FLOATING
CORALS:
A POSSIBLE
DISTRIBUTION
SOURCE
DATA
OF ERRONEOUS
Louis S. Kornicker
A. & M. Collcgc
of Texas
Donald F. Squires
Smithsonian
Institution
ABSTRACT
Some reef corals arc able to float after their cellular structure is filled with air. Expcrimcnts pcrformcd
with corals collcctcd
on Texas’ barrier islands show that when corals
capable of floating arc immersed in water they absorb water slowly and stay afloat for
of floating
varying lengths of time; some float for more than 8 months. The possibility
corals being deposited far from whcrc they originate is real and might lead to erroneous
distribution
data and wrong ecological interpretations.
(Dana), and Solenastrea bournoni ( MilneEdwards and Haimc ) are not uncommon
The ability of corals to float after drying,
on the beaches and among the sand dunes
although not of gcncral knowledge, was of Padre, Mustang, and St. Joseph islands,
recognized as long ago as 1775 by Muller,
Texas.
who gave the specific name natans to ColAll species named above are commonly
pophyllia natans, the “Schwimmenstcin,”
in found living among coral reefs of the West
recognition of its buoy,ancy. Muller noted
Colpophyllia
natans
Indies and Florida.
that to float, the corallum must first be has also been collected alive off the top of
dried so that its inner cells become air filled.
a bank called the “Flower Gardens,” which
Other references to flotation of C. n&ans in- is situated in the Gulf of Mexico about
clude those of Esper ( 1789)) Dana ( 1848),
120 milts south of Galveston, Texas (T.
and Matthai ( 1928). Flotation of the IndoPulley, personal communication).
SolennsPacific species Favia speciosa, which has a trea bournoni
has been reported from
characteristically
light corallum, was noted
Alacran Reef, which is situated in the Gulf
by Guppy ( 1889), Wood-Jones ( 1912)) and of Mexico about 70 milts north of Progreso,
Vaughan ( 1918) who suggested long trans- Yucatan (Kornicker, et al. 1959).
portation of floating coral heads in some
About 35 dry coral specimens from Padre,
instances. Weigelt (1938) called upon floMustang, and St. Joseph islands, ranging in
tation to account for the peculiar occur- weight from 0.52 to 43.6 kg, were tested
rence of a fossil specimen of Colpophyllia
and found to float, and it is postulated that
in the Oligocene of central Europe.
flotation is the principal
mechanism by
The present paper identifies coral species which these corals were transported to the
capable of floating after the cells become coastal islands of Texas. It is possible, of
air filled, considers the internal structure of course, that some specimens were torn off
Floating corals, and presents data on the outcrops of fossils in the offshore waters of
distribution
of floating corals which indithe, Gulf of Mexico. However, these species
cate that the corals were transported con- have not been reported from the near-short
siderable distances from where they lived,
Gulf, and it is unlikely that if this were the
source, the corals on the islands would conTRANSPORTATION
RECORDS
sist mostly of specimens that are capable
Dry corals belonging to the species CoZ- of floating when dry. Storms satisfactorily
account for the presence on the beach of
pophyllia
natans ( Muller) , Colpophyllia
amaranthus ( Muller ) , Solenastrea hyades the small coral Astrangia astreiformis Milne447
INTRODUCTION
448
LOUIS
S. KORNICKER
AND
Edwards and Haime, which lives offshore
and does not float when dry, It is considcrcd unlikely that corals in 11 or more
fathoms of water on offshore banks, which
are situated in water 45-65 fathoms deep,
could be transported to the beach either by
flotation or by bottom currents.
It is possible that some of the floating
corals now found on the coastal islands of
Texas became dry during a low stand of the
sea and then floated to the Texas coast;
however, several specimens collected from
the water’s edge of Padre Island contained
dried polyp tissue showing that these corals
were of recent origin. Many specimens of
coral collected along the water’s edge had
attached to them goose-neck barnacles with
stalks that had not yet decayed, indicating
that they must have been stranded just prior
to discovery.
Additional evidence, supporting the hypothesis that flotation
is the principal
mechanism in the distribution
of these
exotic corals, was supplied by Mr. John W.
Ward, Corpus Christi, Texas, who observed
a specimen of Colpophyllia
floating off
Padre Island in 1954 (personal communication ) . This specimen weighs 34.5 kg, and is
a fragment of a coral estimated to have
weighed in the vicinity of 900 kg. Also, a
ship’s captain reported seeing a large coral
head floating in the Gulf of Mexico off Port
Aransas, Texas, on 27 January 1960 (H. T.
