The life cycle of a gorgonian:

Eunicella singularis (Esper, 1794)

by

Steven Weinberg & Francisca Weinberg

Institute of Taxonomic Zoology, University of Amsterdam, The Netherlands

Abstract 1. INTRODUCTION

The life cycle of the gorgonian singularis has been The biology of members of the subclass Octo-

studied with emphasis on larval behaviour, metamorphosis corallia () is still poorly known. Al- found a mobile and annual growth. Planulae are to have the the well though development of gonads, as as phase lasting from several hours to several days. Once settled, have been studied they metamorphose into a complete primary in approxi- some aspects of larval behaviour, four days. In the first buddingwill yield colonies mately year, rather is known further well, very little about the of a height between 10 and 30 mm. Subsequently, average life of these , their or the from 14 33 Death be growth rates, growth rates range to mm year¯¹. may

due several Predators denude the to to causes. may partly causes leading their death.

gorgonian branches, thus facilitating the settlement of epi- In an attempt to draw a picture of the life cycle turn invade the entire bionts, which in may skeleton, slowly of a typical octocoral, we examined the fate of pushing back the living tissue of the gorgonian. Colonies may

off substratum current from the Mediterranean Sea also be torn their by wave or action, gorgonians during

this sometimes when tall epibionts process being speeded up critical Most observations some phases. were car- such as fast growing bryozoans enhance resistance to water ried out on the White Sea Fan, Eunicella singularis movement. Once toppled, the gorgonians die by necrosis of data their living tissues, or by being buried under sediment. (Esper, 1794), although some were also

Colonies of E. singularis are estimated to reach an of age obtained for the Orange Sea Fan, Lophogorgia approximately 25 to 30 years. Some data have been obtained and the Sea and life of Mediterranean ceratophyta (Linnaeus, 1758) Purple on growth rates spans two other gorgonians, Lophogorgia ceratophyta and Paramuricea clavata. Fan, Paramuricea clavata (Risso, 1826). Recent

descriptions of these gorgonians can be found in Résumé Carpine & Grasshoff (1975) and Weinberg

Dans cette étude le cycle de vie de la Eunicella gorgone (1976). singularis est abordé, et plus particulièrement le comportement Wherever our own observations were insuffi- larvaire, la métamorphose et la croissance annuelle. Les larves consulted authors for planula ont une phase mobile qui dure de quelques heures à cient, we previous additional

quelques jours. Après s’être fixées sur un substrat favorable, information. We that studies of feel this type are elles se métamorphosent en un polype primaire complet en for better of necessary a understanding the ecology l’espace de quatre jours environ. Pendant la première année,

colonies atteindront 10 30 de of these animals. les jeunes entre et mm hauteur par bourgeonnement. Ensuite, les vitesses de croissance moyen- 2. MATERIAL AND nes de mm an La survenir METHODS seront 14 à 33 ¯¹. mort peut par plusieurs causes. Certains prédateurs peuvent partiellement observations Field were invariably carried out dénuder les branches des facilitant ainsi l’installa- by gorgones, means of SCUBA in the of tion d’organismes épibiontiques, qui à leur tour, peuvent diving region Banyuls-

envahir le lentement les tissus squelette entier, repoussant sur-Mer (southern France). Ripe female colonies vivants de la Les colonies gorgone. peuvent également être of Eunicella singularis were under arrachées de leur substrat l’action des du recognized par vagues ou water I small courant. Ce phénomène est parfois précipité lorsque de grands (pi. A) by scratching away a portion tels des la of organismes épibiontiques, bryozoaires, augmentent coenenchyme with a finger nail, thus revealing fois résistance à l’eau. Une renversées, les meurent gorgones the bright pink eggs or planulae when nécrose des tissus vivants, ou recouvrement le present. par par par and in sédiment. Les colonies d’E. de These colonies were collected seawater singularis atteignent un âge kept

environ. données 25 à 30 ans Quelques sur les vitesses de aquaria until spawning of the larvae took place. croissance la durée de vie de deux et autres gorgones médi- Larvae were collected by means of a suction flask terranéennes, Lophogorgia ceratophyta et Paramuricea clavata, ont été obtenues également. (fig. 1).

9

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 128 S. WEINBERG & F. WEINBERG - GORGONIAN LIFE CYCLE

observed in with The released the Metamorphosis was aquaria spermatozoids, into seawater,

running seawater, either by direct observation or enter via the oral apertures into the body cavities

by means of time-lapse cinematography. Tagging of polyps of female colonies, where fertilization

of gorgonian branches was done with numbered takes place. Segmentation of the zygote is holo-

tags of label tape, which were attached by means blastic (Kowalewsky, 1873; Von Koch, 1887) and

of thin, plastic coated electrical wire. takes place in the body cavity of the female polyp.

Although this is the mechanism encountered in

most some in Octocorallia, exceptions exist, e.g.

Clavularia crassa (Milne Edwards, 1848) where

the cleavage of the zygote takes place on the out-

side of the polyps (Kowalewsky & Marion, 1882,

1883; d'Hondt & Tixier-Durivault, 1975; Wein-

as is the case for Cornularia berg, 1978), saga- miensis Utinomi, 1955 (see Suzuki, 1971).

4. THE PLANULA

each The cleavage of zygote eventually leads to

the planula larva, a description of which follows.

