P.S.Z.N. I: Marine Ecology, 16 (2): 165--179 (1995)
if': 1995 Blackwell Wissenschafts-Verlag, Berlin
ISSN 0173-9565
Accepted: April 14. 1995
low-Tide Exposure of Sponges in a
Caribbean Mangrove Community
K. ROTZLER
Department of Invertebrate Zoology, National Museum of Natural History,
Smithsonian Institution, Washington DC 20560, USA.
With 5 figures and 2 tables
Key words: Intertidal P{)r~{era, red-mangrove roots, fouling community. temperature-salinity
stress, desiccation.
Abstract. Sponges on subtidal red-mangrovc prop roots may bccome exposed to air many times per
year during very low tides. Full exposure is stressful and potentially fatal, particularly if occurring in
full sun. Largc root sponges show distinct species zonation bctwcen mean low water and - 0.5 m.
Haliclol/a implex({ormis and Lis.I'oc!?'1ldoryx isodiclyali.l' arc ncar the top while Scopa!illo melderi arc
ncar the lower end of thc rangc. Temporary experimental desiccation resulted in 100 % recovery of all
three species after they had been exposed to either stln or shade for up to 2 h. Scopalina is the least
rcsistant and lost over 90 1}'o tissue within 3 days after the 4-h and 6-h experiments; the remaining cell
mass succumbed to infestation by microbes. Haliclona and Lis.I'odelldoryx recovcred from as much as
6 h in full sun but lost 85% and 80% of the original tissue volume, respectively. In Lissotlelltloryx,
clusters of larvae developed in the regcncrating fragmcnts. Watcr loss tolcrated by thc three species is
estimated as 66 % of wet weight in Haliclolla. 54 % in Lissodelltloryx and 38 % in Sco{Jalilla. Salinity of
interstitial seawater (porc water) extracted from exposcd spongcs rose from ambient 3.5 % to 4.3-4.8 %
aftcr I h, to 5.1-5.9% aftcr 6 h. Most endobionts dicd or left their host during this last phase. Natural
vertical zonation in thesc spongcs rcflects their resistance to tidal exposure.
Problem
Mangroves are the predominant tropical shallow-water community worldwide and
have a distribution parallel to that of coral reefs. For several years we have
studied an example of this ecosystem in Belize in the Central American Caribbean
(RDTZLER & FELLER, 1987, 1995). We distinguish between two types: mainland?
fringing mangrove forests with predominately estuarine character and nutrient and
sediment input from land runoff; and island-mangrove swamps that are char?
acterized by near-oceanic salinity and local nutrient cycling and sediment
production. An important attribute that distinguishes Caribbean mangroves from
most others is a tidal range of less than 0.5 m. Island mangroves are the pre?
dominant type in the extensive lagoon behind the barrier reef of Belize and indeed
throughout the West Indies.
In our study area at Twin Cays, just inside the barrier reef of southern Belize
(ROTZLER & MACINTYRE, 1982), red mangrove (Rhizophora mangle) lines the peri?
meter, interior lakes, and tidal canals and is anchored in the mud-covered peat
U. S. Copyright Clearance Ccntcr Code Statcment: 0173-9565/95/1602-0165$11.00/0
166 RUTZLER
bottom by long, arched prop roots. These roots provide most of the solid (non?
sedimentary) substrate in this environment and support a rich sessile-benthos
community that extends well into the subtidal zone. The main components are
algae, sponges, and tunicates, with many associated invertebrates and a few fishes
that find shelter among the fouling community, particularly inside the sponges.
Sponges are generally not considered resistant to environmental extremes such
as exposure to air during low tide because they are unable to move or close their
aquiferous system. have no protective shells or cuticular structures, and generally
have only one reproductive period per year (only few species are reported as
r-strategists). The present study was prompted by the observations that there is a
vertical zonation of sponge species on mangrove roots and that large subtidal
portions of roots with sponges had become exposed to air during a series of
exceptionally low mid-day tides in the wake of a major EI Nino event in 1983.
These low tides, almost 40 cm belo\\" 0 (mean low water) arrived during periods of
full sunlight and left large, massive sponges air-exposed for periods of 1-4 h or
more. The objectives of this study \-vere to determine for selected species the natural
frequency, duration and tolerance of exposure to air, find possible protective
mechanisms against water loss or aiding recovery from desiccation, and clarify
whether sponge zonation on the roots renect the degree of their resistance to tidal
exposure.
Material and Methods
Surveys of mangrove-root species zonation were made by snorkeling through Hidden Creek, Twin
Cays (16"49.95'N. S8'06.34'W: maps in DE WEERDT ('[ 0/., 1991; RUTZLER. 1993) forO.5 h and measuring
the distance from the water line to the center point of each specimen recorded to the nearest centimeter.
Depth data were corrected against a tidal tracing obtained at the same location and related to mean
low water (0 m). Other tidal information was taken from data published by the National Oceanic and
Atmospheric Administration (NOAA, US Department of Commerce), Tidal Analysis Branch, for Key
West (time correction for Belize: + 14 min for high water. +47 min for low water). NOAA tables were
supplemented by time-series measurements (30-min intervals) from the National Museum of Natural
History's coral reef field station on Carrie Bow Cay (ca. 3 km from Twin Cays) using either a capacitive
or an ultrasonic sensor. The Carrie Bow Cay records are complete for 9 months, January to June.
September. November. and December 1993, partial for rest of that year (the tide gauge was inoperative
during 1994). All records were used to calculate mean lower low water which is the datum (0 em) for
the NOAA tables.
