top 
Today the sponge reefs discontinuously cover about 1000 km2 of seafloor
in Queen Charlotte Sound and Hecate Strait in depths between 165 and 240
metres. Four separate locations have been identified where sponge reef
complexes have formed. The reefs consist of very dense populations of
living hexactinosidan sponges, often more than 1 m in height. The mounds
are up to 21 metres high and often many kilometres wide. Sometimes they
show remarkably steep slopes, up to 90°, on the flanks. In addition
to the mound, or biohermal structures, there are also large biostromal
structures which cover many square kilometres of seafloor. These reach
thicknesses from 2 to 10 m in southern and central Queen Charlotte Sound.
The growth form of individual mounds is characterized by an initial growth
phase as a discrete, symmetrical circular form, usually found on an iceberg
scour berm. These small
mounds grow over time and coalesce to form larger irregular structures.
This aspect of the formation of these features is similar to deep-water
coral (Lophelia) reefs of the northeastern Atlantic (Freiwald et al.,
1999).
Growth rates of hexactinosidan sponges range in the order of 0-7 centimetre
per year (see below). The individual mounds expand by accreting silicisclastic
sediment to the surface of mounds through sediment trapping and seabed current
baffling by the sponges. This results in the development of silicate mounds of
more irregular shapes, including ridges, coalescing lobes and even biostromes.
At scales both of the entire reef complex, and individual mounds, the
reefs elongate in the predominate direction of the seabed current. Small,
circular mounds are seen in the peripheral areas of the complex. In most
areas the biostromal deposits do not achieve the thickness of bioherms
and this suggests a style of growth or accretion of sediment which occurs
at the periphery of the bioconstruction. The stiffness of the reef sediments
increases with depth from watery and soft near the surface, to firm below
about 1 metre depth into the seafloor. Cores show the sponge reef sediments
to be unconsolidated to the base of the reef, with some dewatering of
the sediment apparent.
Dead sponges sediment over and do not lose the structure of the living
organism but become buried in place with little apparent post-mortem change
in the morphology of the skeleton of the sponge. The durable framework
of massive dead hexactinosidan skeletons projecting out of the seabed, provides
a firm substrate which allows young sponges to become anchored to an otherwise
soft, clay covered seafloor. The sponge reefs represent the entire Holocene
section over much of the sediment-starved shelf.
The sediment within the sponge bioherms is characterized by a high levels
of organic carbon. The large amount of organic carbon (more than 3 weight
percent) found in the reef sediments is similar to that found at modern
deltas on the west coast of Canada (Bornhold, 1978). Reducing conditions,
which are usually observed in the recovered cores and grab samples in
the shallow subsurface in reef sediments, are probably explained by this
high organic carbon content. These dys- or anoxic conditions are responsible
for the paucity and low diversity of endobenthic organisms.
Based on the analysis of reef core Tul99A09 the average amount of amorphous
silica is about 14-17% whereas the carbonate content is 10-15%. Noteworthy
is that all carbonate is of biogenic origin, derived from the shells of
organisms. The fine fraction of the siliciclastic sediment consists of
clay minerals. XRD-analysis of 24 samples revealed following clay mineral
contents:
Smectite 42-50% (mean 46%),
Chlorite 3-10% (mean 7%),
Illite/Smectite 23-30% (mean 26%),
Illite 4-8% (mean 6%) and
Kaolinite 12-20% (mean 15%).
top
The sponge fauna of the reefs consist of only 7 hexactinellid species,
and of these only three of them are very abundant and form the framework
of the reefs. They include the hexactinellid hexactinosidan species Heterochone
calyx, Aphrocallistes vastus and Farrea occa. Siliceous sponges that were
often abundant in the reefs, but are not considered to be reef builders,
included the hexactinellid lyssacinosidan (rossellid) species Rhabdocalyptus
dawsoni, Acanthascus platei, Acanthascus cactus and Staurocalyptus dowlingi.
In some areas the reef surface has a dense coverage of monospecific sponges
over many square metres. Farrea occa often form such large clusters.
Large individuals of the funnel-shaped Heterochone calyx are often found,
reaching sizes up to 1.5 metres in height and 70 centimetres in diametre.