The planula of E. singularis is bright pink in

colour, due to the vitelline reserve which is its

only food source during the mobile phase. It is

to pear- worm-shaped, being at average 2.5 mm female colonies in a Fig. 1. Ripe gorgonian (g) are kept and long 0.5 mm wide (fig. 2a-d). seawater aquarium (A) until spawning of the larvae (1)

takes place. Suction produces a vacuum (v) in the suction

flask (SF), causing the larvae to be drawn in (arrow) by

directing the nozzle (n) at them.

3. THE GONADS

The development of octocorallian gonads has been

studied by several authors (De Lacaze-Duthiers,

1864; Von Koch, 1887; Kukenthal, 1919; Moser,

1919; Suzuki, 1971; Vighi, 1972).

Octocoral colonies are either male or female,

six the sperm vesicles and eggs developing on the non-siphonoglyphal septae (Kukenthal, 1925).

takes several in This development months, the case of the female gonads of Corallium rubrum (Lin-

Al- naeus, 1758) even two years (Vighi, 1972).

the though we have not attempted to follow development of the gonads of E. singularis throughout a yearly cycle, we have only seen ripe gonads towards the end of spring (May). A detailed study of the gonads of Eunicella cavolinii

(Von Koch, 1887), a closely related , was carried out by Von Koch (1887). Some micro-

of of Eunicella in I B-E. graphs gonads appear pi. Fig. 2. Variable shapes and sizes of planulae of Eunicella The are and eggs polylecithal, spherical bright singularis (a-d). Anterior side of the larvae is left in the in colour. Certain stimuli pink figures. cause the larvae to contract (e).

Downloaded from Brill.com10/08/2021 12:28:07AM via free access BIJDRAGEN TOT DE DIERKUNDE, 48 (2) - 1979 129

twin" In microscopic sections (pis. I F, II A-B) it can Koch, 1887), mainly of the "siamese type,

be seen that the larva consists of an ectodermal and as a result either of incomplete fusion of two

entodermal The an layer separated by a mesogloea. eggs, or a separate development (without complete cell initial entoderm contains zooxanthellae in the case of separation) of each of the doublet

larvae of E. singularis singularis, the most common (% 3).

with the is the it form and the only one infested these Once planula expulsed by polyp

commensal In the entoderm will start to fall down in vertical . young planulae a position (fig.

is continuous with the central yolk material, which 4a), its density being slightly superior to that of

seawater. In of the current is gradually digested in older larvae, leaving a presence even slightest

the coelenteron. The entoderm also the larvae will be carried the same void, primitive away, adopting

contains contractile with some fibres, which enable the passive position, although increasing current

larvae side the tend be in horizontal to bend to one or other, or to speed they to swept away a

certain such head-on Once larva reaches the it will contract upon stimuli, as position. a bottom,

with other with the substratum its collision larvae, or contact chemi- start crawling over by means of

cals (e.g. formaldehyde) (fig. 2e). The ectoderm cilia, either in a perfect translation (fig. 4b) or

is a ciliated epithelium, which confers mobility to accompanied by a dextrogyrous rotation along its

the larvae. main axis (fig. 4c). The maximal speed we

Each polyp emits several larvae. Theodor measured over short distances in larvae of E.

has estimated that medium-sized female is min 1 thus faster than (1967b) a 18 cm 50% singularis ,

colony of E. singularis emits some 6000 planulae the values found by Theodor (1967b).

the which in larva will often during spawning season, begins June However, a stop, remaining

recorded the first substratum (in Banyuls-sur-Mer we planulae motionless, or exploring the at a given

on 19 June 1976, 8 June 1977 and 13 June 1978, spot by a rapid, mostly counterclockwise rotation

and last till the end of before its initial respectively), may July. (fig. 4d) resuming course. More-

Surface will it will travel colonies (warmer water) spawn over, hardly ever along a straight line, earlier in the season than deeper ones, a phenom- and the resulting mean speed we measured over a

also enon observed by Grigg (1977). distance of 120 cm was at most about 2.5 cm

Among the thousands of larvae obtained when min 1 . We obtained this result by liberating about keeping ripe female colonies in aquaria, several 300 larvae at one spot of a large (diam. ca. 75 cm) dozens will show malformations (see also Von ring-shaped basin used for experiments on larval

Malformations of the “siamese Fig. 3. occurring among planulae, mostly twin” type.

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 130 S. WEINBERG & F. WEINBERG - GORGONIAN LIFE CYCLE

in This basin before phototaxis (Weinberg, preparation). phase to explore a vast surface settling

into 16 the distance down. was divided sectors, average

from to the next 15 We observed that in of favourable one sector being ca. cm. presence a

After 55 minutes the first larvae released in sec- substratum (the underside of gorgonian hold- tor 1 reached sector 9 (the opposite sector), fasts), settlement takes place withinapproximately either by following the left or the right half of the 30 hours. During this period, the slowest larvae

the annular basin, and the number of larvae present can explore substratum over a distance of ca. in each 40 The sector was counted. Numbers of planulae 2 m, the fastest over a distance of ca. m.

and the settlement present in the left sector 2 right sector triggering factors for are a suitable

2 were added together (they had travelled at the roughness of the substratum (Cary, 1914; Gohar,

the same speed); same was done for sectors 3 1940; Theodor, 1967b; personal observations),

this through 8. This yielded a speed spectrum for absence of other organisms (Theodor, 1967b;

shows the and illumination group of larvae; fig. 5 percentage of personal observations) optimum larvae corresponding to each speed interval after a (Weinberg, in preparation). time of 55 minutes. Although these velocities are Once such a place has been selected after the rather low, they will enable the larvae during the exploratory behaviour of fig. 4d, the larva adopts

1 several hours to several days ) of their mobile an upright position, its posterior end touching the

substratum (fig. 4e). Possibly, the cement ne-

for attachment is this cessary secreted during

while sometimes phase, a slow rotation is observed.