Sponges chosen for the experiments belong to different families and orders of the class Demospo/lge?l
(recent descriptions in parentheses): Ha/iclona illlp/cx[fi)/'1llis (HECIITEL), Clla/initlae, Hoploscferitla (DE
\VEERDT ('/ 0/., 1991 :2(2); LissoticlU/oryx isodiclya/i.~ (CARTER), Myxillidae, Pocd/osc!erida (VAN SOliST.
1984:54); and Scopa/i/lll I'IIclzleri (WIEDENMAYER), Axillellidae, Axillellida (WIEDEN~IAYER, 1977: 145, as
U/osa). To deternline resistance to desiccation, specimens of all three species were cut under water into
uniform pieces (ca. 20 x 25 mm ectosomal area, 15 mm thick: S ml volume) and tied to acrylic grid
frames (light-diffuser panels) using 40 Ib (18 kg)-test monofilament fishing line and brass crimps. Each
fragment was kept at least 20 mm apart from the next and. once returned to its habitat, healed and
attached to the new substrate over night. After 6 days (4-10 May 1994) ill siill (but well below low-tide
level) recovery was complete as indicated by new tissue growth. At that point, replicates (4 for each
species and treatment) were exposed to air in sun and shade, for \, 2, 4, and 6 h (10 May 1994) and
immediately resubmergcd and returned to the habitat. The condition of all experimental sponges \vas
monitored daily for 6 days after the treatment. Quantitative estimates of tissue health were made on
days 3 and 6 and compared with controls that were never exposed to air.
Identical but separate treatments were used for histological observations by light and transmission
electron microscopy and for determining water loss rates. Microscopy samples were fixed in 1.5 %
Tidal exposure of mangrove sponges 167
glutaraldehyde bullcrcd in 0.2 M cacodylate with 0.1 M sodium chloride and 0.4 M sucrose (pH 7.2).
For light microscopy, epoxy-rcsin embedded sections were ground. polished, and stained in safranin?
crystal violet according to ROTZLER (1978); semithin (0.5 Jim) TEM sections stained in methylenc blue
were also used. TErvt subsampks wcre post-fixed in 2% osmium tetroxide in the same buller mix.
Rcsults of more detailed electron microscopy will he reported separately (ROTZLER. in press).
An extra set of samples consisting of one small (ca. 30 ml volume) but entire specimen of each of the
experimcntal species was sacrificed to determine water loss over time during exposure to air. It was
suspended in the open shade (air llow. 0-1 m' s '; relative humidity, 68 %) and weighed frcsh (wet
weight to 0.0 I g after quick external blot with tissue paper) and after I h. 3 hand 6 h between 10:00 h
and 16:00 h. Ash-less dry \veight was calculated after weighing (to 0.1 mg) aliquots dried to constant
weight at 70C and ashed in a mullle furnace at 51 () C.
During the period of preparation, experiment. and recovery 12l' April to 16 May, 1994). the tidal
regime at Twin Cays was recorded by a 110at sensor in a still well (4: I gear ralio to recording stylus)
located on a dock in the island's main channel. A self-contained thermistor temperature recorder (24
min time series. O.I'C resolution) \vas installed at Ihe Hidden Creek sponge habitat site just belov,,'
lowest expected tide level for this period. Salinity measurements were taken with a refractometer to the
nearest part per thousand.
Results
1. Physical parameters and zonation
Hidden Creek is typical for tidal channels at Twin Cays. It connects an open bay
(Boston Bay) and a swamp-enclosed lake (Hidden Lake) and allows tidal water
flow between the two. The creek is about 150 m long, 2-4 m wide, and 0.5-2 m
deep. Starting from Boston Bay, it meanders roughly due southwest for two thirds
of its way, then turns north toward the lake. The present study was conducted in
the initial 30 m portion of the channel.
Boston Bay is an expansion of the main channel that separates Twin Cays; it is
in direct connection with the barrier-reef lagoon and has water of similar quantity.
In contrast, Hidden Lake is an isolated, shallow (less than 50 cm) lake with mud
and peat bottom that is subject to extreme fluctuations of temperature and salinity.
During each tidal cycle, the water from one of these two very different environments
flows through the long, narrow channel, reaching at times considerable speed, and
mixes with that of the other. During the sunny experimental period (8-16 May,
1994), temperature cycled with the tides between 28.loC and 35.9 D C, salinity
between 3.3 % and 3.9 %. Previous measurements at the same location peaked at
16'C/39?C and 2.8%;'4.1 (Yo.
The tides in the Carrie Bow Cay area arc microtidal and of the mixed semidiurnal
type (KJERFVE et al., 1982). Records taken during 1993 show a monthly range of
22-48 cm (average 35.4 em). Tidal range at Twin Cays during 29 April to 16 May
1994 was 34 cm (the gauge at Carrie BO\I,/ Bay was inoperative at that time). Tidal
predictions for the past 3 years (1992- I994) were also consul ted to determine
freq uency of ebb-tide exposures to 0 cm (mean lower low water) or below. These
may occur at a peak frequency of9-14 times a month between April to June (Table
1).
The bottom of Hidden Creek is made up of peat covered by mud, decaying
mangrove leaves, bacterial and algal mats, and occasional strands of turtle grass
(Thalassia testudinum). The sides are vertical or undercut, exposing peak banks
behind a phalanx of red mangrove prop roots arching dOV,lIl from the trees that
168
Table I. Frequcncy of ebb tides exposing MLLW (0 m) or lower occurring
during full daylighl hours (10:00 h-16:00 h) in a 3-year period (Key West
predictions correcled for Belize time).