Odum, personal communication).
The largest specimen of floating coral observed by the authors was a complete coral
head of C. natans, which was collected by
Mr. Louis Rewalt on Padre Island.
It
weighed 43.6 kg.
Corals cast upon a beach, dried, and then
taken again into the sea could be transported from the Caribbean area to the Texas
coast through the mechanism of the Gulf of
Water entering the
Mexico circulation.
Gulf of Mexico converges in the coastal
bend area of Texas where these islands are
located. The dominant .wind pattern from
the southeast would also facilitate drift towards these islands.
Because the distance traveled by a float-
DONALD
F. SQUIRES
TABLE
Dry
weight
(!=T)
Spccics
~
1. Field tests of floating
_.-
-- .-
Maximum
dimcnsion
(cm)
Initial
duration
specific
( ap~~Zr?ate
)
::%f
remained
afloat
( days )
.-
Colpophyllicr.
nntnns
366
442
477
1,950
2,005
15
17
31
23
0.81"
0.64"
-
0.73" 17
0.65b 58
0.75" 240 plus
0.73b 24Oplus
51
407
420
9
16
16
0.69" 0.64' 11
0.61" 0.55b 124
0.86" 0.83"
5
Solenust~ea
houmoni
Solenmtrea
? hournoni
103 -
0.61"
-
124
Peclinia
paeonia”
8-
12 -
-
0.75"
0.44b 240 plus
50
n Approximate
specific
gravities
determined
by dividing
dry weight
of specimens
by wet weight
at time specimens
sank.
b Approximate
specific
gravity
obtained
by dividing
dry
weight
of specimen
by volume
of water
it displaced.
Comparison
of this method
with
that described
in footnote
“a”
above,
indicates
that the specific
gravities
obtained
in this
method
are about
0.055
points
lower.
C These Pacific
specimens
were purchased
from
a supply
house.
balance
of specimens
was
collected
on
Padre
Island,
Texas.
ing corallum depends not only upon its
initial buoyancy, but also upon the duration
of the buoyant state, 9 specimens from
Padre Island were placed in a tank of water
and observed for 8 months. The results of
this experiment are shown in Table 1.
Certainly the observed flotation period of
8 months for 4 of the specimens tested is
permissive of transport across large bodies
of water, and of introduction of coralla into
habitats to which they are foreign, Larger
specimens are quite capable of floating sufficiently long for completion of a journey
from the Caribbean to Texas coastal islands,
while those which sink below the surface
may possibly be transported for considcrable distances before they actually touch
bottom. Thcrc is also the possibility
of
beach-to-beach transportation with periodic
drying on the beach alternating with periods
of transport in water currents.
DISTRIBUTION
COMPOSITION
AND
INTERNAL
OF FLOATING
OF FLOATING
STRUCTURE
CORALS
From the evidence acquired during the
course of this study, it is apparent that the
factors causing lowered specific gravity in
coral skeletons are variable within spccics.
For example, field flotation experiments
h ave demonstrated that specimens belonging to the same species float for differing
periods of time (Table 1). Indeed, some
specimens may float while others of the
same species do not. As an example, only
one of about a dozen specimens of Diploria
str@osn (Dana) collected alive and in situ
by Kornicker from the Alacran Reef, Campeche Bank, Mexico, floated after being
dried. However, apparently among some
species such as Colpophyllia
natans, C.
amaranthus, and Fnvia speciosa, specific
gravities are usually well below that of
water and flotation is the general rule.
Primary inspection of coralla from beach
drift might suggest that large cavities caused
by burrowing clams or snails might cause
entrapment of air and enhance the buoyancy of the specimens. Such holes arc not
necessary nor are those caused by other
chemical, physical, or biological means, for
some floating specimens are apparently free
from all such cavities. It is also known that
the period of time elapsed between the
tearing loose of the coral from its substrate
by external forces and subsequent floating
may be small; although the majority of
specimens arc bleached white and fragmentary, indicating a period of wear and tear
on drying on bcachcs, some specimens have
been observed floating with the dried tissues
of the polyp still remaining on the corallum.