This settlement behaviour involving several suc-

cessive explorations of the substratum is typical

of larvae of marine invertebrates many (Meadows

Planula behaviour, larvae Fig. 4. a, Passive vertical floating position & Campbell, 1972). As a whole, octocoral from the translation upon expulsion polyp. b, Crawling: over and their behaviour (see also De Lacaze-Duthiers, substratum of by means cilia, c, Do., accompanied by dextro- 1864, Von Koch, Theodor, 1967b; gyrous rotation, d, Exploratory behaviour: quick (mostly 1900; 1887;

rotation on the same Pre-settlement differ countercllockwise) spot. e, Suzuki, 1971; Grigg, 1977) do not very behaviour: upright standing on substratum, sometimes accom- much from thoseof Hexacorallia (Atoda, 1947a & panied by slow rotation. b, 1951a, b & c, 1953; Lewis, 1974; Vander-

meulen, 1974) except for the fact that the latter

develop a mouth during their pelagic phase.

5. SETTLEMENT AND METAMORPHOSIS

We have never been able to witness the actual

settling of a larva. We know that some 30 hours

after emission, several planulae are found firmly

attached to a suitable substratum. In the termi-

nology of Mileikovsky (1971) the larvae of E.

demersal de- singularis exhibit a lecithotrophic

velopment. Whereas during the pre-settlement

behaviour the larvae touched the substratum with

their tapered (posterior) end, attached larvae have

their thickest part adhered to the bottom. Whether 5. of h - 1 of of Fig. Spectrum average speed (in cm ) a group the larvae have inverted position orsimply changed 300 larvae (see text). Ordinate is number of larvae N

attachment we do not know. (in %) for each speed class. shape during The subsequent phases of metamorphosis have

been studied by time-lapse cinematography of 14 Theodor reports a maximum life of 122 *) (1967b) span settled larvae over a of four and days in vitro, which must be considered exceptional. period days,

Downloaded from Brill.com10/08/2021 12:28:07AM via free access BIJDRAGEN TOT DE DIERKUNDE, 48 (2) - 1979 131

Fig. 6. Metamorphosis of the planula: 1 = settlement; 2 = invagination; 3 = septation; 4, 5 = development of primordial

6 = of 7 = of sclerites; 8 = complete tentacles; appearance primordial pinnules; pinnules clearly visible, appearance primary polyp.

additional several larvae Four observations on other days after settlement the primary polyp is

(figs. 6 & 7). Five to ten hours after attachment complete (pi. II D).

This limited the larvae start to invaginate, accompanied by a chronology (fig. 7), based on a shortening and thickening of their body. After number of polyps, is only approximate, but cor- another five to ten hours interval, the septae are responds remarkably well with the one given by formed (pi. II C), and the metamorphosing larvae Suzuki (1971) for Cornularia sagamiensis, and the reach their minimum length at this point. Their data presented for several Hexacorallia by Atoda

is hollow with & & Variations structure entirely now, a primitive (1947a b, 1951a, b c, 1953)- occur mouth A and internal septae. After a period of rest from one individual to another. complete histo- of ten to twenty hours, the primordial tentacles logical description of the changes involved in

Octocorallia of in start developing, being clearly visible two days the Xenia can be found after settlement, and the lateral pinnules start Gohar (1940), whereas Von Koch (1887) gives growing on each tentacle. One day later, the a description for E. cavolinii. pinnules are well developed, and the first sclerites 6. THE FIRST YEAR white the translucent appear as dots on pink polyp-

walls. start small fraction the larvae ean They forming eight vertical rows, Only a very of reaches

under each and the the one tentacle, spicular sheath primary polyp stage. Many are lost by alighting

invades the of the substrata gradually proximal part polyp. on unfavourable (soft bottoms, or

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 132 S. WEINBERG & F. WEINBERG - GORGONIAN LIFE CYCLE

size from 2 to 10 mm (ca. 5 months old), one

with size 40 17 months a from 10 to mm (ca.

old) and one with a size from 30 to 70 mm (ca.

29 months old). These data suggest a growth

of 8 to 1 least the first rate 30 mm year at during

A colonies year(s). number of one year old (sum-

Rederis mer 1978) at 20 m depth at Cap

(Banyuls) measured between 13 and 21 mm

(mean: 17 mm).

One might expect, however, that with increasing

size, food capture becomes more efficient, and

abrasion or smothering by sediment causing ne-

crosis of living tissues becomes less important,

that rates increase with so growth may age.

with Fig. 7. Average chronology of metamorphosis, average

larval length, based on time-lapse cinematography of several

polyps. Numbers between brackets refer to stages in fig. 6.

eaten fishes plant surfaces), probably many are by

have or other organisms, although we never

actually witnessed the capture of a planula. Thor-

the 8. Histogram of number of colonies of son (1950) assumes depredation to be most Fig. young (N) Eunicella singularis found at Port Vendres in December larvae. 1977, important cause of death of planktonic in height (h) classes of 5 mm each. Curves show (tentatively)

In situ observations over several months of primary three different generations. polyps in several underwaterstations near Banyuls-

sur-Mer have convinced us that only about one or

hundred will survive the first two polyps out of a

sediment year. Abrasion and/or smothering by or

the main death at this algae are among causes

stage. Theodor (1967b) has estimated that of

60000 larvae will survive the emitted, only one

first year.