---------~--------
frcqucncy
month 1992 1993 1994 range average
January 2 0 0 0-2 0.7
February 5 7 8 5-8 6.7
March 9 8 4 4--9 7.0
April 10 9 12 9-12 10.3
May 13 12 12 12-?\3 12.3
June 14 14 II 11-14 \3.0
July 14 14 13 13-14 13.7
August 1 6 4 1-6 3.7
September-December 0 0 0 0 0
----_._-
ROTZl.ER
line the channel. Where the mangroves (Rhizophora mangle) are tall and dense the
channel banks are in deep shade. In other places, where the trees are small or there
are gaps in the canopy, the creek may be radiated by full sun.
Vertical or overhanging peat banks and prop roots are the only solid substrates
along these creeks that have moderate sediment exposure and allow sessile filter
feeders, such as sponges and ascidians, to develop species-rich communities. Lim?
iting for vertical distribution (other than substrate availability) are the fine mud
sediments on the bottom and the water level at the top. The calcified green alga
Halimeda opuntia is a strong competitor for space. A census of the most common
and quantitatively important invertebrates fouling the mangrove roots at the
entrance of Hidden Creek revealed 12 species, lOaf them sponges (Fig. I).
Tide tables and graphs allow estimates of frequency of annual low-tide exposure
(in % of tides) for each position, assuming that it is occupied by an average-sized
specimen (or cluster, in case of the non-sponges) of which one half falls dry for
at least I h. Starting at high tide level, the mangrove oyster, Isognomon a/atus
(Isognomidae, not a true oyster), the only obligatory intertidal species of the group,
occupies the range of +24 to -2 cm (0 = mean lower low water, MLLW);
exposure frequency of an animal in the center is 84 % of low tides. Bel0\..' follow
Haliclona imp/ex(formis (Chalinidae, Haplosclerida) starting at + 13 cm (73 %); the
pale anemone Aiptasia pallida (Aiptasiidae) and H. tubifera (Chalinidae, J-lap/o?
sclerida), +8 cm (44?;;?; Lissodendoryx isodictyalis (lHyxiJ/idae, Poeci/osclerida),
Spongia tubulifera (Spollgiidae, Dictyoceratida), and Geodia papyracea (Geodiidae,
Choristida; studied in this location by RUTZLER, 1988), +3 cm (15%); H. cur?
a~'aoensis (Chalinidae, Hap/osclerida), - 2 cm (130/0); Scopalina ruetz/eri (Axi?
nellidae, Axinellida), Tedania ignis (A1yxillidae, Poecilosclerida) , ?Dictyodendrilla
sp. (Dictyodendrillidae, Dendroceratida), Bienl1la caribea (Biemnidae, Poecilo?
sclerida), -12 cm (0.2 0/0).
2. Effects of exposure to air
The effects of different exposures, indicated by the success rate of recovery after
return to the habitat, are summarized in Table 2. In all, three replicates were lost

?Tidal exposure of mangrove sponges
a
173
Fig. 3. Scopalina ruetz/eri, vacuolated and encapsuled archeocytes. a: light micrograph. b: TEM of
detail (arrows point to spongin-like capsules; scales: a = 50 Jim, b = 2 Jim).
archeocytes with single vacuoles. TEM examination of some of the stressed
spermatogenic cysts reveal disintegrating spermatozoa (Fig. 4c, d). Fragments
regenerating from severely damaged specimens were found to contain healthy
early stages of larvae.
Scopalina ruetzleri is a brilliant orange species with conulous surface skin over
extended, large (2-6 mm and more) choanosomal lacunae. Its consistency is very
soft and fleshy. The ectosomal zone is 100 J.lm thick and the densest of the three
species, packed with cells and spongin. The histology is complex, with great diver?
sity of cell types (Fig. 2a; still under study). Chambers are ovoid to baggy irregular
in outline, 32 x 34-50 x 27 J.lm, lined by 21-23 cells per central cross section;
chambers are tightly crowded together. Archeocytes are large (15 x 7-18 x 811m)
but less common than more rounded cells, 16-23 J.lm, that are filled with charac?
teristic angular, highly refractile inclusions of less than I J.lm. Spermatogenic
cysts are about 22-40 J.lm in diameter. Air-exposed specimens show disorganized
choanocytes and enlarged vacuolated archeocytes (Fig. 2b), with either a single
large or a group of small vacuoles. Some cells are encapsuled by a thin layer of
spongin-like material (Fig. 3), others are disintegrating (Fig. 4a, b). Presence of
non-symbiotic bacteria indicate the start of decay processes that lead to the death
of the entire specimen (Fig. 5).

Tidal exposure of mangrove sponges 175
Fig. 5. Scopalifla ruetzleri, TEM of disintegrating inclusion cell with decay-associated bacteria (scale =
1 j1m).
Discussion
The rich fouling community on prop roots of mangroves like Twin Cays expresses
a distinctive intertidal zonation despite a mean tide of only 15 cm. The effect of
water level changes can be enhanced by certain combinations of astronomical
and meteorological conditions and cause substantial physical disturbance. Such
stochastic processes, also including strong tidal currents, waves, and the effects of
predation, may be at least partly responsible for the conspicuous instability of
these communities (BINGHAM & YOUNG, 1995). On the other hand, species adapted
to resist these disturbances contribute toward community structure and predict?
ability. The present observations show that Haliclona implexiformis and Lis?
sodendoryx isodictyalis occupy a higher and more stressful level in the space
hierarchy of the limited mangrove prop-root habitat than Scopalina ruetzleri and
confirm the original hypothesis that both have substantially higher resistance to
desiccation than the latter species.
