NOTE lit. “funnel cells” G.

Demosponges have either a solid type of construction, such as in the branching red sponge Clathria prolifera, or a hollow type of construction, such as in Haliclona permolis and Halichondria panicea. In the first type, Clathria, water flow is in and out basically through the same tissue mass; thus, the flows are opposing.  In the second type, Haliclona, water flow is unidirectional, entering through myriad ostia on the outer surface and exiting through large oscula (chimneys). Interestingly, despite the difference in construction the two types have similar density of choanocyte chambers (1-1.8x10⁷ chambers ^. cm³) and number of choanocytes per chamber (57-95).  Reiswig 1975 J Morph 145: 493.

Water flows (blue arrows) in Clathria oppose one another

Water flow in Haliclona is unidirectional

NOTE formerly Microciona prolifera, introduced from the Atlantic coast

NOTE it is now uncertain what the species name is for this common west-coast Haliclona; however, permolis is retained here because because many authors continue to use it to describe their specimens

Hexactinellid or glass sponges, such as Rhabdocalyptus dawsoni, live in deep water in British Columbia, a habitat possibly favoured owing to the fragility of their body structures. The major component of the body of an hexactinellid is a multinucleated syncytium, called the trabecular syncytium.It connects through cytoplasmic bridges to various cells in the sponge, such as choanocytes and archaeocytes. The origin of the syncytium is from fusion of early embryonic cells. The syncytium is cytoplasmic, lacks cell walls, possesses multiple nuclei, and extends through the entire body of the sponge.

It is the largest example of a syncytium known in the animal kingdom. Cytoplasm within the syncytium flows bidirectionally. If a cytoplasmic stream is impeded, as by a cut (drawing on left: 1) the flow temporarily reverses itself (2-3) until communication is re-established with the original cytoplasmic track (4-5). The author proposes that food products may be distributed through the sponge via the syncytium and not via cellular transport as in other sponges. Evidence that the embryos are cellular until gastrulation suggests to the author that hexactinellid sponges may have evolved from cellular sponges and that a syncytial construction may not be a primitive trait.  Leys 1995 Biol Bull 188: 241; Leys 2003 Integr Comp Biol 43: 19.

NOTE the hexactinellid sponge Rhabdocalyptus dawsoni is described as having a “flimsy” construction consisting of thin, perforated sheets and filamentous strands draped around a scaffolding of spicules, much like a three-dimensional cobweb. The strands are the ramifications of the trabecular syncytium. The syncytium makes up 75% of the organic matter in the sponge.  Leys & Mackie 1997 Nature 387: 29; for a detailed description of the histology of R. dawsonii see Mackie & Singla 1983 Phil Trans Roy Soc Lond B 301: 365.

In Rhabdocalyptus dawsoni large (1.3mm diameter), highly branched incurrent canals carry water into the sponge and large but less branched excurrent canals carry water out of the sponge.  Along the excurrent canals are many small (60µm in length) flagellated chambers.  The syncytium or trabecular reticulum that makes up the bulk of the sponge is bilayered.  One layer, the primary reticulum, encloses and supports the collar bodies and the cells that produce them (the choanoblasts), while another layer, the secondary reticulum, branches from the primary reticulum and forms a kind of barrier around the collars of the collar bodies.  Nuclei are scattered within the two retucula. Water is drawn through openings, the prosopyles, and moves through the microvilli of the collar bodies and thence into the excurrent canals to the outside.  Leys 1999 Invert Biol 118: 221.

NOTE   the author describes the trabecular reticulum as “multinucleate amoeba strung out as a cobweb”

NOTE   because the flagellated cells in hexactinellids lack nuclei, the name collar bodies is applied to them.  The cells producing them, the choanoblasts, are nucleated

Orientation of bilayered trabecular reticulum in a flagellated chamber of an
hexactinellid sponge. The primary reticulum supports the collar bodies, while the
secondary reticulum is perforated to allow the collars to poke through. Water
flow from the incurrent canal mainly passes through openings, prosopyles, and
is drawn through the microvilli collars of the collar bodies as shown by the
blue arrow on the Right. Some water passes directly through the prosopyles
into the flagellated chamber as shown by the blue arrow on the Left

In addition to the flow caused by flagellar beating, water is passively drawn out of the chimneys by currents of seawater flowing overtop. This is an example of Bernoulli’s principle, where the passage of the water current creates a negative pressure.  It is calculated that this passive flow can equal or surpass the active pumping flow even at low current velocities (e.g., as low as 6cm ^. sec^-1).  Vogel 1974 Biol Bull 147: 443; Vogel 1977 Proc Natl Acad Sci USA 74: 2069.

Seawater flowing over this sponge's oscula will tend to draw water out of them
and thus enhance water flow through the sponge. Photo courtesy Sally Leys, U Alberta

A laboratory study at the Bamfield Marine Sciences Centre, British Columbia confirms that hexatinellid sponges Rhabdocalyptus dawsoni and Aphrocallistes vastus arrest pumping in response to mechanical stimuli and sediment, doing so by propagation of electrical signals through their syncytial layers. The signals are thought to be generated by membrane depolarisation following contact with sediment or mechanical stimulus such as a glass probe, leading to calcium influx into the choanocytes accompanied by cessation of beating. 

The two species differ in their sensitivity to suspended sediment (<25µm), with pumping (confirmed by fluoresceine dye) generally resuming immediately in A. vastus, but only after sometimes prolonged periods in R. dawsoni. Greater than 4h duration exposure to sediments causes gradual reduction in pumping in both species, with recovery taking up to 25h.  During recovery, both species exhibit frequent arrests, and these arrests have a precise periodicity indicative of some sort of pacemaker involvement.  The authors suggest that the different pumping patterns in the 2 species may reflect specialisations for coping with different sediment loads, although what the ecological implications of these might be are not known.  In situ observations show the sponges will stop pumping in the presence of SCUBA-divers, but natural stimuli are likely to be fishes and sediment-disturbance by fishes. Tompkins-MacDonald & Leys 2008 Mar Biol 154: 973. Photograph courtesy Sally Leys, U Alberta.

NOTE  the sponges are collected from locations in Barkley Sound, British Columbia either by SCUBA for boot sponges R. dawsoni (30m depth) or by special manipulator arms on a remotely operated submersible (Canadian Scientific Submersible Facility, Sydney, British Columbia) for cloud sponges A. vastus (160m depth).  The sponges are transported in water collected from depth to special holding facilities at the marine lab

NOTE  the sediment is collected along with the sponges and stored for later use in these experiments.  Before use the sediment is filtered to <25 µm