ILKD+AFOP>K@B AFOMBNO>I ?U MI>KHPLKF@ I>NR>B LC … · 2012. 9. 26. · _]bchic^ 0LLOCENTROTUS FRAGILIS ^_f[s_^ m_nnf_g_hn `il 030 ^[sm, H_gj` %/76/& `ioh^ nb[n p_fca_lm i` nb_ ijcmnbi\l[h]b

BULLETIN OF MARINE SCIENCE, 39(2): 241-256,1986LARVAL INVERTEBRATE WORKSHOP

LONG-DISTANCE DISPERSAL BY PLANKTONIC LARVAE OFSHOAL-WATER BENTHIC INVERTEBRATES AMONG

CENTRAL PACIFIC ISLANDS

Rudolf S. Scheltema

ABSTRACTThe geological origin of central Pacific islands by volcanic activity at "hot spots" through

a pre-existing ocean floor requires that their original colonization by invertebrate organismsmust have resulted from long-distance dispersal. It is proposed that among most tropical,shoal-water, sediment-dwelling invertebrates such dispersal was largely accomplished bymeans of planktonic larvae. Evidence to support this hypothesis comes from plankton samplestaken in the area bounded by 1400E to 1400W longitude and from 400N to 40"S latitude whichshow that larvae ofa wide variety of benthic invertebrates are found in the epipelagic watersof the tropical Pacific, including but not restricted to gastropod and bivalve molluscs, sipun-culans, polychaetes, echinoderms and brachyuran decapod crustacea. Data from drift bottlesin the South Equatorial Current give a model surface velocity between 20 and 25 km day-I,enough to account for dispersal of teleplanic larvae for at least 2,000-4,000 km over the 3-6 months' period of their pelagic existence. The number oflarvae that may be successfullytransported between islands will depend on (a) the number that "escape" the influence of the"parent" or donor island, (b) the mortality oflarvae during transport, and (c) the chance thatlarvae that encounter an island survive to reproduce. It is estimated on the basis of drift-bottle returns (assuming a 1% survival) that once a larva is entrained in the South EquatorialCurrent, its chance of encountering an island is about 3 x 10-4 (ca. I in 3,300). Fecundity(usually large insofar as known) and the size of the "parent" population (difficult to estimate)must also importantly influence the frequency of exchange between islands by teleplaniclarvae. The relatively low endemism and attenuated Indo-Pacific faunas found on centralPacific islands support the hypothesis that there is not only (a) sufficient capacity for dispersalto colonize new islands as they appear, but also (b) that dispersal and the attendant gene flowbetween islands is frequent enough to account for the persistence of most shoal-water, sed-iment-dwelling Indo-Pacific species (rather than the allopatric speciation and high endemismfound among many terrestrial forms). Data on the dispersal of particular species in conjunctionwith information on their genetic variation between islands is required to further test suchan hypothesis.

That planktonic larvae are transported over long distances by ocean currentshas been firmly established for a wide variety of warm-temperate and tropicalbenthic invertebrates (Mileikovsky, 1966; 1968; Robertson, 1964; Scheltema,1964; 1968; 197Ia). This knowledge has been acquired through direct observa-tions of planktonic larvae in the epipelagic waters of the open ocean. Includedhave been the cerianthula and sempers larvae of coelenterates; the metatrocha,chaetosphera, mitraria and rostraria larvae of polychaetes; bivalve and gastropodveligers; zoea and megalopa ofbrachyuran crabs; stomatopod larvae; the naupliiand cyprid larvae of cirripeds; pluteus, bipinnaria and auricularia larvae of echi-noderms; and the tornaria larvae of Enteropneusta. I Among those whose dispersal

, Coe/enlermes-Leloup, 1964; Scheitema, 1968; 1971a; Po/ychaeles-Hiicker, 1898; Scheltema, 1971a; 1974; Sipuncu/ans-Hiicker,1898: Scheltema and Hall, 1975: Diva/I'es-Allen and Scheltcma, 1972; Scheitema, 1971c; Simroth, 1896; Thiede, 1974; Gaslropods-Laurscn, 1981; Lebour, 1945: Scheitema, 1971b; 1977; 1978; 1979; Simroth, 1895; Thorson, 1961; Drachyuran Decapods-Evans,1961; Rice, 1970; S/Omalopods-Lebour, 1934; Cirripeds-Roskcll, 1975; Scheitema and Cariton, 1984; Echinoderms-Monensen,1898; 1913; Scheitema, 1977; EllIeropneusla-Bjornberg, 1955; Hadfield, 1978; Scheltema, 1970.

241

242 BULLETIN OF MARINE SCIENCE, VOL. 39, NO.2, 1986

has been most extensively studied are the larvae of gastropods because theirveligers can be identified by indirect means through comparison ofthe larval shellwith intact proto conchs of identified juvenile or adult museum specimens (Thor-son, 1950; Shuto, 1974; Scheltema, 1977; Jablonski and Lutz, 1980).

Most invertebrate larvae captured in open waters of the North and SouthAtlantic Ocean and subsequently held in laboratory culture retained their com-petence to settle and metamorphose. Veligers of certain gastropods (e.g., Archi-tectonicidae, Cymatiidae) and bivalves (e.g., Pinnidae), the metatrocha larvae ofchaetopterid polychaetes (e.g., Chaetopterus, Phyllochaetopterus), the pelago-sphera larvae ofsipunculans (e.g., Sipuncula, Aspidosipha, etc., Rice, 1981; Halland Scheltema, 1975), the tornaria larvae of hemichordates (e.g., Ptychodera),and the zoea and megalopa larvae of various oxystomatid crabs (e.g., Calappa,Raninoides) survive and metamorphose in culture when removed from open oceanplankton samples (unpubl. obs. unless noted above).

In the laboratory Strathmann (1978) showed that the larvae of many speciesof echinoderms, including holothuroideans, asteroids, ophiuroids, and echinoids,can delay metamorphosis for long periods of time; for example, plutei of theechinoid Allocentrotus fragilis delayed settlement for 252 days. Kempf (1981)found that veligers of the opisthobranch Aplysiajuliana could attain competenceto settle after only 28 days, but were able to delay settlement and retain the abilityto metamorphose up to 311 days in the absence of an appropriate environmentalcue for metamorphosis. Survival of A. juliana larvae in the natural environmentis apparently sufficient to account for its wide Indo-West Pacific distribution.Many other larvae found in the epipelagic waters of the open ocean very likelywill also be found to settle and metamorphose when provided with an adequatesurface and the proper cues. Larvae that delay settlement and that have morpho-logical and physiological adaptations for a long planktonic life (e.g., Cymatiidae,Richter, 1984; Pechenik et aI., 1984) are termed "long distance" (Thorson, 1961)or teleplanic larvae (Scheltema, 1971b, p. 12, from teleplanos, Gr., far wandering).

