BULLETIN OF MARINE SCIENCE, 41(2): 192-203, 1987

DISTRIBUTION AND ECOLOGY OF DEEP- WATER CARIDEAN (CRUSTACEA; NATANTIA) NEAR TROPICAL PACIFIC ISLANDS

Michael King

ABSTRACT Deep-water caridean shrimps have been caught in baited traps set as deep as 800 m off many tropical Pacific Islands. Studies on the biology and distribution of several species (most belonging to the genus Heterocarpus) have been of both commercial and academic interest. The widespread distribution and abundance of these shrimps has encouraged some com- mercial attempts at exploitation, but the economics of the fishing operations have been marginal. Some of the studies on the ecology of deep-water shrimps have produced findings which run counter to competition-based ecological theory. Adult predation may decrease with increasing depth, allowing deeper-water species to have an extended lifespan, an in- creased degree of iteroparity, and a corresponding increase in lifetime reproductive effort. Furthermore, the probabilities of larval survival, which appear to decrease with increasing depth, may be offset by the production of larger eggs by deeper-water species.

Small numbers of deep-water caridean shrimps were found during trawling surveys in Hawaiian waters during 1967 and 1968 (Struhsakerand Yoshida, 1975). These species were later found to be susceptible to trapping (Clarke, 1972; Struh- saker and Aasted, 1974). Deep-water shrimps were subsequently found in the northern Pacific islands of Guam (Wilder, 1977) and the Northern Marianas (Moffitt, 1983). The most extensive surveys have been carried out in Hawaii and the Marianas where traps have been set in many different locations (Oishi, 1983; Ralston, in prep. I; Dailey and Ralston, in prep.2; Moffitt and Polovina, in prep.3). South of the equator, deep-water shrimps have been found in Fiji (Brown and King, 1979; King, 1984) and New Caledonia (Intes, 1978). The Fijian surveys are notable in that they were carried out over an extended time period in one location, beyond the barrier coral reef near Suva. The research vessel TAINUI obtained small numbers of deep-water shrimps in Tahiti in 1978 (P. Hatt, pers. comm. 1980). Catches of shrimps were made during a preliminary deep-water trapping survey in Vanuatu in 1980 (King, 1981 a); further surveys were conducted during 1981 and 1982 (De Reviers et al., 1982). Specimens of caridean shrimps were taken in Western Samoan waters by the French research vessel CORIOLIS(Anon., 1977) and during a fisheries resource survey in 1980 (King, 1980; 1984). In 1981, a preliminary survey was completed in Tonga (King, 1981b; 1984). During a demonstration in the use of traps in a consultancy in Papua New Guinea, small numbers of caridean shrimps were caught (King, 1982). At least 17 species of deep-water caridean shrimps have been found in Pacific islands. Some larger species, particularly those of the pandalid genus Heterocarpus, have elicited some commercial interest (King, 1981c; 1986). Some aspects of deep-water biology have also been of theoretical interest in the study of

I S. Ralston. An intensive fishing experiment for the caridean shrimp Helerocarpus fael'jgalus at Alamagan Island in the Mariana Archipelago. Pacific Scicncc.

2 M. D. Dailey and S. Ralston. Aspects of the reproductive biology, spatial distribution, growth and monality of the deepwater caridean shrimp Helerocarpus lae\'igatus in HawaiI. 'R. B. Moffitt and J. J. Polovina. Distribution and yield of the deepwater shrimp resource in the Marianas.

192 KING: DEEP. WATER SHRIMPS FROM PACIFIC ISLANDS 193

the evolution of life-history patterns at different depths (King and Butler, 1985). This paper reviews what is known of the biology and ecology of deep-water caridean shrimps in Pacific islands.

METHODS

Most surveys carried out from Pacific islands were similar in that baited traps were set, either singly or in a string, along transects running from depths of about 250 m to about 800 m and left in the sea overnight. A typical trap arrangement is shown in Figure I. Traps of several different types have been tested and used in Pacific islands; a typical trap type is shown in Figure I. Most traps were made from steel rod frames and covered with galvanized wire or plastic mesh and operate on the principle that shrimps, attracted to the bait hanging in the traps, enter cone-shaped entrances worked into the trap sides or top. Full descriptions of gear and methods used are given in King (1986).