Reports on truly intertidal sponges are uncommon and mostly restricted to cool
(boreal) climatic zones. The animals are primarily found close to low-tide level and
in damp, sun-protected habitats, such as the lower surfaces of rocks and boulders,
shaded crevices, and areas overgrown by seaweed (DE LAUBENFELS, 1947; BURTO ,
1949; ELVIN, 1976). Specially adapted species may survive direct sun exposure by
growing buried in mud that holds water for hours, as was observed in a Malaysian
mangrove channel (ROTZLER, 1964). In the tropics, sponges subject to drying
during low tide become exposed to four major adverse conditions: loss of oxygen
and food supply provided by circulation of fresh seawater, increase in salinity by
evaporation of interstitial water retained by the animal, increase of exposure to
ambient radiant energy, including temperature and ultraviolet radiation, and,
Tidal exposure of mangrove sponges 169
tide level
[em]
30
25
20
15
10
5
o
-5
-10
-15
-20
-25
-30 ~
o 10
f-'-+--.4
o 10
ebb-tide exposure
[% frequency]
100.0
99.6
99.1
94.3
81.0
55.8
32.8
13.1
2.9
0.6
0.0
Fig. I. Schematic prescntation of ebb-tide frequcncies and vertical distribution (mean ranges) of the
most common large invcrtebrates fouling red-mangrovc prop roots at the Boston Bay entrance of
Hidden Creek, Twin Cays. (Aipl. = A iplasis pallida; Biem. = Biemlla ca,.;beo; Die/)'. = ?DicI.I'odel/drilla
sp.; Geo. = Geodia papyracea: lI. C/I,.. = Ha/iclolla c/lraraoemis: l/. imp. = II. implexijormis; H.
lub. = l/. lubilera; /sog. = /50gIl01ll0f/ a/alus: Lisso. = Lissodc/l(/oryx isodiclyalis: Scop. = Scopalillll
ruetzleri: Spo. = Spof/gia lubulij'era; Ted. = Tedallia i!lf/is).
from their support frame but not more than one per series of four replicates:
controls remained unchanged.
Exposure of I h in shade or sun had only negligible etfects. Development of an
initial, slight bacterial infection (whitish filaments of the multicellular, sulfur-fixing
Beggia/oa) on Scopalina rue/deri disappeared by day 3. Exposure of 2 h was
tolerated \vell, in shade and sun, by both Haliclona implex(lormis and Lissodendoryx
isodictyalis but not by Scopalina. Those replicates of Scopa/ina exposed to the sun
entered a dormant stage, indicated by permanent body contraction with skeleton
fibres left exposed, preservation oflive colour, and lack ofdecay-associated microbe
community. Recovery after extensive tissue reorganization is considered possible
but could not be confirmed during the observation period. Exposure of 4 h resulted
in some tissue loss in Haliclona from the shaded series but only two replicates were
affected. Eventually, both specimens regenerated fully, sloughed the dead cell
and skeleton mass and by recovery day 6 replaced the loss with new growth.
Lissodendoryx showed no signs of stress from air exposure in either shade or
sun. Scopalina on the other hand, suffered extensive cell death as indicated by
discoloration and Beggialoa infection over most orthe body surface of all replicate
specimens. All experimental sponges of this species had become macerated by day
170 RDTZLER
Table 2. Results of desiccation expcriments in shade (L-) and in full sunlight (L+) described in the
text. Quantitative dead-tissuc estimatcs ('~/;, or sponge volume) are averaged for all replicates, unless
otherwise stated. Health status was compared 10 non-exposed controls and judged from colour change
(live n?. dead tissue), turgor. new tissue growlh. and presence or JJc.qgialoa (Begg .? a group of sulfur?
fixing. multicellular bacteria).
recovery time
experiment
L-. I h
l'alie/oJ/a
Lis.\?/J(/t'mloryx
Scopalilla
L-. 2 h
Ilalie/oJ/a
l-issot!cJ/t!oryx
Smpalillll
L-. 4 h
Ila/ic1olla
l.iss/l(/clldoryx
S'('opalilla
I day
A
A
HI
c: 5% dead
A
D: 90 (~/~ dead
3 days
A
A
;\
A
A
A
B'
A
D: 95 (~/;) dead
6 days
A
A
;\
A
A
A
A
E~
comments
11J?'.Cf!/. at base of explants
I I replicate lost from support frame
Idead tissue in 2 replicates only
~regenerating.dead tissue sloughed
'fully recovered
~macerated except for spongin filters;
disease community with feeding ciliates
L-. I h
Ilalie/olla c: 35% dead C: 42 (~/;, dead
Lis.wJdclldor.'?x C: 46 '~.~, dead C: 10'% dead
S('opalilla C: 4~% dead D: 91 (~'~, dead
L+. I h
Halie/oJ/a A AI
Li.uo(/elulor.\'x A A
Sl'opalilla A A
L+.2 h
Ilalie/oJ/a Al A
LissodcJ/t/or.'?x A A
Scopalilla BI B
L-I-.4 h
Ila/ie/ollo AI A
LI:"sOtlelldoryx A A
Scopalilla C" D!: 91 % dead
L ....... 6 h
Ilaliclolla C: 78% dead C": 85 '~'~, dead
Liss/}(/elldoryx C: 79 (~/;) dead C I: 80'~/;, dead
Scopa/illa E E
------
A
A
A
A
A
B
A
A
E
A. E'
A. E1
E
I fully regenerated, dead tissue sloughed
'replicates filled with mature larvae
II replicate lost
I/Jeqg. at base of explants
II replicate lost
!/Jegq. infested
)HeM? infested
"live cells in 2 replicates
'I replicate fully recovered. 3 dead
(A = excellent: full turgor. new growth; B = good: somewhat contracted but no signs of decay, no new
growth: C = passable: stressed. contacted. signs of infection. tissue loss < 50% or volume; D = bad:
infection and tissue loss> 50%: E = dead.)