Owing to their geologic origins, oceanic islands present a singular opportunityfor investigating the role of long-distance dispersal for the biogeography ofsublittoraF invertebrate species. The Central Pacific or Polynesian islands all havearisen on the large Pacific plate from volcanic activity and therefore show discretediscontinuities with the sea floor from which they arise. Differences in age ofislands along a chain are the result of their sequential formation as the Pacificplate moved toward the northwestward (e.g., the Hawaiian archipelago, Dalrympleet aI., 1973) and passed over "hot spots" resulting from convection currents withinthe earth's mantle (Morgan, 1972). Since the sea floor pre-dates the origin of allPolynesian island arcs, no connection can have ever existed with continental landmasses, and hence colonization can only have come about as the result of someform of long-distance dispersal. There are two principal ways that long-distancedispersal may occur among benthic invertebrates across otherwise impassablebiogeographic barriers; these are (1) rafting (Scheltema, 1977, p. 91; Scheltemaand Carlton, 1984) under which will also be included "island integration" (Ro-tondo et aI., 1981), and (2) the transport of planktonic larvae on ocean currents(Scheltema, 1971a). Anecdotal evidence exists for rafting of a variety of inver-tebrate taxa (Highsmith, 1985; Scheltema, 1977). Especially interesting recentdocumented instances are the dispersal of sexually mature polyps of the herma-

2The term "sublittoral" is llsed here as in Hedgepeth (1967, p. 18, fig. I) 10 include those species that occur subtidally to approximately200 m. As oceanic islands do not have shelves, the terminology of Ekman (1953, Chapter I), "warm-water fauna of the shelf," isinappropriate.

SCHELTEMA:LARVALDISPERSALOFCENTRALPACIFICINVERTEBRATES 243

typic coral Porites on drifting pumice (Jokiel, 1984) and the transport of intertidaland sublittoral mollusks on the carapace of marine turtles (Frazier et aI., 1985).

Clonal species (e.g., compound ascidians) usually have either a very short plank-tonic stage (often less than an hour) or completely lack a free-drifting larva, yetmany have very wide geographic ranges (Jackson, 1986). Because clonal benthicforms usually attach to hard substrata they include those species that are mostlikely to be rafted. On the other hand, among infaunal species, only few are likelyto be rafted since most are unable to survive outside of sediments for protractedperiods. Some epibenthic species have the possibility for more than a single modeof transport (e.g., barnacles, serpulids, oysters, small gastropods, etc.) and maybe dispersed by either rafting or planktonic larvae.

This paper is not concerned with rafting but rather deals with the evidence forlong-distance dispersal of planktonic larvae and its possible consequences for thegeographic distribution of shoal-water sublittoral invertebrate species among cen-tral Pacific islands. Specifically, I (1) demonstrate that larvae ofa wide variety ofbenthic invertebrate taxa actually occur in surface waters of the central PacificOcean and are available to colonize oceanic islands; (2) give evidence from theliterature that shows that the currents are sufficient to account for the transportof such larvae over long distances and between islands; (3) consider the probabilitythat a larva will encounter an island once entrained within an Equatorial Current;and finally (4) argue that the exchange oflarvae between the populations of variousislands is very likely sufficient to account for the low endemism of shoal water,sediment-dwelling invertebrate species of the central Pacific islands.

METHODS

Data on the occurrence of teleplanic larvae in the Pacific are based on 210 oblique plankton towscollected over a period of 26 years between 1953 and 1979. Recent samples specifically taken for thepresent studies consisted of 36 20-min oblique tows with nets ofO. 75 m diameter from approximately150 m to the surface on a voyage of the R/V KNORRbetween Wellington, New Zealand and Honolulu,Hawaii from 18 January to 8 February 1979 (Scheltema and Williams, 1983). These samples werereduced to a volume of 0.5-1.0 liter, depending upon the quantity of zooplankton taken, and subse-quently preserved in 10% formalin with sodium borate added to control pH. Detailed data for theabove cruise of the R/V KNORR (Cruise 73) can be found in the archives of the Woods Hole Ocean-ographic Institution or at the National Oceanographic Data Center. The remaining 174 samples camefrom the plankton collections of the Scripps Institution of Oceanography and lie within the regionbounded by 140·E to 140·W longitude and 400N to 40·S latitude. The detailed data for these collectionsare given by Snyder and Fieminger (1965; 1972). The samples selected were all obtained by obliquetows with plankton nets of I m diameter. Material was examined from the following expeditions: (I)Transpac, November 1953, Snyder and Fieminger, 1965, p. 49, Chart 8,15-25 min, 110-150 mdepth, Sta. 123A-136A incl. (2) Capricorn, November-January 1952-1953, Snyder and Fieminger,1965, p. 39, Chart 12,25-30 min, 200 m depth, Sta. 1-17 incl. (3) Troll, March 1955, Snyder andFleminger, 1965, p. 63, Chart 25, 15 min, 200 m depth, Sta. OA-5A; 7; 8A-16A. (4) Equapac (R/VHORIZON),August 1956, Snyder and Fleminger, 1965, p. 78, Chart 8, 0.5 h, 400 m depth, Sta. 2-41incl. (5) Equapac (R/V STRANGER),August-September 1956, Snyder and Fleminger, 1965, p. 79, Chart8, 15 min, 140 m depth, Sta. 1-46 incl. (6) Monsoon, September 1960; March-April 1961, Snyderand Fleminger, 1965, p. 105, Chart 36-37, 30 min, ca. 200-300 m depth, Sta. 3-5; 34-40 incl. (7)Circe, April 1968, Snyder and Fleminger, 1972, p. 28, Chart II, 15 min, ca. 200 m, Sta. 1-8. (8) Scan,June-July 1969, Snyder and Fleminger, 1972, p. 122, Chart 39, 15 min, 200 m depth, Sta. 4-6, 40P,450,460,490, Leg IV; 1-8 Leg V. (9) Antipode, August 1970, Snyder and Fleminger, 1972, p. 9,Chart 5, 0.5 h, Sta. 4, Leg IV. In all instances, the entire sample was sorted under a Wild stereomi-croscope at 12 x magnification.