It is probable, however, that the majority
of floating specimens have spent some time
in residence on a beach near where they
lived, for the mechanism required to break
the corallum loose would probably be storm
‘waves and these are quite likely to cast the
specimen upon the beach, Later, after the
pore spaces of the corals have become air
filled, high storm waters may remove the
corals from the beach and start them on
1:heir journey.
Because of the possibility that residence
CORALS
449
on the shore with exposure to air, fresh
water, and sunlight might result in conversion of the original aragonite to calcite, resulting in a lower specific gravity (2.94 to
2.72), specimens were checked for mineraX-ray spectrographic
logical composition.
analysis made by Dr. Brian Mason, American Museum of Natural History, showed no
differences in skeletal mineralogy between
floating and nonfloating types.
The mechanism of flotation is obviously
the result of reduction of specific gravity by
the introduction
of large volume of airfilled spaces. A coral skeleton composed of
vertical elements, the walls and septa of the
corallitcs, and horizontal elements such as
dissepiments, aided by vesicular exothccal
tissue, is remarkably well designed for such
a feat. Specific gravities of 9.44 were obtained for specimens of Pectinia paeonia indicating that the volume of air entrapped
is considerable in relation to the mass of the
skeletal material. Other specific gravities of
floating specimens are given in Table 1.
Superficial examination of the rclativc
sizes of pore spaces in floating and nonfloating specimens of the same species does
not yield informative data. Differences of
a very small magnitude in the thickness of
the walls of the cellular calcareous tissue
can introduce a considerable difference in
the specific gravity of the entire corallum.
This aspect of the variation within species
must receive continued attention.
As shown in Figure 1, corals immersed in
water initially
absorb water rapidly, but
after about 30 days, water is absorbed more
slowly. The specific gravity of the coralla,
including those which sank soon after immersion, never reached that of aragonite
(2.94). Corals should float for proportionatcly longer periods in water with high
salinity and low temperature because of the
resulting increase in water density. It is
thcrcforc considcrcd possible that corals
which become waterlogged and sink below
the surface while in the open sea might remain suspended in denser subsurface water
and be transported by deep currents.
As a check on field experiments of flotation duration, several specimens were
450
__
LOUIS
S. KORNICKER
AND
I
I
1
I
I
-o--L
I
DONALD
F. SQUIRES
granting the assmnption that all entrapped
air was removed from the specimens, To
ascertain the validity of this assumption,
thin, flat fragments of septa several millimeters square which were free from all
angular protuberances or re-entrants were
floated in methylene-iodide
acetone solutions togcthcr with standard density cubes,
The following specific gravities were obtaincd:
Species
Specific gravity
CoZpophyZZiu cf. C. umurunthus
(Muller ) _..._... . .. . .._._.___
_ _._._.._---___--2.69 2 0.01
SoZenust~eu
hyudes (Dana)
._...-___.._...2.69 2 0.01
Pectin.iu pueoniu ( Dana ) .._. _-.._____..._._..
about 2.8
These data indicate that despite the high
vacuum, entire specimens retain much air,
suggesting that there must be large pore
spaces which are closed or scaled. If this is
so, it also is suggestive that these pore
spaces are filled with gas during the life of
the polyp. This line of speculation needs
8 0.
120
160
0
40
field study and careful laboratory corroboTIME
DAYS
ration,
FIG. 1. Graphs showing the change in specific
gravity of corals immcrscd in water. Open circles
Specific gravities of minute septa1 partirepresent spccimcns of Colpophylliu
natans colcles are also much lower than those exlccted on Padre Island.
Closed circles represent
pectcd for purely aragonitic skeletal elespecimens of Solenastreu
bownoni
collected
on
ments. It is expected that the reduction is,
Padre Island. Circles with bottom half filled rcprein part, a reflection of the amount of subsent an unidentified
spccimcn collected on Padre
Island.
Circles with right half filled
represent
strate materials within the skeletal element,
spccimcns of Pectin&z pueoniu, a Pacific
coral,
upon which the aragonite is deposited.
which was purchased from a supply store for USC
This material, a mucopolysaccharide -like
in this expcrimcnt.
Corals floated when the speciEic
substance (Goreau 1959), is prcscnt in
gravity was below one (shaded arca). The initial
approximate
specific gravity of corals that did not
unknown quantities in these corals. Dissink and of the unidentified
coral that did not float
crepancies between theoretical values of
was determined by dividing the weight of the dry
specific gravity Candthose observed for small
spccimcn by the volume of water it displaced.