6 and only the skeleton and a predatory microbial and ciliate community remained.
Exposure of 6 h in the shade caused immediate and substantial (almost half) tissue
loss in all three species. The condition in Haliclona and Scopalina deteriorated
further over the next 3 days but while all replicates of Scopalina had died by day 6,
Haliclona recovered and replaced all dead material with new growth. Lissodendoryx
from the same experiment recovered steadily and fully and by day 6 showed new
Tidal exposure of mangrove sponges 171
growth in all replicates and developing and mature larvae in three of the specimens
(only early stages of embryos were noted in the controls and were barely noticeable
without usc of a microscope). Exposure of 6 h in the sun resulted in the immediate
death of Scopalina and in the demise of t\\'o replicates in Haliclona, three in
Lissodendoryx. Surprisingly, parts of the remaining specimens (amounting to 20?1t,
of the original sponge volume) survived and, ultimately (day 6) one replicate of
each of the two species regenerated into a healthy specimen.
Weight determinations of water loss during suspension in air show that Lis?
sodendoryx holds the most water (37.1 g' g-I ash-less dry weight), followed by
Haliclona (31. 7 g' g-I AD\V), and Scopalina (16.0 g' g-I ADW). Lissodendoryx
benefits from the slowest water loss (by drip and evaporation combined): 5.2 1%
after I h, 21 % after 3 h, 54?;;, after 6 h. Haliclona loses at the rate of: 6.9% (I h),
26?1i, (3 h), 66 IX> (6 h); ScojJalina: 8.9 % (I h), 38 % (3 h), 82 1Yo (6 h). This simulates
the water retention of the three species at their estimated exposure tolerance
thresholds during the recovery experiments, that is about 3 h for Scopalina and 6
h for the other two species. Salinity of interstitial water retained by these sponges
was measured in different replicates after I hand 6 h (3 h not taken). It rose to
4.3 % (1 h) and 5.4 % (6 h) in Lissodendoryx, to 4.5 % (I h) and 5.9 % (6 h) in
Haliclona, and to 4.8 % (1 h) and 5.1 %, (6 h) in Scopalina. These data indicate that
loss by drip (suggested by the I-h value) is lowest in Lissodendoryx, highest in
Scopalina, loss by evaporation (6-h value) is lowest in Scopalina, highest in
Haliclona. Because gravity drives water to the lower portions of the air-exposed
sponge, associated species of endofauna such as polychaetes, crustaceans, and
ophiuroids first migrate there, later leave the sponge or die.
3. Histological expression of stress
The histology of specimens flxed immediately after 6 h exposure to air and sun was
examined and compared with untreated controls (Figs 2-5). Processes of possible
cell repair or replacement over time could not be observed by this method.
Haliclona impex(lormis is a pinkish violet, soft but clastic-compressible cushion?
shaped sponge. The surface is punctate, the interior has a dense, bread-like porosity.
Most aquiferous canals are under 0.5 mm in diameter; canals near the oscula may
reach 2 mm in diameter. The ectosome is about 500 Jim thick, primarily a loose
reticulation of spicules supporting strands of cellular tissue. Choanocyte chambers
are more or less spherical, 25 x 27 Jim in diameter, with 12-15 choanocytes per
central cross section. Numerous large archeocytes (7-8 tLln) and spermatogenic
cysts (16-20 tlln) in different developmental stages (some with spermatids and
spermatozoa) are the most conspicuous elements in the healthy choanosomc.
In air-exposed samples, most choanocyte chambers are disorganized, with many
choanocytes dispersed at random throughout the mesohylc. Some archeocytes
appear inflated (9-13 tim) and enclose large (3-4 jim) vacuoles.
Lissodendoryx isodictyalis is pale grayish green (experimental specimen) to blue,
firm and tough in consistency. The surf~lce is skin-like smooth, the interior coarsely
porous from aquiferous canals measuring 0.5-2.0 mm (average, 0.8 mm) in diam?
eter; 5 mm or larger canals are near the oscula. The ectosomal zone is dense
with cells and reinforced by tangentially placed spicules; it measures 60-100 fll11.
176 RDTZLER
evcntually, loss of cellular watcr upon draining and drying of the internal cavity
and aquiferous system. These conditions may be enhanced, accelerated, or com?
plicated by unusually long-lasting low tides and adverse weather conditions during
exposure, such as unobstructed mid-day sun, strong wind, and heavy rain storms.
Sponges circulate water by means of choanocyte-lined chambers that connect a
complex inhalant and exhalant canal system. When undisturbed, they pump
water- steadily or with cyclic activity - at a great rate. In situ measurements of
three common reef sponges in Jamaica showed water transport rates ranging from
0.2 1to almost 1 1per cm3 sponge per h and a transport efficiency of 20-23 1water
per ml oxygen consumed (REISWIG, 1974). Temporary shut-down of water flow
(for hours or a few days) occurs in the field (REISWIG, 1971) and under unfavourable
aquarium conditions (RUTZLER, unpublished), but this state lacks other stress
factors associated with exposure to air.