RESULTS

A summary of the relative frequency at which major taxa of invertebrate larvaeare found in the central Pacific (Table 1) shows an essential similarity to that of

244 BULLETIN OF MARINE SCIENCE, VOL. 39, NO.2, 1986

Table I. Percentage occurrence of planktonic larvae of benthic invertebrate taxa in surface watersof the Central Pacific Ocean (Based on 210 stations shown in Figs. 1-4)

Taxon and larvallype

Sipunculan pelagospheraGastropod veligerPolychaete "chaetosphera," metatrocha, etc.Brachyuran zoea and megalopaBivalve veligerEnteropneust tomariaEchinoderm (pluteus, bipinnaria, auricularia)Phoronid actinotrocha

No. stations

171163126106995129

5

Percentage occurrence

81.477.660.050.547.124.313.82.4

the tropical Atlantic Ocean, with gastropods and sipunculan larvae the mostcommon in both oceans (Scheltema, 1971a). These two taxa, wherever they occur,include species that have teleplanic larvae either morphologically and physiolog-ically adapted fOf a delay in settlement and a long planktonic life.

Among gastropod veligers found in the Pacific (Table 2, Fig. 1), the 12 mostcommon families are also represented in the Atlantic. Indeed, some species be-longing to families appearing in Table 2 either occur or have analogs in theAtlantic. Among the bivalves (Fig. 1) there are also similarities. For example, theveligers of the Pinnidae and Teredinidae commonly occur in samples from bothoceans (Scheltema and Williams, 1983). A common bivalve encountered is"Planktomya henseni," a form which occurs in tropical waters throughout theworld. The larva (perhaps to more than one species) is of the superfamily Lep-tonacea and probably the family Montacutidae (=?Sportellidae) (Moore, 1983).It has not been possible to induce Planktomya henseni to metamorphose in thelaboratory; however, members ofthe Montacutidae are all commensal, and prob-ably will not settle without the presence oftheir host species. Larvae of the familyPinnidae have been previously reported in Polynesian waters (Scheltema andWilliams, 1983).

Pelagosphera larvae of sipunculans were found at more than half the locations(Fig. 2); most appear to belong to the same "forms" (possibly genera) as those inthe Atlantic. Although living sipunculan larvae from plankton samples are readily

Table 2. Gastropod families whose veliger larvae are commonly found in surface waters of theCentral Pacific Ocean (Based on 210 stations shown in Fig. I)

Taxon

All gastropodsArchitectonicidaeCymatiidaeNaticidaeNeritidaeCypraeidaeThaididaeTriphoridaeCoralliophilidaeBursidaeTonnidaeStrombidaeOther*

No. stations

16388634228272523191211II

129

Percentage occurrence

77.641.930.020.013.312.911.911.09.05.75.25.2

61.4• Includes: Cassidae, Columbellidae, Cerilhiidae, Turridae, Ovulidae, Conidae, Litiopidae and Muricidae at

SCHELTEMA: LARVAL DISPERSAL OF CENTRAL PACIFIC INVERTEBRATES 245

400

. Maldeg·StarbuCj<

Marquesas :

. iles Tubuai

• gast roped vehgers

o bivalve vehgers() botho negative stations

. \:'Kermadec

/0()

••

ftql\-• qi'• ·0 1'11)~ l.s

.~. I~~'()Cl(J. O'.s..~••

Norfolk\ .

NEWZEALAND

()()

Figure 1. Distribution of epipelagic gastropod and bivalve veligers from 210 plankton tows takenin the central and west Pacific Ocean. For families represented see Table 2. Symbols are defined atlower right.

maintained to settlement in the laboratory, their identification presents a difficultproblem owing in part to their morphologic plasticity and to changes during post-larval ontogeny.

Families of polychaetes whose teleplanic larvae are most commonly found inthe central Pacific include Spionidae and Chaetopteridae (Table 3, Fig. 2). "Chae-tosphera" larvae (Hacker, 1898) ofSpionidae, common in the Atlantic (Sche1tema,1971a, fig. 5), are also abundant from the central Pacific. According to Bhaud(1984) at least six kinds of "chaetosphera" larvae which can be referred to thegenus Scolelepis are found off New Caledonia. The Chaetopteridae include thegenus Chaetopterus and probably also Phyllochaetopterus and Spiochaetopterus;the latter two genera must be confirmed by inspection of the modified seta of the4th segment using scanning electron microscopy (Bhaud, 1978; Scheltema, 1974).

Ophiopluteus, echinopluteus and bipinnaria larvae of ophiuroids, echinoids,

246 BULLETIN OF MARINE SCIENCE, YOLo 39, NO.2, 1986

20°

0°

20°

40°

40°160°

. iles Tubuai

. Maldea·Starbuc;j<

Marquesas :

o polychaete larvae• pelagosDhaera() bothlO negahve stations

. .}·Cqok

o

o

Samoa

•

()Tonga

180°

()o

NEWZEALAND

160°

JAPAN

'.~/()()

Figure 2. Distribution of epipelagic polychaete and sipunculan larvae from 210 plankton tows takenin the central and west Pacific Ocean. For families represented see Table 3. Symbols are defined atlower right.

and asteroids, respectively, are found in the North and South Equatorial Currentsof the Pacific (Fig. 3) (Mortensen, 1898; Scheltema, 1977, shows Atlantic distri-bution of echinoderm larvae). The skeletal elements of plutei can be used toidentify some genera and species (Mortensen, 1921 and subsequent papers), butthese structures do not preserve well in samples with a low pH. Most ofthe larvaefrom the older Scripps samples examined are therefore not identifiable.

Finally, the locations at which zoea and megalopa larvae of decapod crustaceawere recorded in tropical epipelagic waters of the Pacific are shown in Figure 4.Families included, but were not restricted to, Calappidae, Raninidae and otheroxystomatids; Dromiidae, Portunidae and Grapsidae. These families are amongthose known to include widely distributed tropical Pacific species (Yaldwyn, 1973).

The evidence shown in Figures 1 through 4 and summarized in Tables 1 through

SCHELTEMA: LARVAL DlSPERSALOF CENTRAL PACIFIC INVERTEBRATES 247

20°

0°

. i1esTubuai

Marquesas :

. MaldeQ.Starbuq,'k

• eChinopluteuso ophiopluteus() oatho auricularia+ bipinnariao negative stations

° °. C k'0

Tonga

NEWZEALAND

--

()o °0:, Marianao 0 ~0 IslandsLJ!

carolineOd%lands. . ~t\ '"... , ~;.

Figure 3. Distribution of epipelagic bipinnaria, echino- and ophiopluteus larvae from 210 planktontows taken in the central and west Pacific Ocean. Symbols are defined at lower right.

3 demonstrates that larvae of a wide variety of benthic invertebrate taxa actuallyoccur in surface waters of the central Pacific Ocean.

