Charles Messing's Crinoid Pages: Feeding Mechanisms

However, in the feather stars and stalked isocrinids, which account for over 85% of living species, arms and pinnules also function in locomotion, a dual obligation that requires a skeleton rigid enough to stand erect against a passing current, yet flexible enough to permit movement (Lawrence 1987, Messing et al. 1988, Shaw & Fontaine 1990). The muscular articulations of arms and pinnules fit this requirement as follows: as in other echinoderms, crinoid ligaments include unique mutable collagenous tissue (or catch connective tissue) capable of rapidly altering between flaccid and stiffened states (e.g., Wilkie 1983, 1984, 2005, Motokawa 1984, 1985, 1988, Wilkie & Emson 1988, Ribeiro et al. 2011).

Contraction of muscles on the ambulacral (oral) side of the fulcral ridge curls or rolls an arm inward toward the mouth and flexes pinnules toward the arm axis. When the muscles relax, the combined elasticity and contractility of the large antagonistic ligament on the aboral side of the ridge extends arms and pinnules outward for feeding (Motokawa et al. 2004). Once extended, stiffened ligaments allow the arm and pinnules to maintain an extended posture passively against a current for food gathering. Individual articulations have limited scope but an arm of over 200 segments or a pinnule with more than 50 may have great flexibility (see also Baumiller 1997, 2008). The same mechanism functions in crawling and swimming.

Fine fingerlike tube feet (or podia, singular: podion), the terminal branches of the water vascular system, occur in groups of three (triplets, or triads) along both sides of the pinnular ambulacra (food grooves). Each triad consists of a long, medium and short (or primary, secondary and tertiary) tube foot. In most photos of living crinoid pinnules, only the primary podia are visible, as rows of fine short threads along the pinnules. A similar arrangement across five families examined so far suggests that the arrangement is common to all feather stars, and probably to all living crinoids (Nichols 1960, Meyer 1979, Byrne & Fontaine 1981, LaHaye & Jangoux 1985, Holland et al. 1986, Lawrence 1987). The primary podia, 0.4-0.9 mm long (Meyer 1979, Byrne & Fontaine 1981) to at least 2.0 mm in deep-sea stalked Hyocrinidae, alternate with flap-like lappets along the ambulacral margin and, when extended, project almost at right angles outward from the central groove. The bases of secondary podia are fused to the inner surface of the lappets; their contraction pulls the lappets inward, covering the groove. In Antedon bifida and Florometra serratissima, secondary podia curl outward at an angle over the lappets (Lawrence 1987), while in Oligometra serripinna, they project upward from the groove (Holland et al. 1986). The short tertiaries extend vertically from the groove margins. Like ophiuroids, but unlike most other extant echinoderms, crinoid tube feet lack a terminal sucker. They bear papillae tipped with cilia (ostensibly sensory) and containing mucus-producing cells (Holland 1969, McKenzie 1992).  

Crinoids rely on the direct interception of particles, which reflects the operation of their feeding apparatus as an adhesive fiber filter rather than a simple sieve. The latter would retain all particles larger than the mesh size while allowing all those smaller to pass through. Crinoid tube feet are adhesive, and they clearly capture particles smaller than the spaces between them (Holland et al. 1986). The structure and function of such filters are best interpreted by aerosol filtration theory, which was first applied to biological filter feeders by Rubenstein & Koehl (1977). Briefly, although a crinoid may experience either turbulent or laminar water movements through its feeding array, at the fine-scale level of the primary tube feet where particles are actually captured, water flow is governed by viscous forces and laminar fluid motion (Vogel 1994).

Since then, the theory has been widely applied to crinoids, including fossil stalked species (Ausich 1980, Kammer 1985, Kammer & Ausich 1987), and explains a great deal about crinoid feeding biomechanics and behavior, including variations in length, spacing and posture of primary tube feet, pinnules and arms, and variations in habitat, all relative to current flow (Meyer 1979 1982a b, Liddell 1982 Leonard et al. 1988, Leonard 1989). In particular, the theory predicts that different filter arrays will function optimally under different flow regimes (Baumiller 1997, 2008, Holterhoff 1997). As an example, semi-cryptic reef-dwelling Comatulidae feather stars subject to weaker multidirectional flow have longer, more widely spaced tube feet, whereas species that perch in the open and are subject to laminar flow have shorter more crowded tube feet (Meyer, 1973 1979, Liddell 1982). However, stalked, deep-sea Hyocrinidae that feed well above the substratum also have long, widely spaced tube feet. Baumiller (1997) provides thorough review of feeding and filtration from a biomechanical standpoint.

When a suspended food particle comes in contact with a primary tube foot, it flicks, bends or curls rapidly inward, forcing the particle into the food groove. The shorter podia and lappets vary somewhat in function among species. In A. bifida and F. serratissima, the shorter podia (and, in F. serratissima, the lappets as well) scrape particles off the primaries and retain them in the groove; in A. bifida, secondary podia can also capture food (Nichols 1960, Byrne & Fontaine 1981, LaHaye & Jangoux 1985, Lawrence 1987). By contrast, the primary podia of O. serripinna perform all "conspicuous small-scale feeding acts" unassisted by secondary podia (Holland et al., 1986). Byrne & Fontaine (1981), Holland et al. (1986) and Leonard (1989) also describe the coordinated activity of multiple adjacent podia associated with capture of larger or mobile particles.

The primary podia at least are adhesive, but the role of mucus in food capture apparently varies among species. Several authors (Magnus 1963, Rutman & Fishelson 1969, Nichols 1960) describe food capture via entanglement in mucous threads and strands. Nichols (1960) describes the podia of A. bifida as forcibly ejecting mucous strands when contacted by a food particle (not a preformed mucous web), but La Touche contends that mucous threads "apparently do not occur in A. bifida" (personal communication in Byrne & Fontaine 1981, p. 17). Florometra serratissima produces mucous threads but does not shoot them out upon particle contact. Although Byrne & Fontaine (1981) suggest that food collection via threads could be an important resource, perhaps during dense plankton blooms, they consider that direct impingement of particles on adhesive primary podia is the typical collection method, a conclusion also reached by Meyer (1982) and Holland et al. (1986) for all feather stars. The latter authors found no mucous threads in O. serripinna.

Wiping of podia against each other and against the current generated by ciliary tracts on the groove floor wraps captured particles in mucous secretions and forms them into boluses which are then transported mouthward by the cilia (Nichols 1960, Byrne & Fontaine 1981, LaHaye & Jangoux 1985, Lawrence 1987). Holland et al. (1986) discuss three possible mechanisms of particle transport via ciliary action. Pinnule grooves run into arm grooves which converge like tributary rivers on the mouth.