The
fragments are due to imperfections in tcchapproximate
specific gravity of corals that at first
floated but then sank was determined by dividing
nique and in particular to buoyancy derived
the weight of the dry spccimcn by the weight of
from the fundamentally
fibrous structure
the specimen after it just sank. Interim specific
(thus,
perhaps,
porous)
of
the coral skelegravities were obtained by dividing the dry weight
ton.
of the specimen by the wet weight at the time
In general, it is apparent that the flotameasurements were made.
tion of the coral skeleton is dependent upon
placed in water-filled jars and subjected to the entrapped gases within the cellular
high vacuum. In some instances, the speci- organization of the skeletal elcmcnts. This
mens remained under the high vacuum for cellular construction, coupled with an app,arent reduction of density resulting from
as long as 72 hr, yet when sp&iEic gravities
compositional aspects, is probably variable
were determined at the end of this period,
to a very large extent not only from region
they were in the range of 1.8 to 1.96;which
to region, but locally with a given populais far below that of aragonite. These low
tion. Whether local environmental
diffcrvalues of specific gravity are impressive
DISTRIBUTION
OF FLOATING
cnces alone affect the type and density of
skeletal material deposited, or whether
there is also some genetic variation within
Rcduccd
is not known.
8 population,
specific gravity is probably related directly
to growth rates, although there are insufficient data on this point to make any
generalization.
ZOOLOGICAL
AND
GEOLOGICAL
SIGNIFICANCE
The possibility of floating corals serving
for the attachment and dispersal of marinc
organisms is obvious. Marine algae and
goose-neck barnacles have been observed
growing on coral specimens recently deposited at the water’s edge, This potential
has already been recognized for floating
coral by Guppy ( 1889) as it has been for
floating pumice by Guppy ( 1889) and
more recently by Richards ( 1958).
Because floating corals need not always
bc worn or fragmentary, caution must be
used in interpreting deposits of coralla. For
example, recent coral specimens were found
up to 11 ft above sea level on the side of
clay dunes near Port Isabel, Texas (Price
and Kornickcr 1961). Apparently the coral
had been swept up the side of the dune by
hurricane winds. In the absence of other
evidence, the presence of large amounts of
coral at that height might have been interpreted as a result of uplift. Correct interpretation in this instance was aided by the
presence of other floating material including wood, asphalt, and pumice. Accumulations of debris such as this can bc used in
the identification
of old hurricane strand
lines (Price and Kornicker 1961). Floating
corals from beach deposits on the Texas
islands are frequently unworn and may be
quite angular. Nonfloating coralla, on the
other hand, are usually quite worn and
rounded. In several instances coralla still
having the dried flesh of the polyps on
them have been observed to float,
Perhaps the most serious problem arising
from the distribution of coralla by flotation
is the possibility of their presence being
misinterpreted
in assessing environments.
Because all of the corals observed to float
are reef-building
(hermatypic)
types, they
CORALS
451
are important indicators of environmental
conditions, Most reef corals live in a very
rcstrictcd climate in regard to tempcraturc,
salinity, and depth-all
characters of importance, particularly
in the interpretation
of past environments.
Inclusion of one or
more spccimcns of floating coralla in a fossil
assemblage enclosed in a sedimentary rock
might cause this deposit to be interpreted
as having been laid down under tropical,
shallow-water conditions. Discrimination of
this circumstance as against a more lcgitimate association of reef corals might be
difficult.
As mentioned, coralla of floating
corals seldom show the physical effects of
their possible long transport, and secondary
effects such as the deposition of minerals
on the corallum during the process of fossilization might make testing of the floating
hypothesis impossible.
Erroneous distribution
records can well
arise by means of floating coralla. Specimens of Favia speciosa have been observed
considerable distances from the places in
which they grew, and similarly, ColpophylZia has been collected far from its true
habitat. It is difficult to know if any existing distribution records reflect the flotation
of coralla, but the possibility
must be
watched for carefully in the future. As
examples of the types of instances in which
the floating corals can become an intcrpretation problem, the following are cited.
In 1960, Dr. Harry K. Wells, Duke University, submitted a number of corals to
Squires for identification.
Among them was
a specimen of Colpophyllia
labclled as “a
fragment of a large mass at Cape I-Iatteras.”