H. calyx is easy to identify due to their very characteristic fingerlike,
radial protrusions of lateral body wall sometimes longer than 20 centimetres
and 5 cm in diametre. These protrusions often have small opening at their
ends. The body wall is up to 1 centimetre thick. The terminal osculum
is round to oval and wide open. In colour this sponge is white to bright
yellow. H. calyx attaches only to hard substrate.
Aphrocallistes vastus has a funnel- or cup-like morphology with radial
flat hollow and ovenglove-like extensions closed at their ends. A. vastus
is usually slightly smaller than H. calyx but can reach also remarkable
sizes greater than 2 metres in height. The terminal osculum is round and
open. The body wall is never thicker than 5 mm and usually around 3 mm
thick. The colour of A. vastus is also yellow, but darker than H. calyx.
It is quite abundant in the reefs, but never forming such big cluster
as H. calyx or Farrea occa (see below). A. vastus is firmly attached to
hard substrate by spreading a basal plate onto rocks, boulders or other
sponge skeletons.
The body shape of Farrea occa is a dichotomously branching round tube
with open lateral branches. Oscula are round to oval. F. occa is very
thin-walled, not thicker than 2 mm, consisting only of two layers of fused
hexactin spicules resulting in a very brittle skeleton. It forms tremendous
large stocks bigger than 15 metres in diametre. F. occa is the most important
sediment baffler in the reef. Due to its enormous size, F.occa slows bottom
currents, transporting sediment across the reef leading to the deposition
of suspended sediment. The baffled sediment is an organic rich siliciclastic
clay (see below), transported in suspension as flocs by seafloor currents.
Like all hexactinosidan sponges, F.occa is firmly attached to hard substrate
with its basal plate.
The accompanying lyssacinosidan hexactinellid sponges (Rhabdocalyptus dawsoni,
Staurocalyptus dowlingi, Acanthascus cactus and A. platei) are all boot-shaped
tubes with a large atrial cavity, a single round osculum and up to 50
centimetres in height. They have a whiteish colour but often their dense
hypodermal spicule veil is filled up with sediment and a community of
microorganisms called "spicule jungle" (Mackie & Singla,
1983) and turns their colour to greyisch or brownish. Of these four species
Rhabdocalyptus dawsoni and Staurocalyptus dowlingi are most abundant.
They occur randomly within the reefs as well as outside of them. Lyssacinosan
sponges settle normally in soft sediment but can also attach to firm or
hard substrate such as rocks or boulders.
Demospongiae occur within the siliceous sponge reefs as well, but mostly occur as
isolated individuals outside the reef. This group is represented only
by several species. Mycale bellabellensis and branching Iophon chelifer are predominant.
These sponges usually dwell on large boulders. Rare specimens of Geodia
sp. were also observed.
top
Taxa identified in association with the sponge reefs include several
species of annelid worms (Terebella sp., serpulids), bryozoans (often
encrusting dead sponge skeletons), bivalves and gastropods (both very
rare). Several species of rockfish (e.g. Yellow eye, Sebastes ssp.) occur
in some areas of the reefs utilizing openigs and niches between the sponges.
It is possible these fish are using the reefs as a "kindergarten"
during their juvenile stage. Species of spider crab and King crab, shrimp,
prawns and euphausids are locally very abundant. Many different species
of echinoderms, notably seastars and urchins, were common in those parts
of the sponge reefs, where sponges are dying or obviously in poor condition.
The relative amount of these echinoderms can thus be used as an indicator
for the state (health) of the sponge reefs.
As mentioned above, endobenthic and semiinfaunal organisms are rare. Among
those are terebellid serpulids most common. Living bivalves or any other
larger organisms were only rarely detected. Sparse pelecypods have been
recovered from cores including Thyasira gouldi, and Thyasira flexuosa,
(Conway, et al., 1991) and these are species adapted to low oxygen levels
within the sediment (Paul Johnston - personal communmication, 2000). This
is a further indication of the low oxygen concentration and reducing conditions
within the fine grained sediment.
To date we have no indication that anything is feeding on living sponges.
This is no surprise as living hexactinellid sponges have only very little
organic tissue which could serve as nutrition.