DISCUSSION

The relationship between passive long-distance dispersal by planktonic larvaeand the geographic distribution of benthic invertebrates in the tropical Pacific hasbeen the subject of considerable intermittent speculation (Ekman, 1953; Garth,1966; Zinsmeister and Emerson, 1979; Horikoshi, 1982). Most of these discus-sions have been based on inferences from or the knowledge of other regions (suchas the Atlantic); practically no direct evidence existed for the transport of inver-tebrate larvae in tropical Pacific waters (Marumo and Kitou, 1956; Scheltemaand Williams, 1983). The data presented here show not unexpectedly that larvaeof a wide variety of benthic invertebrate taxa do indeed occur in the epipelagic

248 BULLETIN OF MARINE SCIENCE, VOL. 39, NO.2, 1986

200

200

400

o

iles Tubual

• megalopao zoea() botho none

l-1600

oSamoa 0

"-0 0TongC? 0 0

.0

o

¢ '"~

o '\(, man -8.',p Imyra

o • Wash. ngl6hf o· Famlng

g '- , "Christmas A\oll~ 0 'Jarvis

p.§I1t:loenlx 0 ~ I

O

• . '008 ~ +- . Maldea'Starbu'tl'

Tok!!.lau 0 MarQuesas :

. \'Kermadec. 10

1800

o

°Ellice

NEWZEALAND

1600

Figure 4, Distribution of epipelagic zoeae and megalopae of brachyuran decapod crustaceans from210 plankton tows in the central and west Pacific Ocean. Symbols are defined at lower right.

of the central Pacific and that potentially may provide a means for colonizationof newly formed islands as well as for genetic exchange between shoal-waterbenthic populations of existing islands. Successful dispersal of larvae betweenislands will depend in part also upon (1) the proximity of islands to one another,and (2) the position of islands with respect to ocean currents.

Although it is quite impossible to follow individual larvae over long distancesat sea, one can nonetheless, from knowing the location of its capture, gain aninsight about the probable origin as well as the route a larva may have taken hadit remained adrift. Since larvae can only be carried "downstream" along an oceancurrent, the direction and rate of transport can be predicted, although only ap-proximately, from a knowledge of the net direction and velocity of surface currents.

The Pacific circulation forms two very large anti-cyclonic gyres, one in thenorthern and the other in the southern hemisphere (Knauss, 1966; Warren, 1966).Their tropical portions are designated the North and South Equatorial Currents,

SCHELTEMA: LARVAL DISPERSAL OF CENTRAL PACIFIC INVERTEBRATES 249

Table 3. Polychaete families whose larvae are commonly found in surface waters of the central PacificOcean (Based on 210 stations shown in Fig. 2)

Taxon

All polychaetesSpionidaeChaetopteridaeAmphinomidaeSabellariidaeOther*

·Owcniidae and other families at 2,000 km the rate is 26.5 km day-I,whereas for those that drifted

250 BULLETIN OF MARINE SCIENCE, VOL. 39, NO.2, 1986

Table 4. Positions of release, point of recovery and mean rate of transport for drift bottles releasedin the central Pacific Ocean·

Rate ofDislancet Dayst drift

drifted to re· (kmPosition of release Point of recovery (km) covery day-I)

6°18'N, 175°33'W (34) 7"31'N, I 52"05'E; Onari Is. 3,573 208 17.27°1O'N, 171°13'E; Ine, Amo Atoll 1,463 74 19.8

2°52'N, 179°34'W (37) 5°19'N, 162°59'E; Tafunsak, E. Caroline Is, 1,954 79 24.73°08'N, I72°54'E; Makin Atoll, Gilbert Is. 837 15 55.8

1033'N,171"OI'E (38) 1°24'S, 138°45'E; St. Mathias, New Guinea 4,486 86 52.27°25'N, 151°47'E; Epitar, Caroline Is. 3,180 122 26.1

0014'N, 177°44'E (39) 0055'N, I73"00'E, Maiana, Gilbert Is. 532 8 66.5§532 13 40.9532 34 15.6

0°56'5, 176°22'E (40) 7°47'S, I 56°22'E; Vella Lavella, Solomons 2,341 257 9.15"00'5, I 55°20'E; Green Is., New Guinea 2,376 57 41.72°40'S, 147"00'E; Nihau Is., Manus, New Guinea 3,265 86 38.0

10"28'8, 150"30'E; Finschhafen, New Guinea 3,047 230 13.22°25'S, 175°11'E (41) 5°1O'S, 154°30'E; Bougainville, New Guinea 2,493 58 43.03°27'S, 173°10'E (42) I 1025'S, I52"00'E; Conflict Group, Papua 2,493 238 10.5

10049'S, I65°57'E; Nambalue, Santa Cruz Is. 1,140 147 7.84°57'S, I 72°35'E (43) 8°21'5, 162°40'E; Sikaiana, Solomon Is. 1,158 97 11.92044'S, 159°34'W (46) 1043'S, 142°54'E; Wulvulu Is. 1,854 454 4.170s, 142°W (WHOI) 5°30'S, 145°45'E; Madang, Papua New Guinea 7,975 569 14.0• Data from Barkley et aI., 1964, pp. 20-22, Cruise 55, R/V CHARLESH. GILBERT.Numbers in parentheses after the position of releaserefer to station numbers in original source. Source oflast entry (WHOl) from "Anonymous, 1982" (see Literature Cited).t "'Distance drifted" is the "great circle" or shonest distance, usually underestimates the actual distance.* "Days to recovery" overestimates the number of days adrift since there is a variable and unknown time interval between strandingand recovery of bottles.§ Includes II bottles treated as one observation.

Twenty-eight or 1.3% of the 2,127 bottles released were eventually recoveredon islands (Table 4; "WHOI" bottle has not been included in calculations becauseit is not known whether other bottles were cast at this location at the same time).All these recoveries were from bottles released in the South Equatorial Currentbetween January and early April. If the frequency of returns is computed solelyon the 980 bottles released in the South Equatorial Current, then the percentagebecomes 2.9%. This estimate for the stranding of drift bottles is probably con-servative for a variety of reasons, e.g., many returns may not be reported becausethe finders could not read the card written in English inside the bottle or had noway to post a reply. Nonetheless, from the frequency of returns it can be proposed,as an hypothesis, that once entrained within the South Equatorial Current, theodds that a larva encounters an island (other factors disregarded) lies between 2to 3%.

Such an estimate does not take into account losses from mortality by predationor disease. However, few published estimates exist of predation on larvae innatural populations (Mileikovsky, 1959), and the diseases oflarvae, such as fungi,are mostly known from contamination of laboratory cultures.

Even when mortality is very high, the frequency at which larvae reach islandsmay, nonetheless, be significant because this number depends also on the numberof larvae entrained into the current system and thus on (1) the fecundity of aspecies and the size of the source population, and (2) the chance that a larva willescape the local circulation around an island and get carried out to sea to enterthe ocean circulation.

SCHELTEMA:LARVALDISPERSALOFCENTRALPACIFICINVERTEBRATES 251

140 160 180 160

20

o

20

000 ac:

FIJI4

O·

SAMOA

"'"

-,

~.