There are no obvious protective features in intertidal sponge species, such as
shells that reduce evaporation. Unlike some other invertebrates studied in the
same mangrove swamp (FERRARIS el al., (994) they are unable to move from
unfavourable habitats and lack organs that control ionic gradients between
environment and cell. The sponge body between the interior aquiferous system
and the exterior surrounding seawater is enveloped by pinacoderm and cho?
anocyte cell layers and is composed of mesohyle (an intercellular matrix) in
which cells and skeleton (spongin, spicules) are embedded. Because sponges
tolerate a considerable range of salinity but lack proper body fluids, it has long
been suggested that they use an intracellular mechanism of osmoregulation
comparable to protozoans (HART~lAN, 1985). Indeed, contractile vacuoles have
been demonstrated in pinacocytes (and in several other cell types) of the
freshwater family Spongillidae. The vacuoles surround the nucleus and are
connected to a system of nephridial tubules and lacunae radiating through the
cytoplasm (WEISSENFELS, 1974; BRAUER, 1975). To date, no comparable structures
are known in marine sponges. The vacuolization of some archeocytes observed
in the present study (Figs 2b, 3) are interpreted as stress related, not an
indication of osmoregulation. Based on ionic comparisons of marine sponges
and habitat water (PROSSER, 1967), the necessity of contractile vacuoles (at least
under normal conditions) has been questioned because osmotic water loss rather
than gain is to be expected (SIMPSON, 1984). However, several ecologic and
experimental studies show that marine sponges readily invade brackish-water
environments, particularly the euryhaline c1ionids ('oyster pests') that can live
and function normally in 1.5-2.0 % salinity and can recovered from several days
of exposure to salinities as low as 1.0 I~O (HOPKINS, 1956; HARTMAN, 1958).
Unusually harsh conditions for sponges have been described from brackish
Chilka Lake (India, connected to the Sea of Bengal) where spongillids coexist
with several species of marine demosponges (AJ'.."NANDALE, 1915). Although long?
term measurements of habitat conditions are not known, the spongillids were
found in salinities of 0-1.0 ?lrl, the marine species under conditions ranging from
o (or 'quite fresh') to 2.00/0, some even to full salinity (3.6%).
Few studies are available describing the response of marine sponges to hyp?
ersaline environments. In the laboratory, dissociated cells of Microciona pro/ijera
re-aggregated at salinities ranging from 1.2 to 4.3 % (GALTSOFF, 1925), similar
experiments with lotrochota birofu/afa resulted in re-aggregates between 2.4 and
Tidal cxposurc of mangrovc spongcs 177
3.8 % (DE LAUBENFELS, 1932). These observations, however, do not clarify regu?
latory mechanisms and are not good indicators of salinity resistance in whole
sponges.
It appears that for a sponge the most important initial defense during air exposure
is to prevent or resist water loss. Scopalina starts out \vith a water content that is
lower than that of the other species, which is unimportant as long as the water is
retained. However, Scopalina, also initially (1-3 h), loses a greater proportion of
its water than the other two sponges and by 6 h has lost a dramatic 82 %, probably
by run-otT from the large internal lacunae. The effcct of this loss is reflected by
tissue deterioration after 2--4 h in sun or shade (Table 2). Subsequent to initial
watcr loss by drainage, watcr is lost by evaporation. Thus, sponge cells are bathed
in hypersaline water, a condition to which they may be able to adapt. Typically,
cells of invertebrates exposed to hypersalinity initially accumulate inorganic ions
intracellularly but water follows to restore cell volume. This has the effect that
most intracellular componcnts are restorcd to near thcir original concentrations
while inorganic ion concentrations remain high. Subsequently, if the stress persists,
long-term adaptation involves rcplacemcnt of elevated intracellular inorganic ions.
This is necessary because elevated cellular sodium (Na), potassium (K), and chlor?
ine (CI) will eventually disrupt macromolecular function. Inorganic ions are typi?
cally replaced with non-perturbing osmolytes, such as amino acids,
trimethylamines, and polyols (FERRARIS, 1993, and rcference therein). Whether
sponge cells utilize these mechanisms is as yet unknown. During earlier experiments
with Lissodendoryx isodictyalis and Tedania ignis at T\vin Cays (FERRARIS & RUTZ?
LER, unpublished, 1987) we found that these sponges do not undergo predicted cell
shrinkage when exposed to dcsiccation. This may indicate thc existence of a limiting
phase of ccll-volume regulation via accumulation of inorganic ions, as outlined
above. That sponge cells process volume-regulatory ability as well may be indicated
by the response of the species in the current study. Interstitial salinity clearly
increases in all three species and two of them, Haliclona and Lissodendoryx, recover
completely implying some mechanism ofadaptation. It will be of interest to pursuc
a comparison of the cell-volume regulatory abilities of these sponges.
It is also important to distinguish between tolcrance of short-term extremes and
capability to adapt to long-lasting or repeated conditions that cause physiological
stress. Periods of adverse conditions can be bridged by closing pores and ceasing
pumping activity (HARTMA:'-I, 1958; REISWIG, 1971) and by the formation of gem?
mules which are best known from spongillids but are also common in some
marine species (SIMPSOl'\ & FELL, 1974). It has been shown, however, that although
gemmules protect cells from salinity extremes - indeed their formation and ger?
mination are influenced by osmotic pressure changes - they do not provide
increased resistance to desiccation (FELL, 1975).
Even less well understood than osmoregulation are the effects of solar radiation
on sponges, particularly if the cooling and spectral-filter action of the water is
lacking. Tissue temperature in tide- and sun-exposed sponges can rise rapidly
(13 cC . h -I in tempered climate; ELVII'\ 1976) but this effect may be offset by cooling
from evaporation. More severe effects can be expected from ultraviolet radiation
that is known to damage and kill even submerged sponges in 1-2 days (JOKIEL,
1980).
Extended full exposure eventually leads to drainage of interstitial seawater from
178 ROTZLER
large lacunae which in certain species are presumed to be adaptations allowing
conservation of a water reserve (AN:--lANDALE, 1918). Among the species used in
the present experiments, however, Scopalina ruetzleri has the largest internallacu?
nae which were shown to drain faster than the smaller and more complex interstitial
space system of the other two species where friction and other hydraulic forces
help to retain pore water. Evaporation, on the other hand, is controlled by the
presence of an external barrier, such as a dense ectosome, which is well developed
in S. ruetzleri and Lissodendoryx isodict.valis but missing in Haliclona implexijormis.