The microfauna is dominated by abundant foraminifera, whereas diatoms
are rare but always present. A study (Jean-Pierre Guilbault, Montreal;
Canada) of 31 samples (subsurface core samples and surface samples) collected
on the bioherms during the cruise yielded more than 200 species of foraminifera,
of which some only occur in small numbers. The proportion of planktonic
foraminifera ranges from 0 to 10 % in all samples, probably indicating
comparable surface conditions at all sampling points. With minor random
variations, all samples contained the same assemblage, where the two major
constituents are Epistominella vitrea and Bolivina decussata. Other major
species recorded are Eilohedra laevicula, Seabrookia earlandi, Angulogerina
spp., Lobatula fletcheri, Cassidulina reniforme and Astrononion gallowayi.
Attached to sponge skeleton were the following genera: Lobatula, Ramulina,
Tumidotubus, cf. Crithionina, Trochammina and Trochamminella.
The coarse fraction (fauna bigger than 1 mm) contains more arenaceous
species, whether more primitive (Psammosphaera, Rheophax scorpiurus, Haplophragmoides,
Cribrostomoides) or more complex (Martinottiella, Karreriella, Rhumblerella).
These genera are quite common on the shelf off British Columbia. The occurrence
of Ammodiscus arenaceus, however, is a surprise as this species is usually
found in the deep sea, at abyssal depths. Some of the foraminifera are
supension feeders (e.g., Rhabdammina, Ammodiscus) which suggests low sedimentation
rate and non-turbid waters.
Surface samples (done with the slurp gun) are characterized by a high
proportion of deep sea tubular forms (e.g., Rhabdammina abyssorum, Pelosina
ssp., Tolypammina vagans, Ammodiscus arenaceous). Obviously these samples
were collected at locations where sedimentation rate is very low, both
temperatures and salinity are uniform and the food supply was constant.
In effect, the foraminferal assemblage mimics deep sea conditions.
top
All modern and most fossil hexactinosidan sponges need a hard substrate
to settle. This hard substrate was provided by icebergs, grounding and
ploughing the sea floor. The fine grained sediment fraction of the berms
on both sides of the iceberg scours was washed out and the remaining cobbles
and boulders form a suitable substrate for the initial settlement of Hexactinosida.
After the death and mazeration of hexactinosan individuals, their skeleton
plays an important role as a substrate for other hexactinosidan sponges.
Their larvae attach to these skeletons and develop into a juvenile sponge
using the fibres of the substrate skeleton for a solid fixation. Little
is known about the larval stage of hexactinellid sponges. Our investigations
cannot contribute to understanding of this early stage of sponge development
and we have to refer to OKADA (1928). The skeleton of juvenile hexactinosidan
sponges cannot be used for a sound taxonomical identification as the size,
shape and architecture of the juvenuiles spicules differ remarkably from
the adult individuals. Juvenile sponges of Lyssacinosida, for example, have
a rigid skeleton whereas the spicules of adult lyssacinosidans are isolated
not fused together (Mehl, 1992). The skeletons investigated are fairly
irregular without a obvious (visible) prefered orientation. The size of
very young sponges is around 1 mm in diametre and globular in shape. In
later stages of development, the shape is more variable, and depends mostly
on the orientation of the substrate skeleton.
A juvenile sponge attaches by means of tendril-like spicules wrapping
around the substrate skeletons. The tendrils form a dense reticulate meshwork
which, culminates in the formation of a completely closed envelope or
layer around the substrate spicule. The thickness of the tendrils varys
between less than 10µm and more than 50µm (Neuweiler, 2000).
Usually there is a small gap between the tendrils and the substrate skeleton.
This gap probably is due to the sponge tissue enveloping the spicules
of the juvenile sponge. Later, this gap is closed by additional silica
deposition which begins with the forming of pillar-like structures. These
tendrils are long and cover a relatively large area of the later formed
often root-like basal plate.
After reaching a discrete size (about 2-3 mm), the young sponge slowly
starts to develop more regular hexactin spicules. The rays of these hexactins
are linked by tendril-like extensions. After this stage of development,
the sponge finally begins to form a characteristic architecture of its
spicules tracts. As we only had dead sponges and their skeletons respectively,
nothing can be said with regard to the microscleres. These diagnostically
important spicules were not preserved together with the dictyonal framework.