~----~ -_.~~~.=--..

CAROLINE IS.

Figure 5. Locations of drift bottle release and recovery in the central and west Pacific Ocean. (Datafrom Barkley et a1., 1964; R/V CHARLESH. GILBERT,cruise #54, Sta. 54-12 through 54-43; cnlise#55, Sta. 16 through 63). Lines connect points of release and recovery and suggest possible routes ofsurface transport. Data on distance and rate of transport are given in Table 4. Bottles along transect"a" were released between the end of September and the beginning of December 1961; there were norecoveries. Bottles along transect "b" were released in January 1962; recoveries were obtained from8 of the 28 points of release (28.6%). However, all came from 10 release points between 6°N to 50Slatitude, where the transect crosses the South Equatorial Current. Bottles along transect "c" werereleased between the last week in March and first week in Apri11962; recoveries were obtained fromonly I of the 20 points of release (shown by dashed line). This return was from I of 8 release pointswithin the South Equatorial Current. Forty bottles were released at each of the locations along thethree transects except where closely spaced near the Hawaiian Islands; here 10 were released. Location"d" indicates a single release point northwest of the Marquesas Islands (Anon., WHOI, 1981) madeby E. A. Balloch in January 1981. See text for discussion.

The fecundity of species with teleplanic larvae, in the few instances where thisis known, is generally quite high. Wilson (1985, p. 76) notes that some tropicalcowries (Cypraeidae) may have from 10,000 to 500,000 veligers from a single eggmass. Tun shells (Tonnidae) have large egg masses with 300,000-600,000 ova(Scheltema, 1971b, for references). Thais haemastoma, a thaidid with a teleplanicveliger, produces 500,000 larvae per year (Butler, 1954). However, the fecundityof most invertebrate species with teleplanic larvae is not precisely known.

The question of retention of invertebrate larvae around oceanic islands andtheir subsequent recruitment to local populations has been almost unstudied eventhough it has been suspected for some time that hydrographic mechanisms mustexist to keep planktonic organisms around oceanic islands (Boden, 1952; Emery,1972). Hadfield (1978) offered biological evidence that seems to show that tomarialarvae of Ptychoderaflava are retained in local waters near Hawaii for 3-9 months.Studies by Lobel and Robinson (1983) show how coral-reef fish larvae may beretained around Hawaii by off-shore quasigeostrophic eddies, and propose thatlarvae probably find their way back to their "parent" or other nearby populations.Since the occurrence of these eddies is seasonal it is probable that the escape of

252 BULLETIN OF MARINE SCIENCE, VOL. 39, NO.2, 1986

invertebrate larvae from the influence of island circulation will also prove to varyseasonally. Indeed, timing of reproduction may become important for the reten-tion, or alternatively the "escape," of larvae from island circulations. The fore-going considerations show that the dispersal between islands is probably com-plicated by numerous variables that need further research.

Once a teleplanic larva is entrained within the major oceanic circulation suchas the South Equatorial Current, the theoretical chance that it will encounter anisland (assuming a 99% mortality) would be in the order of 3 x 10-4 [0.03(=estimate of island encounter from drift bottles) x 0.01 (=an assumed survival),Scheltema, 1971b; 1978 for further discussion]. How many larvae will actuallysettle and survive to sexual maturity is, however, another question. The abovetheoretical example shows that if the number oflarvae entrained is large (500,000or more, not an unrealistic assumption for gastropods) then the odds that at leastsome reach another island become relatively good, even though much temporalvariation may occur. Dispersal need not be continuous and may vary from yearto year and still be important for chance colonization and the genetic exchangebetween islands. To summarize: What is not known for understanding the chance-for and frequency-of exchanges between islands are (1) data on the numbers oflarvae that "escape" their parent island to become entrained in the oceanic cir-culation, (2) the extent oflarval mortality in the open sea, and (3) the chance thatpost-larvae that encounter another island will survive to reproduction.

What is known is that larvae of many shoal-water, benthic species are carriedout to sea and can be consistently found there. Teleplanic larvae can and dosurvive for long periods in the epipelagic waters of the open ocean. Given sufficienttime, the chance that some of these larvae will settle and survive is virtuallycertain (Simpson, 1952), but to fully understand the significance oflong-distancelarval dispersal it will be necessary to obtain information on the other areas ofignorance outlined above.

The proportion of endemic species on oceanic islands provides indirect evidencefor their isolation or conversely the frequency of successful immigration andexchange with other islands. Endemism can also be affected by (1) the geologicage of an island (i.e., enough time must have passed for species to evolve); (2)historic events over geologic time, for example (a) changes in sea level which canresult in an increase or decrease in the number of "stepping" stones, the guyotsand islands that act as intermediate points for the spread of species, (b) climaticchanges that may limit latitudinal range, and (c) changes in the position of islandsrelative to major oceanic currents; and (3) ecological constraints, the biologicalinteractions which may determine whether or not a species will survive, or becomeextinct. Specific examples show that the percent endemism of shoal-water benthicgastropods, brachyuran decapods, polychaetes and echinoderms among those cen-tral Pacific islands where it is known, ranges between 2 and 20%, much lowerthan that of terrestrial fauna and flora (Table 5). For example, the endemism ofinsects of Hawaii is 93.8% (Carlquist, 1974). Although the estimates shown inTable 5 are open to modification as the fauna of central Pacific islands becomesbetter known, it is unlikely that the conclusions from these data will changesubstantially.

The benthic marine invertebrates of central Pacific islands represent an atten-uated Indo-Pacific fauna (Salvat, 1967) and even include some species found alsoin east Africa and the Red Sea. According to conventional wisdom, for a speciesto exist over such a large range and on such widely scattered islands, gene flowbetween the separated populations must be at least great enough to offset any

SCHELTEMA: LARVAL DISPERSAL OF CENTRAL PACIFIC INVERTEBRATES 253

Table 5. Some representative published values for endemism among benthic invertebrates on centralPacific islands

PercentTaxon Location No. species endemic Reference

Gastropoda Ducie Atoll 38 12.0 Rehder and Randall (1975)Line Islands 229 2.0 Kay (1971)Hawaiian Islands 598 19.2 Kay (1979)Easter Island 12.0* Steele (1957)

Brachyura Easter Island 21 9.5 Garth (1973)Polychaeta Easter Island 43 9.3 Kohn and Lloyd (1973)Echinodermata S.E. Polynesia 20 15.0 Marsh (1974)• This figure may be an underestimate. Rehder (1980) in a monograph on the EaSIer Island mollusks estimates 42% endemism. butthis value is inflated because 40 of the 48 endemic species are newly described. Among the 98 species of gastropods found by Rehder(excluding the pelagic forms, i.e., Recluzia and Jan/hina), 39 (ca. 40%) are newly described and mostly minute species. If new speciesare excluded, then only 17% are endemic to Easter and closely adjacent islands, 42% arc widely distributed Indo· Pacific forms, theremainder (41%) 3fC West Pacific or Pacific Plate species.

differentiation that may occur by genetic drift. And indeed, it is the prosobranchfamilies that include mostly species with teleplanic larvae (e.g., the Architecton-icidae, Coralliophilidae, Cymatiidae and Bursidae) that are also those that arewidely distributed on central Pacific islands. On the other hand, prosobranchfamilies such as the Volutidae, Vasidae and Cancellariidae, that lack species withplanktonic larval stages are quite restricted in geographic range and are absentfrom central Pacific islands.