We have tried raise in vitro. to primary polyps 9. of unbranched Fig. Comparison two young colonies, one

differs Secondary polyps formed by budding sometimes (A) and one ramified (B). Although total length in have the A and the third both, they same age, as colony occurred as soon as a fortnight after completion branch of colony B took the same time (approximately) to of the After four primary polyp. months, budding reach height h.

had yielded colonies of 2-5 polyps, with a 7. ANNUAL GROWTH maximum height of 8 mm (pi. HE). Shortly

died after, these experimental colonies due to a Before attempting to determine annual growth, we

technical accident, making further observations have to define what annual growth in a gorgonian

impossible. is. Cary (1914) and Grigg (1974) measured

Young colonies measured at 18 and 26 m depth increase in colony height, whereas Velimirov

Vendres in in total the near the harbour pier of Port December (1975) measured increase length, i.e.

fall into three size classes of the of all the branches of 1977 (fig. 8), pre- sum lengths a colony.

different with sumably three generations: one a Colonies are formed by budding sequences, some

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 48 - BIJDRAGEN TOT DE DIERKUNDE, (2) 1979 133

buds If the and 11 10 which started starting lateral branches. we compare at 15 m depth at m depth,

unbranched colony of fig. 9A with the ramified November 28th, 1976. The first series was soon

of would of the all one fig. 9B, we say they were equal interrupted, as tags nearly disappeared.

when as a Another series of 26 branchlets age taking colony height criterium, tagged replaced

whereas on the basis of total length the second the initial 15 m experiment on April 12th, 1977.

than the number lost colony would be considered much older Although a of tags were here too,

first. branches of both the 10 m and 15 m series were

Our field observations have shown that each measured with intervals until July 1978 when

branch grows independently from the others. 7 and 11 tagged branches subsisted, respectively

A Hence, assuming equal growth rates, colonies (fig. 11).

and B have the branch Due to intrinsic and external factors individual same age, as 3, growing

4 and took fact independently from branches 1, 2, 5, as growth rates can vary to a large extent, a

A h. For esti- also Some branchlets long as colony to reach height age reported by Grigg (1974).

mates it is better therefore to measure colony showed negative growth, probably due to abrasion

height than total length. This method is an and/or predation. These few branchlets were left

approximation, however, as in most cases colony out of our calculations. The others showed indi-

height is smaller than the real maximum size for vidual growth rates ranging from 0 to 49 mm

1 i.e. the 1 This is faster than the to m . 5.2 21.5 colony growth, longest budding sequence year year

in shown in 10A B. The Velimirov for the Mediter- a colony, as fig. & latter reported by (1975)

the is wider E. but slower than the figure represents case of a colony that ranean species cavolinii,

than its height, in which case colony height yields growth rates observed for Gorgonia flabellum

estimate of its It is accurate 1758 1 and Plexaura a poor age. even more Linnaeus, (0-83 mm year )

1 to count growth rings, which have annual peri- flexuosa Lamouroux, 1821 (5-55 mm year ) the Californian odicity (Grigg, 1974), especially since growth near Florida (Cary, 1914) or

differ from individual rates may considerably one Muriceacalifornica (Aurivillius, 1931) (0-60 mm

will be shown later 1 Kiikenthal to another, as on. However, year ) reported by Grigg (1974).

for and since undertook with E. practical reasons, counting growth (1909) growth experiments

is measured He off branch of colonies rings a destructive method, we length singularis. cut tips

from distant and measured a gorgonian foothold to the most maintainedin an aquarium, a growth branch of 60 in 22 995 1 This tip. mm days (= mm year )!

Our first attempt to measure growth rates of E. singularis consisted in the tagging of 47 branch-

lets lie The near Grosse, Banyuls-sur-Mer. first

experiment consisted of two series, 10 branchlets

of Fig. 11. Average increase in length (∆L) as a function of branches of time in months (M) for a number tagged

Eunicella singularis at 10 m depth (open circles) and 15 m

Ile Numbers 10. of is depth (black dots) near Grosse, Banyuls-sur-Mer. Fig. Length longest budding sequence (black) a number of branches found back time of better dimension for colony size than its height, especially in indicate tagged at

-1 the of colonies that broader than their like = 14.6 case are height, each measurement. Increase is linear; ∆L mm year

B. -1 colony at 10 m, 12.1 mm year at 15 m.

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 134 S. WEINBERG ft F. WEINBERG - GORGONIAN LIFE CYCLE

growth rate is obviously exceptional, and caused whereas 4 branchlets of Lophogorgia ceratophyta

if observation is reliable 1 at . by the cutting, the all, yielded an average growth rate of 28.5 mm year

which A 26 we doubt. similar experiment carried out at m depth

based of all Rederis with of Paramuricea Curves on average growth given at Cap a young colony

clavata branches in each spot (fig. 11) strongly suggest a (6 branches) yielded an average growth

such 1 linear and no seasonal influences of 12.5 mm growth rate year .