This mode of attachment is common to all analysed specimens and is clearly
visible by means of SEM.
top
No direct observation or measurement was done to establish the growth
rates of hexactinosidan sponges. Long term observations in the Antarctic
indicate a very slow growth rate of rossellid Lyssacinosida (Hexactinellida).
Over 10 years of observation there was no remarkable change in body size
(Dayton, 1978).
Individuals of Aphrocallistes vastus, some 60 centimetres high were found
on an underwater cable which had been in place for 52 years (Levings&
McDaniel, 1974). If these animals settled immediately after cable installation,
the average growth rate would be about 1 centimetre per year.
In June 2000, a gas pipeline was surveyed by means of an underwater video
system at Secret Cove (near Pender Harbor; British Columbia, Canada).
Moderately large populations of Aphrocallistes vastus occurred on the
pipe at depths of 100 to 120 feet. Their form was irregular but roughly
ellipsoidal. The size of the sponges were determined relative to the pipe
diametre (25,4 cm). Many had a height of 27-30 centimetre while one was
about 57 centimetre and another about 64 centimetre. The pipeline was
surveyed in September 1991 one year after installation. At that time there
was little evident biota settled on the pipe. Based on a maximum time
interval of nine years from sponge settlement to measurment, the minimum
average growth rates range from 3 to 7 centimetres per year (if the sponges
settled in 1991). Aphrocallistes vastus often reaches sizes of 1 metre
and more. At a growth rate of 7 centimetre per year, a sponge could reach
1 metre in height in 14 years, assuming growth rates continued unchanged.
If settlement had occurred some time later, then the growth rate is much
higher. It is likely that growth rates depend on nutrient availability,
reproductive activity, stability of the water and oceanographic conditions
and will also be highly variable during the lifespan of each individual
sponge (written commun. Dr. Bill Austin, Marine Ecology Station, Sidney; B.C., Canada).
Three years of observation on the lyssacinosidan sponge Rhabdocalyptus dawsoni
(Leys & Lauzon, 1998) provided new and reliable data on growth rates
of this hexactinellid species. The mean growth rate (measured by 19 individuals)
is around 2 centimetres per year (maximum 5,4 cm per year; minimum -0,7
cm per year). A negative growth rate is the result of a possible shedding
of their outer spicule coat between November and March (Leys & Lauzon,
1996). Noteworthy is that some sponges showed no growth while other individuals
increased in size within the observed time span. The growth rate also
depends on the size (= age). While smaller sponges (length less than 25
centimetre) grew 12,6% per year in relation to their length, larger sponges
grew (longer than 25 centimetre) 6% per year. Smaller sponges thus grow
faster than larger sponges.
This indicates that the growth rates of hexactinellid sponges are within
the same range of the growth rates of modern corals and are furthermore,
comparable to the growth rates of demosponges.
top
In order to estimate the age of hexactinellid sponges, the growth rates alone are not sufficient data, because the volume of the sponge body increases to the power of three while the length increases only linearly. A better way to estimate the age is by means of the increasing volume of the sponge body. After Leys& Lauzon (1998) a mean growth rate of 2 centimetres per year corresponds to 167 ml. A sponge, 32 centimeter tall with a volume of 5,8 l, would be 35 years old (if the growth rate is constant. Large individuals (e.g. 36,8 l and nearly 1 m long) are at least 220 or more years old (calculated with a constant average growth rate of 167 ml per year). Growth rates are usually not constant, but vary within broad limits and decrease as the sponges increase in size. Therefore we consider the above mentioned arithmetical ages to be fairly reliable.
top
Siliceous sponge spicules consist of amorphous biogenic silica. This
opaline material is supposed to dissolve shortly after the death and maceration
of the sponge. This is discussed by numerous authors, mainly dealing with
fossil or sub-fossil sponges (e.g., Friedman et al., 1976; Land, 1976;
Lang, 1989; Narbonne & Dixon, 1984; Rigby, 1986).
Thorough investigation on hundreds of samples by means of SEM no evidence
was seen for dissolution marks or corrosional patterns on the spicules.