Lewontin (1974) has pointed out "a migration rate as small as one individualin a thousand per generation is sufficient to prevent differentiation arising [fromrandom drift] between populations of moderate size." This conclusion suggeststhat the numbers of larvae needed to maintain genetic exchange between popu-lations is relatively small, and may be intermittent; it is probably less than thatrequired for the initial establishment of species on islands. The broad geographicrange and small tendency toward speciation found among the known sublittoralinvertebrate faunas of central Pacific islands support the hypothesis that a con-tinuous gene flow exists not only among these islands, but indeed also over largeareas of the Indo-Pacific. One way that such genetic exchange is likely to occuris by the dispersal of larvae. Further tests of such an hypothesis require data onboth the genetic variation among islands and the specific identification of larvae.

ACKNOWLEDGMENTS

I wish to thank I. P. Williams for her assistance in sorting some of the plankton samples and E. M.Hulburt for collecting samples between New Zealand and Hawaii. This research was made possiblein part by a grant from the National Science Foundation (OCE-8410262). Contribution Number 5838ofthe Woods Hole Oceanographic Institution.

LITERATURE CITED

Allen, J. A. and R. S. Scheltema. 1972. The functional morphology and geographical distributionof Planktomya henseni Simroth 1895, a supposed neotenous pelagic bivalve. J. Mar. BioI. Ass.,U.K. 52: 19-31.

Anonymous. 1982. Unknown oceanographer aids cruise passenger in drift bottle experiment. News-letter, Woods Hole Oceanographic Institution (September 1982).

Barkley, R. A., B. M. Ito and R. P. Brown. 1964. Releases and recoveries of drift bottles and cardsin the central Pacific. U.S. Fish Wildl. Ser., Spec. Sci. Rept. No. 492. 31 pp.

Bhaud, M. 1978. Morphological variations of the modified setae of chaetopterids during ontogenesis.Ophelia 17: 199-206.

--. 1984. Nouvelles donnees sur les larves Chetospheres de polychCtes (note preJiminaire).Oceanis 10: 697-706.

254 BULLETIN OF MARINE SCIENCE, VOL. 39, NO.2. 1986

Bjornberg, T. K. S. 1955. S6bre quarto tornarias do Atlantico e de Mediterriineo. Bol. Inst. Oceanogr.(Sao Paulo, Brasil) 6: 197-207.

Boden, B. P. 1952, Natural conservation of insular plankton. Nature 169: 697-699,Butler, P, A. 1954. The southern oyster drill. Proc. Nat. Shellfish Ass. 1953, Pp. 67-75.Carlquist, S. 1974. Island biology. Columbia University Press, New York. ix + 660 pp.Dalrymple, O. B., E. A. Silver and E. D, Jackson, 1973. Origin of the Hawaiian Islands. Amer. Sci,

61: 294-309,Ekman, S. 1953, Zoogeography of the sea. Sidgwick and Jackson, London. xiv + 417 pp.Emery, A. R. 1972. Eddy formation from an oceanic island: ecological effects. Carib. J. Sci. 12: 121-

128.Evans, F. 1961. The planktonic Crustacea of the Petula Transatlantic Expedition. Proc. Linn. Soc.,

Lond, 172: 189-207.Frazier, J., D. Margaritoulis, K. Muldoon, C. W, Potter, J. Rosewater, C. Ruckeschel and S. Sales,

1985, Epizoan communities on marine turtles. 1. Bivalves and gastropod Mollusca. P.S.Z.N.I.Marine Ecology 6: 127-140.

Garth, J. S. 1966. On the transport of crab larval stages, Proc. Symp. on Crustacea. Symp. Ser. 2,Mar. BioI. Ass, India, Pt. I, pp. 443-448.

--. 1973. The brachyuran crabs of Easter Island. Proc. California Acad. Sci. 4: 311-336.Hacker, V, 1898. Die pelagischen Polychaeten und Achaeten Larvae der Plankton-Expedition. Ergb.

d. Plankton Exped. Humboldt-Stiftung. Bd. II H. 50 pp, v pI.Hadfield, M. G. 1978. Growth and metamorphosis of planktonic larvae of Ptychoderajlava (Hemi-

chordata: Enteropneusta). Pages 247-254 in F. S. Chia and M. E. Rice, eds. Settlement andmetamorphosis of marine invertebrate larvae. Elsevier, New York and Oxford. xi + 290 pp.

Hall, 1. R. and R. S. Scheltema. 1975. Comparative morphology of open-ocean pelagosphaera. Pages183, 197 in M. E. Rice and T. Todorovic, eds. Proc. Internal. Symp. BioI. Sipuncula and Echiura,Vol. I. Inst. BioI. Res., "Sinisa Stankovic" Belgrade, Jugoslavia and Natl. Mus. Nat. Hist., Wash-ington, D.C. xxiii + 355 pp.

Hedgepeth, J. W. 1967. Classification ofthe marine environments. Pages 17-28 in J. W. Hedgepeth,ed. Treatise on marine ecology and paleoecology, Vol. 2. Oeol. Soc. Amer. Mem. 67. viii +1,296 pp.

Highsmith, R. C. 1985. Floating and algal rafting as potential dispersal mechanisms in broodinginvertebrates. Mar. Ecol. Prog. Ser. 25: 169-179.

Horikoshi, M. 1982. Limit of larval transport and the endemism in the shell fauna of the Pacific.Pages 118-122 in Fauna and hydrobiology of the shelf zones of the Pacific Ocean (Proc. XIVPacific Sci. Congr. 1979), Section "Marine Biology," Issue 4, Vladivostok. 148 pp. (In Russian:English summary)

Jablonski, D. and R. A. Lutz. 1980. Molluscan larval shell morphology: ecological and paleontologicalapplications. Pages 323-377 in D. C. Rhoads and R. A. Lutz, eds. Skeletal growth of aquaticorganisms. Plenum, New York. xiii + 750 pp.