I as observed for E. cavolinii by Velimirov (1975). Table summarizes the data obtained by the

Linear correlation co- different methods. Whereas to have regression yields very high tagging seems

and the of efficients (0.996 for the 10 m experiment a negative effect on growth of branches

0.998 for the 15 m experiment) and the obtained E. singularis, P. clavata with its much thicker

rates 14.6 1 in the branches is affected this average growth are mm year not by method.

1 in 2 first case and 12.1 mm year the second ).

Table I These values seem a little low when compared to

1 observation of colonies Growth rates mm ) of three Mediterranean those obtained by young (in year- gor-

gonians as determined by three different methods. A = (preceding paragraph). We suspected that damage average growth rates of tagged branches; B = average growth caused the which forced the by tagging itself, of = rates branches of untagged colonies; C estimated wounds maximal after observation of branchlets to repair inflicted by friction growth rates gorgonians on of known sewer pipes age. of the wire, might have slowed down growth rates.

We therefore undertook a second attempt, i.e. Eunicella Lophogorgia Paramuricea

size substrata of known singularis ceratophyta clavata measuring colony on age,

method used several of coral a by investigators A 14.5 — 18.3

B 22.4 growth (Grigg, 1974; Buddemeier & Kinzie, 28.5 12.5

C 33.0 24.0 — In counted 142 1976). December 1977 we gor-

the gonians at 20 m depth on a section of sewer

8. AGEING AND DEATH pipe of Port Vendres. Ten of these gorgonians

measured 160-180 mm. As this concrete pipe was Years of observation on gorgonians never led us

installed in March 1972, the oldest colonies had to witness obvious signs of ageing in these animals.

in the small settled summer of 1972, and were therefore In and large colonies alike, the tissues and

5.5 old when measured them. This the base oldest of the years we polyps at (the part colony)

maximal and the yielded a growth rate (over longer always look as healthy active as those at

of 33 1 3 formed branch Whereas individual periods) mm year ). newly tips.

The between these data and the and the discrepancy polyps may age die, regeneration capacity preceding ones led us to undertake a third attempt. of the colonies (Kiikenthal, 1909; Cary, 1914;

On December 9th, 1977, a number of stones Lang da Silveira & Van 't Hof, 1977) will lead to

rack individuals. bearing gorgonians were installed on a metal quick healing or replacement of lost

Bear. It other at 20 m depth at Cap Instead of tagging, is obvious, on the hand, that for each each colony was carefully drawn and all branches species there is a maximum colony size never ex- measured. The Two same was repeated on June 22nd, ceeded. possible causes, determinate growth

death this 1978. In these undisturbed branches growth rates or at size, are unlikely. Whereas death

in 0-60 is if the varied from 0-32 mm 195 days, i.e. mm hard to conceive no ageing takes place,

1 The obtained for nature of these animals year . average growth rate 25 very colonial, branching branchlets 1 of E. was 22.4 mm year makes the of determinate singularis , hypothesis growth equal-

ly improbable. As integration of the colony is 2 ) For comparison: of 26 colonies of Paramuricea tagging there be for very poor, seems to no reason budding 24 clavata at m depth yielded an average growth rate of 18.3 to at the branch since the 1 1.6-36.9 1 stop tips, especially mm year (range: mm year ). 3 Likewise, 48 is ) counting gorgonians at 31 m depth on the colony probably not able to "tell" it has reached of which installed sewer pipe Banyuls-sur-Mer, was spring a certain size. Many workers on Hexacorallia have maximal of 24 1 the 1965, yielded a growth rate mm year for also assumed or indeterminate in species Lophogorgia ceratophyta. proved growth

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 48 - BIJDRAGEN TOT DE DIERKUNDE, (2) 1979 135

Schismo- most colonial corals (e.g. Buddemeier & Kinzie, Porella cervicornis (Fleming, 1828) and

if Whereas 1976). If we assume indeterminate growth and pora avicularis (Lamouroux, 1812). death is the result of these animals have small adherence not ageing, only one pos- only a fairly

the sibility remains: destruction of colonies by zone on the gorgonian itself, therefore not directly

both external agents. killing the colony in the way previously men-

tioned fast ulti- One such agent could be predation. We never octocorals do, their growth will

the to either witnessed, however, any important damage due to mately cause gorgonian topple over, by

know these predation. We of no larger organism (e.g. the sheer weight of calcareous colonies, or by

The the in currents fish) feeding on Mediterranean gorgonians. increased resulting drag water

of E. have ob- only predator singularis that we (pi. HID).

is small ovulid served a gastropod, Neosimnia The observation of toppled colonies, many of spelta spelta (Linnaeus, 1758), capable of de- them entirely devoid of epibionts (pi. II F) even-

the natural nuding small portions of the branches of their tually led us to the hypothesis that living tissue (pi. Ill A) (see also Theodor, 1967a). death of gorgonians is also dictated by the forces

A similar behaviour is the hold- displayed by Pseudosimnia exercised on the fan. Up to a certain size,

will this size is carnea carnea (Poiret, 1789), feeding on deep- fast be able to resist; whenever water gorgonians (e.g. Eunicella verrucosa (Pallas, exceeded, on a day of strong current or heavy

