is present between the In all the wing muscles studied, the lack of significant differanterior and posterior parts of the gastrocnemius (Table 5). ences found between muscle regions for all the parameters This slight regionalization suggests that gastrocnemius is not (Scheffé's multiple comparison test) indicates that none of the as functionally specialized as wing muscles, possibly owing to wing muscles have great regional variations in capillarization the use of the gastrocnemius in both aquatic locomotion and or in percentage of oxidative fibres by area. This morphology terrestrial maintenance of posture.
might be a consequence of the wing muscles' specialization for gliding as a result of an adaptive constraint imposed by the need to be airborne for long periods of time. Wing Muscles. Capillary density in the pectoralis muscle of yellow-legged gulls is the lowest reported for the pectoralis of all the bird species listed in Table 3. This table shows that species performing hovering or flapping flight, such as hum-Acknowledgments mingbirds or pigeons, have higher capillary densities in their pectoralis than the two species of gulls, who have a marked This research was supported by the Direcció n General de Incomponent of gliding flight. Moreover, fibre density values vestigació n CientıB fica y Técnica grant PB93-0740. The authors found in both gull species are the lowest reported in the literagratefully acknowledge the collaboration of Mr. Marc Bosch ture for birds (Table 3). Since high capillary density and small (Departament d'Ecologia, Universitat de Barcelona) and the fibre sizes are known to be related to the high oxidative capacity Direcció General de Medi Natural of the Departament d'Agriof muscles (Schmidt-Nielsen and Pennycuick 1961; Romanul cultura, Ramaderia i Pesca (Generalitat de Catalunya) in sup-1965), these data may reflect the low oxygen demands imposed plying wild yellow-legged gulls during a campaign for the conon the pectoralis by gliding flight. Even in those fields of the trol of the population of this species in Medes Islands (Girona, other wing muscles that have oxidative proportions greater Spain). We are grateful to Dr. Mercè Durfort (Departament than 70%, lower capillary density values were found (924de Biologia Cel.lular Animal i Vegetal, Universitat de Barce-1,182 capillaries mm 02 ) as compared to the values shown in lona) for her helpful comments on microscopy. We also thank Table 3. This could be a consequence of the involvement of Jessica Whitmore for her help in editing the text. This work wing muscles in maintaining the extension of the humerus is dedicated to the memory of Prof. Dr. Josep Planas Mestres and the wrist during gliding, pulling the tendons by means of isometric contraction, which is less costly than isotonic con- (1926 -1995