Even on those spicules from the base of the bioherms, derived from cores,
was there any indication of dissolution features. These spicules are up
to 9000 years old and show remarkable preservation, there being essentially
no difference from the youngest, most recent sponge spicules obtained
from living sponges at the top of the bioherm. This is probably due to
the fact that the fine-grained matrix sediment consists of large proportions
of clay minerals. In this siliciclastic regime, a high saturation of dissolved
silica in the seawater prevents spicules from dissolution.
Transformations from unstable amorphous silica into more stable modifications
is a maturation process which depends on facies, burial and time. A possible
diagenetic alteration of the sponge spicules was examined by means of
Raman spectroscopy. The results clearly show that no transformation into
a more stable phase of silica took place. Even the oldest sponge spicules
(9000 years old) from the base of the mounds remain unaltered and still
consist of amorphous silica. It is possible that the available time span
is too short and/or that the burial depth (the mounds are "only"
18 m high) is too small to provide the required conditions for alteration.
top
Bornhold, B.D. (1978): Carbon/nitrogen ratios (C/N) in surficial marine sediments of British Columbia. - Current Research, part C, Geological Survey of Canada, Paper 78-1C, Ottawa.
Conway, K. W., Barrie, J. V., Austin, W. C. & Luternauer, J. L. (1991): Holocene sponge bioherms on the western Canadian continental shelf. - Continental Shelf Research 11: 771-790, 10 figs., 1 tab., London.
Dayton, P. K. (1978): Observations of growth, dispersal and population dynamics of some sponges in McMurdo Sound, Antarctica. - In: Lévi, C. & Boury-Esnault, N. (eds.) Observations of growth, dispersal and population dynamics of some sponges in McMurdo Sound, Antarctica. - Colloques Internationaux du Centre National de la Recherche Scientifique, 291: 271-282, Paris.
Freiwald, A., Wilson, J. B. & Henrich, R. (1999): Grounding Pleistocene icebergs shape recent deep-water coral reefs. - Sedimentary Geology 125: 1-8, 3 figs., Amsterdam.
Friedman, G. M., Syed, A. A. & Krinsley, D. H. (1976): Dissolution of Quartz accompanying Carbonate Precipitation and Cementation in Reefs: Example from the Red Sea. - Journal of Sedimentary Petrology 46(4): 970-973, Tulsa.
Land, L. S. (1976): Early dissolution of sponge spicules from reef sediments, North Jamaica. - Journal of Sedimentary Petrology 46(4): 967-969, 3 figs., Tulsa.
Lang, B. (1989): Die Schwamm-Biohermfazies der Noerdlichen Frankenalb (Urspring; Oxford, Malm): Mikrofazies, Paloekologie, Palaeontologie. - Facies 20: 199-274, 26 figs., 9 pl., 3 tab., Erlangen.
Levings, C. D. & McDaniel, N. G. (1974): A unique collection of baseline biological data: benthic invertebrates from an underwater cable across the Strait of Georgia. - Fisheries Research Board of Canada, Technical Report 441: 1-19.
Leys, S. P. & Lauzon, N. R. J. (1996): Hexactinellid ecology: Growth and seasonal regression in Rhabdocalyptus dawsoni. - Abstract, International Conference on Sponge Biology, pp. 29, March 12.-16., 1996 Otsu, Japan.
Leys, S. P. & Lauzon, N. R. J. (1998): Hexactinellid sponge ecology: growth rates and seasonality in deep water sponges. - Journal of Experimental Marine Biology and Ecology 230: 111-129, 8 figs., Amsterdam.
Narbonne, G. M. & Dixon, O. A. (1984): Upper Silurian lithistid sponge reefs on Sommerset Island, Arctic Canada. - Sedimentology 31(1): 25-50, Amsterdam.
Neuweiler, M. (2000): Untersuchungen an Kieselnadeln rezenter hexactinellider Schwaemme. - Thesis Universitaet Stuttgart, 166 pp., 165 figs., Stuttgart.
Okada, Y. (1928): On the Development of a Hexactinellid Sponge, Farrea sollasi. - Journal of the Faculty of Science Imperial University of Tokyo, Sect. 4 Zoology 2: 1-27, 22 figs., 8 pl., Tokyo.
Rigby, J. K. (1986): Sponges of the Burgess shale (Middle Cambrian), British Columbia. - Palaeontographica Canadiana 2: 1-105, 27 figs., 20 pl., Toronto.
last changes 20.01.2005