1766)) in the Mediterranean. In the Caribbean, swell, these colonies will be torn off their sub- related ovulid gastropods, Cymbula acicularis (La- stratum. Similar observations have been made by marck, 1810) and Simnialena uniplicata (Sowerby, Grigg (1977) for colonies of Muricea in Califor-

various and for 1848) feed on colonies of gorgonians, as nia, Caribbean shallow-water gorgonians, does after Once the "Flamingo Tongue" snail Cyphoma gib- especially a hurricane (Cary, 1914). bosum 4 and will (Linnaeus, 1758) ) the polychaete the gorgonians are toppled over, they die by

Hermodice carunculata (Pallas, 1766). necrosis of the living tissues in contact with the

Whereas the wounds inflicted these will be buried under sedi- by pre- bottom, or they simply dators are certainly not lethal, the denuded por- ment. of of this tions the skeleton may become the site settle- In an attempt to verify hypothesis, we

in ment of other organisms. Two Octocorallia are measured the largest colonies a population at

in the colonization Par- it that specialized of gorgonians: different depths. First of all was seen the erythropodium coralloides (Pallas, 1766) and tallest colonies in the population were always less

than others. ramified Rolandia rosea (Philippi, 1842) (see also Wein- ramified Broader, more berg, 1975, 1977, 1978). Once settled, these colonies never attained the maximum height for a animals maximum fan grow very fast, gradually pushing back given depth. The surface remained

2 the living gorgonian tissue and investing the entire rather constant however (2145 cm for a 55 cm

2 colony. This behaviour is apparent in the photo- high poorly ramified specimen, 2257 cm for a of graph pi. Ill B. It is not rare to find gorgonian 37 cm high colony with many ramifications, both

either of these mechanistic axes completely overgrown by at 25 m at Cap 1'Abeille), making our Octocorallia (pi. Ill C) 5 ). explanation for gorgonian death rather plausible.

invertebrates It that with Many other or algae may settle on was further observed increasing depth denuded portions of gorgonian axes. Frequently (and decreasing water movement) the maximal

in encountered are the lamellibranch Pteria hirundo height each population of E. singularis increases.

At station measured (Meuschen, 1787) and several large bryozoans, one near Cap Rederis we a such 64 as Hippodiplosia fascialis (Pallas, 1766), maximal height of 33 cm at 12 m depth, cm

at 25 m depth and 81 cm at 40 m depth. These

tend confirm that destructive 4 of findings to water ) Many species the gastropod family are asso- ciated with gorgonians all over the world (Cate, 1973). movement is the main factor for limiting gor- In the of the °) Caribbean, hydrocorals genus Millepora are gonian growth. able invest to gorgonians in much the same way (personal These data, combined with those on observations). growth

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 136 S. WEINBERG & F. WEINBERG - GORGONIAN LIFE CYCLE

and the lead to conclusion that colonies of Euni- 1951a. The larva postlarval development of rates, the , reef-building corals. III. Acropora briiggemanni (Brook). cella singularis may live as long as 25-30 years, J. Morph., 89 (1): 1-13. over whereas Lophogorgia ceratophyta may 1951b. The larva and development of the grow , postlarval and clavata A corals. IV. Galaxea Quelch. J. 35, Paramuricea over 50 years. reef-building asperea Morph., 89 (1): 17-30. similar was estimated for colonies of Muricea age of 1951c. The larva and development some , postlarval The fast californica (see Grigg, 1974). growing reef-building corals. V. Seriatopora hystrix Dana. Sci.

Tohoku Caribbean gorgonian Gorgonia flabellum (see Rep. Univ., (4, biol.), 19 (1): 33-39. the 1953. The larva and postlarval development of attain obser- , Cary, 1914) may 150 cm (personal reef-building coral. Sci. Rep. Tohoku Univ., (4, biol.), also life of 20-30 vations), indicating a span years. 20 (1): 105-121.

BUDDEMEIER, R. W. & R. A. KJNZIF., 1976. Coral growth.

Biol., 14: 9. CONCLUSION Oceanogr. mar. 183-225. de la CARPINE, C. & M. GRASSHOFF, 1975. Les gorgonaires of characterized E. singulars are by 71 Populations Mediterranee. Bull. Inst, oceanogr. Monaco, (1430):

long-lived individuals (low turnover rate) with a 1-140. and CARY, L. R., 1914. Observations upon the growth-rate rather short reproduction period. In the concept oecology of gorgonians. Pap. Tortugas Lab. Carnegie Inst. of the r-K continuum, as worked out by Pianka Wash., 5 (182): 81-90.

revision of the (1970) this corresponds to K selection, charac- CATE, C. N., 1973. A systematic recent cypraeid family Ovulidae (: ). teristic of populations living at their maximum Veliger, 15 (Suppl.): i-iv, 1-116. size in a stable environment, to which the indi- GOHAR, H. A. F., 1940. The development of some Xeniidae

Pubis, charac- some viduals are well adapted. However, another (Alcyonaria) (with ecological aspects).

mar. biol. Stn. Ghardaqa, 3: 27-79. teristic of E. singularis is its high rate of recruit- GRIGG, R. W., 1974. Growth rings: annual periodicity in ment with larval and juvenile mortality, high two gorgonian corals. Ecology, 55: 876-881.

of The found of two corals. typical r selection. same was by , 1977. Population dynamics gorgonian

Ecology, 58: 278-290. Grigg (1977) for populations of Muricea in Cali- D'HONDT, M.-J. & A. TIXIER-DURI VAULT, 1975. Clavularia fornia. Most coral populations probably occupy steveninoae nouvel Octocoralliaire de n. sp., intermediate in the r-K Mediterranee. Cah. Biol, 16: 585-592. such an position con- mar.,