Jackson, J. B. C. 1986. Dispersal and distribution of clonal and aclonal benthic invertebrates. Bull.Mar. Sci. 39: 588-606.

Jokiel, P. L. 1984. Long-distance dispersal of reef corals by rafting. Coral Reefs 3: 69-76.Kay, E. A. 1971. The littoral marine molluscs of Fanning Island. Pacific Sci. 25: 260-281.--. 1979. Hawaiian Marine Shells. Bernice P. Bishop Museum, Special Publication 64. Bishop

Museum Press, Honolulu, Hawaii. xvii + 653 pp.Kempf, S. C. 1981. Long-lived larvae of the gastropod, Aplysia juliana: do they disperse or just

slowly fade away? Mar. Ecol. Prog. Ser. 6: 61-65.Knauss, J. A. 1966. Equatorial currents. Pages 264-265 in R. W. Fairbridge, ed. The encyclopedia

of oceanography. Reinhold, New York. xiii + 1,021 pp.Kohn, A. 1. and M. C. Lloyd. 1973. Marine polychaete annelids of Easter Island. Int. Rev. Oes.

Hydrobiol. 68: 691-712.Laursen, D. 1981. Taxonomy and distribution ofteleplanic prosobranch larvae in the North Atlantic.

Dana Rpt. No. 89. 44 pp.Lebour, M. V. 1934. Larval Crustacea (Decapods and Stomatopods). Expedition S.A.R. Prince

Leopold of Belgium, Duke of Brabant, to the extreme east (1932). Bull. Mus. Roy. D'Hist. Natr.Belgique 10: 19-24.

--. 1945. The eggs and larvae of some prosobranchs from Bermuda. Proc. Zool. Soc. Lond.114: 462-489.

Leloup, E. 1964. Larves de Cerianthaires. Discovery Rpts. 33: 251-307.Lewontin, R. C. 1974. The genetic basis of evolutionary change. Columbia Univ. Press, New York.

xiii + 346 pp.Lobel, P. S. and A. R. Robinson. 1983. Reef fishes at sea: ocean currents and the advection oflarvae.

Pages 29-38 in M. L. Reaka, ed. The ecology of deep and shallow coral reefs. Symp. Ser. UnderseaRes. NOAA's Undersea Research Program, Vol. 1(1). 149 pp.

http://www.ingentaconnect.com/content/external-references?article=0722-4028(1984)3L.69[aid=7507274]http://www.ingentaconnect.com/content/external-references?article=0007-4977(1986)39L.588[aid=527918]http://www.ingentaconnect.com/content/external-references?article=0007-4977(1986)39L.588[aid=527918]http://www.ingentaconnect.com/content/external-references?article=0173-9565()6L.127[aid=8660519]

SCHELTEMA: LARVAL DISPERSAL OF CENTRAL PACIFIC INVERTEBRATES 255

Marsh, L. M. 1974. Shallow-water Asterozoans of Southeast Polynesia. I. Asteroida. Micronesia 10:65-104.

Marumo, R. and M. Kitou. 1956. Distribution of pelagic larvae of littoral animals in the open sea.Bull. Jap. Soc. Sci. Fish. 22: 225-228.

Mileikovsky, S. A. 1959. [Interrelations between pelagic larvae of Nephthys ciliata O. F. Miiller,Macorna baltica L. and Mya arenaria L. of the White Sea.] Zool. Zhurnal 38: 1889-1891. (InRussian)

---. 1966. [The range of dispersal of pelagic larvae of bottom invertebrates by ocean currentsand its distribution and role on the example of Gastropoda and Lamellibranchia.] Okeanologiya6: 482-493. (In Russian)

---. 1968. Some common features in the drift of pelagic larvae and juvenile stages of bottominvertebrates with marine currents in temperate regions. Sarsia 34: 209-216.

Moore, D. R. 1983. The little bivalve Planktornya unmasked. Am. Inst. Inv. Mar., Punta de Betin13: 123-132.

Morgan, W. J. 1972. Plate motions and deep mantle convection. Pages 7-22 in R. Shagan, ed. Studiesin earth and space science. Geol. Soc. Amer. Mem. 132. ix + 683 pp.

Mortensen, T. 1898. Die Echinodermenlarven der Plankton-Expedition. Ergb. Plankton-Exped. J.Humboldt-Stiftung Bd. II J. 120 pp, x pis.

---. 1913. Die Echinodermenlarven der Deutschen Siidpolar-Expedition 1901-1902. DeutscheSiidpolar-Exped. 1901-1903, Vol. 14: 67-111.

---. 1921. Studies of the development and larval forms of echinoderms. Vid. Medd. Dansk.Naturh. Foren. 71: 133-160.

Pechenik, J. A., R. S. S. Scheltema and L. S. Eyster. 1984. Growth stasis and limited shell calcificationin larvae of Cyrnatiurn parthenopeurn during trans-Atlantic transport. Science 224: 1091-1096.

Rehder, H. A. 1980. The marine mollusks of Easter Island (Isla de Pascua) and Sola y Gomez.Smithsonian Contr. Zool. 289. 167 pp.

-- and J. E. Randall. 1975. Ducie Atoll: its history, physiography and biota. Atoll Res. Bull.No. 183.40 pp.

Rice, A. L. 1970. Decapod crustacean larvae collected during the Indian Ocean Expedition. FamiliesRaninidae and Homolidae. Bull. Brit. Mus. (Nat. Hist.) Zool. 21: 1-24.

Rice, M. E. 1981. Larvae adrift: patterns and problems in life histories of sipunculans. Amer. Zool.21: 605-619.

Richter, G. 1984. Die gehauseentwicklung bei den larven der Cymatiiden (Prosobranchia: Tonnacea).Arch. Moll. 115: 125-141.

Robertson, R. 1964. Dispersal and wastage oflarval Philippa krebsii (Gastropoda: Architectonicidae)in the North Atlantic. Proc. Acad. Nat. Sci. Philadelphia 115: 1-27.

Roskell, J. 1975. Continuous plankton records: a plankton atlas of the North Atlantic and the NorthSea. Suppl. 2. The oceanic cirripede larvae, 1955-1972. Bull. Mar. Ecol. [Hull] 8: 185-199.

Rotondo, G. M., V. G. Springer, G. A. F. Scott and S. O. Schlanger. 1981. Plate movement andisland integration-a possible mechanism in the formation of endemic biotas, with special ref·erence to the Hawaiian Islands. Syst. Zool. 30: 12-21.

Salvat, B. 1967. Importance de la faune malacologique dans les Atolls Polynesiens. Cah. PacifiqueNo. II. Pp. 7-49.

Scheltema, R. S. 1964. Origin and dispersal of invertebrate larvae in the North Atlantic. Amer.Zool. 4: 299-300. (Abstract)

---. 1968. Dispersal of larvae by equatorial currents and its importance to the zoogeography ofshoal-water species. Nature 217: 1159-1161.