It may well be the combination of anatomical advantages with physiological
disposition that makes L. isodict.valis the most resistant to low-tide exposure in the
group.
Summary
Twin Cays, Belize, as most mangrove islands in the Caribbean, have a small tidal
amplitude. I-Iidden Creek, one of the deeper drainage channels, is subjected to a
mean tidal range of 35 cm and supports a rich fouling community that covers peat
banks and prop roots of red mangrove (Rhizophora mangle) below the high-water
line. Tidal currents bring about extreme and abrupt changes in temperature and
salinity and astronomical and meteorological factors may cause as many as 14 ebb
tides below mean lower low water per month. Sessile organisms occur in vertical
zonation that reflects their degree of adaptation to intertidal exposure. In this
habitat, sponges that are typically subtidal in distribution can also be found in the
lower intertidal where they are subject to air exposure and desiccation.
Field experiments involving controlled exposures of three sponge species to air
in shade and sun show that Haliclona implexiformis and Lissodendoryx isodictyalis
recover from a treatment of 4 h out of water and in full sunlight. In these species,
portions of specimens may survive 6 h in full sun and regenerate into healthy
sponges after a few days. Morphological adaptations help to slow water loss by
drip and evaporation. Scopalina ruetzleri survives only 2 h of exposure to air and
has poor morphological water retention quality.
Histological and cytological examination of air-exposed specimens shows stress
primarily in the form of disorganized choanocyte chambers, enlarged vacuolated
archeocytes, and disintegrating spennatozoa. Evidence, such as the absence of cell
shrinkage, suggests that a cellular volume-regulatory mechanism exists in these
sponges.
Acknowledgements
The author is grateful to JOAN D. FERRARIS, National Institutes of Health, Bethesda, Maryland, for
valuable contributions to the discussion. She and Hm?JRY M. REISWIG, McGill University, Montreal,
Quebec, provided expert reviews of the manuscript. Thanks also to KATHLEEt' SMITH for laboratory
assistance, particularly the preparation of Fig. l. MOLLY RYAN and MIKE CARPENTER aided in field
work and with the preparation of illustrations. ANDREA BL.AKE sectioned and took electron micrographs;
HILARY SMITH helped with photographic printing. GEORGE HAGERMA:-lt', SeaSun Power Systems,
Alexandria, Virginia and JIM HUBBARD, NOAA's Tidal Analysis Branch, Silver Spring, Maryland,
provided tidal data. This is contribution number 443, Caribbean Coral Reef Ecosystems Program
(CCRE), National Museum of Natural History, Washington, D.C.
Tidal exposure of mangrove sponges
References
179
AN:-JA:-lDALE, N .. 1915: Fauna of the Chilka Lake, Sponges. Mem. Ind. Mus. Calcutta, 5: 21~54.
--, 1918: Zoological results ofa tour in the Far East. Mcm. Asiat. Soc. Bengal, 6: 193?216.
BI:-lGHAM, B. L. & C. M. YOU~G, 1995: Stochastic events and dynamics of a mangrove root epifaunal
community. P.S.Z.~. I: Marine Ecology, 16: 145-163.
BRACER, E. B., 1975: Osmoregulation in the fresh water sponge, Spol/gil/a laeus/ds. J. Exp. Zool., 192:
181-192.
BURTOK M., 1949: Notes on the ecology of sponges. Br. Sci. News, 2: 83-85.
ELVIN. D. W., 1976: Seasonal growth and reproduction of an intertidal sponge, Haliclona permollis
(BowERHANK). BioI. Bull., 151: 108-125.
FELL, P., 1975: Salinity tolcrance and desiccation resistance of the gemmules of the brackish-water
spongc, Haliclona loosanojJi. J. Exp. Zool., 194: 409-412.
FERRARIS, J. D., 1993: Molecular approaches to the study ofevolution and phylogeny of the Nemertina.
In: R. Gmso:-J, J. MOORE & P. SUNDBERG (Eds.), Advances in Nemertean Biology. Hydrobiologia,
266: 255-265.
K. FACCHALD & B. KENSLEY. 1994: Physiological responses to fluctuation in temperature or
salinity in invertebrates. Adaptations of Alplleus viridari (Decapoda, Crustacea), Terebellides parl'o
(Polychacta) and Goljingia cylindrata (Sipunculida) to the mangrove habitat. Mar. BioI., 120: 397?
406.
GALTSOFF, P. S., 1925: Regeneration after dissociation (an experimental study on sponges). 1. Behaviour
of dissociated cells of lHierociona prol{rera under normal and altered conditions. J. Exp. Zool., 42:
183-221.
HARTMAN, W. D., 1958: Natural history of the marine spongcs of southcrn New England. Bull. Pcabody
M us. Nat. Hist., 12: 1-155.
HOPKI:-lS, S. H., 1956: Notes on the boring sponges in Gulf Coast estuaries and their relation to salinity.
Bull. Mar. Sci. Gulf Caribb., 6: 44-58.
JOKIEL, P. L., 1980: Solar ultraviolet radiation and coral reefepifauna. Science, 207: 1069-1071.
KJERfVE, B. K., RUTZLER & G. H. KmRsPE, 1982: Tides at Carrie Bow Cay, Belize. In: K. RDTZLER &
I. G. MAClt'TYRE (Eds.), The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Belize, I,
Structure and Communities. Smithsonian Institution Press, Washington, D.C.: 47-51.