G. 1887. Die Gorgoniden des Golfes von Neapel known the KOCH, VON, tinuum, one exception being oppor- und der angrenzenden Meeresabschnitte. Exster Theil tunistic Red Sea species Stylophora pistillata einer Monographic der Anthozoa Alcyonaria. Fauna (Esper, 1791) (see Loya, 1976). Flora Golf. Neapel, 15: i-x, 1-99, pis. I-X. KOWALEWSKY, A. O., 1873. On the development of Coelen-

terata 1-36, I-YIII (Moscow Univer- ACKNOWLEDGEMENTS [in Russian]: pis. sity Press, Moscow). Labora- We wish to thank Dr. Jacques Soyer, Director of the KOWALEWSKY, A. O. & A.-F. MARION, 1882. Sur le develop- toire at for the hospitality and Arago Banyuls-sur-Mer hebd. Acad. Sci., pement des Alcyonaires. C. r. Seanc. working facilities in his laboratory. Mr. Jean Lecomte, CNRS- Paris, 95: 562-565. photographer, is kindly acknowledged for his help in making & 1883. Documents l'histoire embryo- , pour and the the time-lapse film on metamorphosing planulae genique des Alcyonaires. Annls. Mus. Hist. nat. Mar- photography of pi. II E. Prof. Dr. Hajo Schmidt and Dr. seille, 1 (4): 1-50, pis. I-V. Wolfgang Schafer (Zoologisches Institut, Heidelberg Uni- KUKENTHAL, W., 1909. Beobachtungen an einigen Korallen- versity) made the photographies of pi. I D-E, and Dr. Ernst tieren des Adriatischen Meeres. Aus der Natur, 5 (11): (Lehrstuhl fur Zellmorphologie, Bochum Univer- Kniprath 321-328. of I F and II A-B. We sity) those pis. are very grateful to bei , 1919. Eireifung und Spermatogenese den Gorgo- Albus and them, as we are to Drs. Henk Mr. Jean Mabit, narien. Zodl. Anz., 50 (6-7): 164-166. CNRS-diver, who helped with underwater measurements. 1925. Erste Unterklasse der Anthozoa: Octocorallia. In: , 87-117 from The senior author was supported by the grant T. KRUMBACH ed., Handbuch der Zoologie, I: 690-769 Netherlands Organization for the Advancement of Pure (Walter de Gruyter & Co., Berlin/Leipzig).

Research (Z.W.O.). naturelle du LACAZE-DUTHIERS, H. DE, 1864. Histoire corail.

Organisation, reproduction, peche en Algerie, industrie REFERENCES et commerce: i-xxv, 1-371, pis. I-XX (J. B. Bailliere et

ATODA, K., 1947a. The larva and postlarvil development of fils, Paris).

Coralliaires du Golfe du Lion. some corals. I. Pocillopora damicornis 1900. Alcyonaires reef-building ,

Sci. Tohoku Zool. 353-462. cespitosa (Dana). Rep. Univ., (4, biol.), Archs. exp. gen., 3 (8):

18 (1): 24-47. LANG DA SILVEIRA, F. & T. VAN'T HOP, 1977. Regeneration 1947b. Plexaura flexuosa Octo- , The larva and postlarval development of some in the gorgonian (, reef-building corals. II. Stylophora pistillata (Esper). corallia). Bijdr. Dierk., 47 (1): 98-108.

Sci. Rep. Tohoku Univ., (4, biol.), 18 (1): 48-64. LEWIS, J. B., 1974. The settlement behaviour of planulae

Downloaded from Brill.com10/08/2021 12:28:07AM via free access BIJDRAGEN TOT DE DIERKUNDE, 48 (2) - 1979 137

larvae of the hermatypic coral Favia fragum (Esper). J. micrograph by S. Weinberg). D, Detail of the peripheral

Biol. 15: 165-172. of of exp. mar. Ecol., cytoplasm an ovocyte a gorgonian (Corallium rubrum),

LOYA, Y., 1976. The Red Sea coral Stylophora pistillata is an containing electron-dense yolk granules (dg), electron-light

r strategist. Nature, Lond., 259 (5543): 478-480. yolk granules (lg) and lipid droplets (li); the ovocyte is

MEADOWS, P. S. & J. I. CAMPBELL, 1972. Habitat selection surrounded by the mesogloea (mg) and entoderm (en) of the

10: W. by aquatic invertebrates. Adv. mar. Biol., 271-382. polypean septum; 4500 X (electron micrograph by

of larval in cell of E. residual MILEIKOWSKY, S. A., 1971. Types development Schafer). E, Immature sperm cavolinii;

marine bottom invertebrates, their distribution and cytoplasm (re) of spermatocyte surrounds the nucleus (n)

are visible; ecological significance: a re-evaluation. Mar. Biol., Ber- and the flagellum (f) between which centrioles

lin, 10: 193-213. the flagellum is separated from the cytoplasm by a flagellar

und be MOSER, J., 1919. Eireifung, Spermatogenese erste Ent- canal (fc); no acrosome or mitochondrial sheath are to

wicklung der Alcyonarien. Zool. Anz., 50 (6-7): 159- seen; 24000 X (electron micrograph by H. Schmidt). F,