---. 1970. Two new records ofPlanctosphaera larvae (Hemichordata; Planctosphaeroidea). Mar.BioI. 7: 47-48.

---. 1971a. The dispersal oflarvae of shoal-water benthic invertebrate species over long distancesby ocean currents. Pages 7-28 in D. Crisp, ed. Fourth European Marine Biology Symposium,Cambridge Univ. Press. ix + 599 pp.

---. 1971b. Larval dispersal as a means of genetic exchange between geographically separatedpopulations of shallow-water benthic marine gastropods. BioI. Bull. 140: 284-322.

---. 1971c. Dispersal ofphytoplanktotrophic shipworm larvae over long distances. Mar. BioI.11:5-11.

---. 1974. Relationship of dispersal to geographical distribution and morphological variation inthe polychaete family Chaetopteridae. ThaI. Jugosl. 10: 297-312.

--_. 1977. Dispersal of marine invertebrate organisms: paleobiogeographic and biostratigraphicimplications. Pages 73-108 in E. G. Kauffman and J. E. Hazel, eds. Concepts and methods ofbiostratigraphy. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania. xiii + 658 pp.

---. 1978. On the relationship between dispersal of pelagic veliger larvae and the evolution ofmarine prosobranch gastropods. Pages 303-322 in B. Battaglia and J. A. Beardmore, eds. Marineorganisms-genetics, ecology, and evolution. Plenum Press, New York and London. x + 757 pp.

http://www.ingentaconnect.com/content/external-references?article=0025-3162(1970)7L.47[aid=8660506]http://www.ingentaconnect.com/content/external-references?article=0025-3162(1970)7L.47[aid=8660506]http://www.ingentaconnect.com/content/external-references?article=0028-0836(1968)217L.1159[aid=8149693]http://www.ingentaconnect.com/content/external-references?article=0028-0836(1968)217L.1159[aid=8149693]http://www.ingentaconnect.com/content/external-references?article=0003-1569(1981)21L.605[aid=7795982]http://www.ingentaconnect.com/content/external-references?article=0003-1569(1981)21L.605[aid=7795982]http://www.ingentaconnect.com/content/external-references?article=0036-8075(1984)224L.1091[aid=8660508]http://www.ingentaconnect.com/content/external-references?article=0036-8075(1984)224L.1091[aid=8660508]http://www.ingentaconnect.com/content/external-references?article=0025-3162()11L.5[aid=8660513]http://www.ingentaconnect.com/content/external-references?article=0025-3162()11L.5[aid=8660513]

256 BULLETINOFMARINESCIENCE,VOL.39, NO.2, 1986

---. 1979. Dispersal of pelagic larvae and the zoogeography of Tertiary marine benthic gastropods.Pages 391-397 in A. J. Boucot, J. Gray and J. Gray, eds. Historical biogeography, plate tectonics,and the changing environment. Oregon State Univ. Press, Corvallis, Oregon. xii + 500 pp.

--- and J. T. Carlton. 1984. Methods of dispersal among fouling organisms and possible con-sequences for range extension and geographical variation. Pages 126-133 in J. D. Costlow andR. C. Tipper, eds. Marine biodeterioration: an interdisciplinary study. Naval Inst. Press, Annap-olis, Maryland. xxi + 384 pp.

--- and J. R. Hall. 1975. The dispersal of pelagosphaera larvae by ocean currents and thegeographical distribution of sipunculans. Pages 103-115 in M. E. Rice and T. Todorovic, eds.Proc. Internat. Symp. BioI. Sipuncula and Echiura, Vol. I. Inst. BioI. Res. "Sinisa Stankovic"Belgrade, Jugoslavia and Natl. Mus. Natur. Hist., Washington, D.C. xxiii + 355 pp.

--- and I. P. Williams. 1983. Long-distance dispersal of planktonic larvae and the biogeographyand evolution of some Polynesian and western Pacific mollusks. Bull. Mar. Sci. 33: 545-565.

Shuto, T. 1974. Larval ecology of prosobranch gastropods and its bearing on biogeography andpaleontology. Lethaia 7: 239-256.

Simpson, G. G. 1952. Probabilities of dispersal in geologic time. Bull. Amer. Mus. Nat. Hist. 99:163-176.

Simroth, H. 1895. Die gastropoden der Plankton-Expedition. Ergb. Plankton-Exped. Humboldt-Stiftung. Bd. II F.d, 206 pp, xxii pIs.

---. 1896. Die Acephalen. Ergb. Plankton-Exp. d. Humboldt-Stiftung Bd. II F.e. 44 pp, iii pI.Snyder, H. G. and A. Fleminger. 1965. A catalogue of zooplankton samples in the marine invertebrate

collections of the Scripps Institution of Oceanography. S.1.0. Ref. No. 65-14 (15 Oct. 1965).--- and ---. 1972. A catalogue of zooplankton samples in the marine invertebrate collections

of Scripps Institution of Oceanography, Accessions 1965-70. S.I.O. Ref. No. 72-68 (I August1972).

Steele, P. H. 1957. Easter Island shells. Nautilus 70: 111-113.Strathmann, R. R. 1978. Length of pelagic period in echinoderms with feeding larvae from the

northeast Pacific. J. Exp. Mar. BioI. Ecol. 34: 23-27.Thiede, J. 1974. Marine bivalves: distribution of meroplanktonic shell-bearing larvae in Eastern

North Atlantic surface waters. Palaeog. Palaeochim. Palaeoecol. IS: 267-298.Thorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. BioI. Rev. 25:

1-45.---. 1961. Length of pelagic life in marine invertebrates as related to larval transport by ocean

currents. Pages 455-474 in M. Sears, ed. Oceanography. Pub!. 67. Amer. Assoc. Adv. Sci.,Washington, D.C.

Warren, B. 1966. Ocean circulation. Pages 590-597 in Encyclopedia of oceanography. Reinhold,New York. xiii + 1,021 pp.

Wilson, B. R. 1985. Direct development in southern Australian cowries (Gastropoda: Cypraeidae).Aust. J. Mar. Freshw. Res. 36: 267-280.

Yaldwyn, J. C. 1973. Decapod crustacea from South Pacific reefs and inlands. Pages 503-511 inOceanography of the South Pacific 1972. Compo R. Fraser, New Zealand Nat. Comm. forUNESCO.

Zinsmeister, W. J. and W. K. Emerson. 1979. The role of passive dispersal in the distribution ofhemi-pelagic invertebrates, with examples from the tropical Pacific Ocean. Veliger 22: 32-40.

DATEACCEPTED: January 14, 1986.

ADDRESS: Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543.