Colpophyllia probably does not occur within
600 miles of Cape IIattcras, the most likely
source of the specimen being the northern
Bahamas. If the fact that the specimen
floated had not been recognized, the occurrcnce of the large corallum could have
been taken as an anomalous record of a
Colpophyllia living in cool water on a sand
bottom-both
conditions foreign to the
normal habitat of the genus.
Recently, trawls taken by Messrs. Stetson
and Pratt, of Woods Hole Oceanographic
Institution, in depths of 400 fathoms on the
452
LOUIS
S. KORNICKER
AND
Blake Plateau off Charlcstown, S. Carolina,
yielded a large specimen of Colpophyllia
over a foot in largest diameter. Associated
with the specimen were numbers of living
deeper water corals constituting the normal
fauna at that depth. The Colpophyllin
barely floated several weeks after collection, but was quite obviously a formerly
waterlogged specimen which had become
lodged with the other materials. Interpretation of the biological assemblage, had it
been collected in a fossil situation, might
have been difficult.
An instance where floating coral may
have lead to an erroneous archeological
record was also encountered during this
study. A human head .mask carved from
Solennstrea sp. had been given to Mr. Louis
Rewalt in 1958 with the information that it
had been picked up among the sand dunes
of Padre Island. The carving floated when
tested in water, and remains of marinc
organisms in the hollowed-out eyes showed
that it had been in the water after being
carved. The face of this crudely fashioned
mask is covered with a plaster-like material,
which effectively covers the coral and forms
a smooth surface except where it has been
worn by erosion. Photographs of the mask
have been sent to many anthropologists, but
the cultural source of the specimen is unknown.
SUMMARY
AND
CONCLUSIONS
Although it is not generally known, some
reef coral skeletons are able to float after
their cellular structure is filled with air.
The possibility
of floating coralla being
misinterpreted
in both modern and fossil
accumulations is real and must be guarded
against. In fossil assemblages, identification of floating coralla may be difficult
because of the difficulty in directly testing
the specimens. Recognition of other species
of modern corals which float is necessary
so that the proportions of the fauna can bc
measured.
DONALD
F. SQUIRES
Flotation mechanism seems to be largely
related to air entrapped in the cellular construction of the corallum. Apparently for
most species a period of drying out in air
is a requisite for flotation.
Evidence accumulated from preliminary laboratory testing and from the discovery of specimens
floating with dried polyps still attached to
the corallum suggests that gas may be present in the corallum during the lift of the
coral. Causes of variation in corallum specific gravity from individual
to individual
as well as between species is not understood.
REFERENCES
Expcd.
DANA, J. D. 1848. U. S. Exploring
Zoophyte.
Nurn&pEn, E. J. C. 1789. Die Pflanzenthiere,
berg.
COREAU, T. F. 1959. The physiology of skeleton
formation in corals.
I. A method for measuring the rate of calcium deposition by corals
Biol. Bull., 166:
under different
conditions.
59-75.
Islands.
GUPPY, H. B. 1889. The Cocos-Keeling
Scottish Geogr. Mag., 5: 281-297.
KORNICKER, L. S., F. BONET, R. CANN, C. M.
Reef,
Campcche
HOSKIN. 1959. Alacran
Bank, Mexico.
Publ. Inst. Mar. Sci. Univ.
Texas, 6: l-22.
MATTIIAI,
G. 1928. Catalogue
of the Madrcporian Corals in the British Museum, Natural
History, 7 : London.
NurnMULLER, I'. L. 1775. Linne Natursystcm,
berg.
PRICE, W. AJXMSTRONC, AND J,OUIS S. KOHNICKEH.
1961. Marine and lagoonal deposits in clay
dunes, Gulf
Coast, Texas.
J, Std. Pet.,
31(2):
245-255.
distribution
BICIIAHDS, A. F. 1958. Transpacific
of floating pumice from Isla San Benedicto,
Mexico.
Deep-Sea Res., 5: 29-35.
VAUGHAN, T. W.
1918. Papers Dept. Marine
Biol. Carnegie Inst., 9.
1943. Revision of
AND J. W. WELLS.
->
the suborders,
families
and gcncra of the
Sclernctinia.
Geol. Sot. Amer., Spcc. Pap.,
44: 298 pp.
6.
WEIGELT, J. 1938. Palcobiologica,
WOOD-JONES, F. 1912. Coral and Atolls, Reeve,
London, 392 pp.