164. Longitudinal section of a planula larva of E. singularis;

and K selection. Am. (en) can be PIANKA, E. R„ 1970. On r Nat., 104 ectoderm (ec), mesogloea (mg) and entoderm

592-597. distinguished; the barely visible cilia (ci) are 20-30 /im long;

Cornularia Al- SUZUKI, H., 1971. Notes on (Stolonifera, 60 X (phase-contrast micrograph by E. Kniprath). cyonaria) found in the vicinity of the Manazuru Marine

Biological Laboratory. Sci. Rep. Yokohama natn. Univ., Plate II 2 (18): 1-6. A, Detail of cross section of planula larva of E. singularis; 1967a. l'etude des THEODOR, J., Contribution a gorgones ectoderm with cilia and entoderm des (ec) (ci), mesogloea (mg) (VI): la denudation des branches de gorgones par

(en) are visible; 580 X (light micrograph by E. Kniprath). mollusques predateurs. Vie Milieu, 18 (1A): 73-78. B, Detail of planula larva of E. singularis; the ectoderm 1967b. Contribution a l'etude des , gorgones (VII): consists of epithelial cells (ep) with microvilli (mv), cilia ecologie et comportement de la planula. Vie Milieu, 18 (ci) and osmiophilic vacuoles (arrows) and of glandular cells (2A): 291-301. (gc); the entoderm contains vitelline material (vi) and THORSON, G., 1950. Reproductive and larval ecology of symbiotic zooxanthellae (2); both layers are separated by the marine bottom invertebrates. Biol. Rev. Cambridge phil. mesogloea (mg); 2000 X (electron micrograph by E. Knip- Soc., 25 (1): 1-45. Cross section settled larva of E. V rath). C, through a singu- ANDERMEULEN, J. H., 1974. Studies on reef corals. II. Fine in of ectoderm laris the process septation; (ec), mesogloea structure of planktonic planula larva of Pocillopora (mg) and entoderm (en) surround the gastric cavity (gc) damicornis, with emphasis on the aboral epidermis. Mar. resulting from the digestion of the vitelline material from the Biol., Berlin, 27: 239-249. planula larva; septae (s) originate from mesogloea and ento- VELIMIROV, B., 1975. Wachstum und Altersbestimmung der derm (see inset); 90 X, inset 750 X (light micrograph by Gorgonie Eunicella cavolinii. Oecologia, 19: 259-272. S. D, Cross section through primary polyp of la du Weinberg). VIGHI, M., 1972. Etude sur reproduction Corallium E. singularis; holes resulting from the dissolution of sclerites rubrum (L.). Vie Milieu, 23 (1A): 21-32.

can be seen in ectoderm and Contribution de (sc) (ec) mesogloea (mg); WEINBERG, S., 1975. a la connaissance mesogloea and entoderm (en) are continuous with the Parerythropodium coralloides (Pallas, 1766) (Octo- septae (s) in which the musculature (m) is visible; columnar corallia, ). Beaufortia, 23 (298): 53-73. epithelium of ectoderm surrounds the stomodaeum (st) with , 1976. Revision of the common Octocorallia of the its ciliated siphonoglyph (si); 90 X (light micrograph by Mediterranean circalittoral. I. Gorgonacea. Beaufortia, S. Weinberg). E, Four months old colonies of E. singularis; 24 (313): 63-104. 6 X (photograph by J. Lecomte). F, One normally attached 1977. Revision of the common Octocorallia of the , colony of E. singularis (upper left) and several torn-off Mediterranean circalittoral. II. Alcyonacea. Beaufortia,

colonies lying on the bottom, some of them (barely visible, 25 (326): 131-166. arrows) are only skeletons, having lost their (white) living 1978. Revision of the common Octocorallia of the , tissue (photograph by S. Weinberg). Mediterranean circalittoral. III. Stolonifera. Beaufortia,

27 (338): 139-176.

Plate III LEGENDS TO THE PLATES

A, The ovulid gastropod Neosimnia spelta spelta feeding on Plate I branch of 8 of a E. singularis; X. B, Colony the alcyonacean of A, Diver at a depth 20 m in Eunicella-“meadow”, col- Parerythropodium coralloides (dark coloured) spreading over female colonies of lecting ripe (photograph by S. Weinberg). the basal parts of a colony E. singularis. C, Colony of

of Eunicella in the stoloniferan B, Ovocyte (o) singularis septum; septal E. singularis, completely overgrown by Ro-

musculature is visible; 120 X (m) clearly (light micrograph landia rosea next to healthy colonies (right). D, Colony of by S. Weinberg). C, Ovocyte of E. singularis surrounded E. singularis with several epibionts, the octocoral Parerythro- by a layer of entodermal cells (en); cytoplasm (c), nucleus podium coralloides (pc) and the bryozoan Hippodiplosia

(n) and nucleolus (nu) can be distinguished; 240 X (light fascialis (hf). (All photographs by S. Weinberg.)

Received: 13 September 1978

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 138/I S. WEINBERG & F. WEINBERG - GORGONIAN LIFE CYCLE

Plate I

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 48 - BIJDRAGEN TOT DE DIERKUNDE, (2) 1979 139/II

Plate II

Downloaded from Brill.com10/08/2021 12:28:07AM via free access 140/III S. WEINBERG & F. WEINBERG - GORGONIAN LIFE CYCLE

Plate III

Downloaded from Brill.com10/08/2021 12:28:07AM via free access