LAuBE~FELs, M. W. DE, 1932: Physiology and morphology of Porifera exemplified by IOfrochola
birotulata HIGGIN. Publ. Carnegie Inst. Washington, 435: 37-66.
--,1947: Ecology of the Sponges ofa brackish water environment, at Beaufort, N.C. Ecol. Monogr.,
17: 31-46.
PROSSER, C. L., 1967: Ionic analyses and c1Tects of ions on contractions of sponge tissue. Z. vgl. Physiol.,
6: 69-74.
REISWIG, H. M., 1971: In situ pumping activity of tropical Demospongiae. Mar. BioI., 9: 38-50.
--, 1974: Watcr transport, respiration and energetics of three tropical marine sponges. J. Exp. Mar.
BioI. Ecot., 14: 231-249.
RUTZLER, K., 1964: Aspects oflittorallife in the Indian Ocean. Atti Scm. Stud. BioI. Bari, I: 1-17.
--, 1978: Sponges in coral reefs. In: D. R. STODDART & R. E. JOHANNES (Eds.), Coral Reefs: Research
Methods. Monographs on Oceanographic Methodology. 5. Unesco, Paris: 299-313.
--, 1988: Mangrove sponge disease induced by cyanobacterial symbionts: failure of a primitive immune
system? Dis. Aquat. Org., 5: 143-149.
--,1993: Caribbean Coral Reef Ecosystems program, progress report 1991-1993. National Museum
of Natural History, Smithsonian Institution, Washington, D.C.; 20 pp.
-- -, C. FELLER, 1987: Mangrove swamp communities. Oceanus (Woods Hole Oceanographic Insti?
tution), 30(4): 16-24.
- -, 1995: Caribbean mangrove swamps. Scientific American (in press).
-- & I. G. MACINTYRE, 1982: The habitat distribution and community structure of the barrier reef
complex at Carrie Bow Cay, Belize. Smithson. Contrib. Mar. Sci., 12: 9-45.
SIMPSON, T. L., 1984: The biology of sponges. Springer-Verlag, New York, Berlin, Heidelberg, Tokyo;
662 pp.
-- & P. E. FELL. 1974; Dormancy among the Porifera: Gemmule fonnation and germination in fresh?
water and marine species. Trans. Am. Microsc. Soc., 93: 544-577.
SOEST, R. W. M. VA).], 1984: Marine sponges from Cura~ao and other Caribbean localities. Part III.
Poecilosclerida. Studies on the Fauna of Cunl~ao and other Caribbean Islands, 66(112): 167 pp.
WEERDT, W. H., K. ROTZLER & K. P. S~UTH, 1991: The Chalinidae (Porifera) of Twin Cays, Belize,
and adjacent waters. Proc. BioI. Soc. WasIl., 104: 189-205.
WEISSE:"FEl.S, N., 1974: Bau und Funktion des SiiHwasserschwamms l:-jJhydatiafluviatilis L. (Porifera).
I. Das Nephridialsystem del' Pinacocyten. Cytobiologie, 8: 269-288.
WIEDENMAYER, F., 1977: Shallow-water sponges of the western Bahamas. Exp. Suppl. 28, Birkbauser
Verlag, Basel & Stuttgart; 278 pp.
P.S.Z.N. I: Marine Ecology. 16 (2): 180 (1995)
:(: 1995 Blackwell Wissenschans-Vcrlal.!. Berlin
ISSN 0 173-9565 ~
Book Review
HAYWARD, P.l., l.S. RYLAND & P.O. TAYLOR (Eds.): Biology and Palaeobiology of
Bryozoans. Proceedings of the 9th International Bryozoology Conference, School
of Biological Sciences, University of Wales, Swansea, 1992. Olsen & Olsen
(Fredensborg), 1994. 240 pp., ISBN 87-85215-23-6
This volume is curre11lly the last one in a long series published by the 'International Bryozoology
Association', \",hich organizes conferences of high scientilk standard at triennial inten'als: the first was
held in Milan in 1968, the last one in Wellington. New Zcaland in 1995. From the very beginning
these Bryozoology Conferences had been intcnded to olrer an opportunity for both zoologists and
palaeontologists to present results of their studies on Bryozoa. Bryozoology in the broadest sense of
the word has always been the goal of thesc mcetings and, therefore, the wide range of topics involved
include: taxomony, biogeography, ecology, biochemistry, pollution problems, etc.
Following this fundamental framework, the 45 publications (out of 58 papers presented at the
conference) included in this volume concern an impressive variety of studies within bryozoology: larval
release and settlement, larval morphology, astogenetic and ontogenetic development. distribution
patterns, growth and survival strategies, food sources, feeding elliciency and rates, a bryozoan-sponge
symbiosis, taxonomy, zoogeography. freshwater bryozoans, etc. With respect to palaeontology, papers
concern bioimmuration, taphonomy, budding patterns. evolution, palaeoecology. and palaeo?
biogeography. including faunas from the Ordovician (Ireland. Australia) through the Devonian. Car?
boniferous. Permian. Triassic. and Cretaceous up to the Tertiary (Austria, Egypt. Spain). Using this
approach, this volume offers an excellent survey of progress and activities in the field of bryozoan
research. Author, keyword and systematic indexes arc valuable and useful extras.
A referee system was applied. guaranteeing a high scientilic standard; excellent quality of printing
and illustrations and a rather moderate price (ca. US S120) arc an additional recommendation for this
book. The purchase of this volume is therefore an utmost necessity for all those biological and
palaeontological institutes. museums. etc. concerned with the study of ecology and biodiversity in
invertebrates in general.
NORBERT VAVRA. Vienna/Austria
U. S. Copyright Clearance Center Code Statement: 0173-9565/95/1602-0180$11.00/0