RESULTS Species Found.-At least 17 species of caridean shrimps have been found during trapping surveys in Pacific islands. A list of all species found in Fiji, Vanuatu, Western Samoa, Tonga, the Marianas and Hawaii is given in Table 1; surveys in Vanuatu, Western Samoa and Tonga were preliminary. Descriptions and a key to the species found in southwestern Pacific islands are given in King (1984). Most published work refers to the distribution and biology of the Heterocarpus species which, because of their large size, are generally regarded as having the best commercial potential. Of the Heterocarpus species, H. ensifer has been found in only small quantities in Fiji, Vanuatu, Tonga (King, 1984) and New Caledonia (Intes, 1978) but in larger quantities in northern Pacific islands such as Hawaii (Struhsaker and Aasted, 1974), Guam (Wilder, 1977), and the Marianas. Individuals of this species caught in the southern Pacific Ocean appear smaller than those caught in the northern Pacific. H. sibogae, on the other hand, is more commonly found in the southern Pacific. H. gibbosus, which forms a large part of the catch in Fiji, has only occasionally been found in other southern Pacific islands and rarely in the northern Pacific. H. laevigatus is widely distributed in Pacific islands, and because of its large size and abundance is of greatest interest in fisheries development. Several other organisms have been commonly caught in traps set for deep-water shrimps. The more notable of these include snapper (Lutjanidae), particularly Ete/is and Pristipomoides species (in depths less than 300 m) and several species of eels at all depths. In Fiji, Nautilus pompilius has been taken (up to 12individuals per trap) in depths of less than 380 m (King, 1984). Another nautilus, N. mac- romphalus which is unique to New Caledonia, has been caught in deep-water traps (Intes, 1978). The deep-sea red , Geryon has been caught in depths greater than 600 m in both Fiji and New Caledonia. Amphipods and isopods, including a giant Bathynomus species, sometimes enter traps in large numbers and devour any bait not protected by plastic or small mesh containers. Distribution by Depth. - The distribution of deep-water shrimps is highly related to depth, with each particular species occupying different but overlapping depth ranges. As traps are set at increasing depths and distances from the shore or reef, catches of each species reach a maximum before decreasing and being replaced by another, and often larger, species in deeper water. In general, shallow water catches (from less than 400 m) consist of small shrimps such as Parapandalus serratifrons and Plesionika longirostris. Medium-sized Het- 194 BULLETIN OF MARINE SCIENCE, VOL. 41, NO.2, 1987

8 mm ROPE

Figure 1. The general arrangement of a trap fishing rig with a typical four-entrance box trap shown in the inset. The length of line between the buoys and the anchor is equivalent to the depth of water plus an excess of at least 25%.

erocarpus species predominate in catches deeper than 400 m. One of the largest species found, Heterocarpus laevigatus, is common in depths of more than 500 m. The general depth distribution of the eight more common species of deep- water shrimps is shown in Figure 2. It should be noted, however, that the depth of maximum abundance of each species may vary between islands; the reported depth distributions of one of the most common species, Heterocarpus laevigatus, are compared in Figure 3. It is difficult to compare the relative catch rates of deep-water shrimps in different islands due to the effects of using different traps, baits and soak times in the various surveys. Mean catch rates at optimum depth ranges, in areas where surveys have been conducted using small traps (volume between 0.2 and 0.3 m3), vary from about 1 to 3 kg per trap per night. Larger traps, more recently preferred by commercial fishermen in Hawaii, are reported to catch at least five times more shrimps than small traps (Methot, 1984). The Fijian surveys have provided some information on the possible seasonality KING: DEEP· WATER SHRIMPS FROM PACIFIC ISLANDS 195

Table I. Caridean shrimps found in deep-watertrappingsurveys in Fiji (Fij), Vanuatu (Van), Western Samoa (Sam), Tonga (Ton), Marianas (Mar) and Hawaii (Haw). The trap abundance of each species is indicated as C (common), 0 (occasional), R (rare) or no letter (not recorded). From King (1984), Anon. (1984), and R. Moffitt (pers. comm. 1986)

Species Fij Van Sam Ton Mar Haw Parapandalus (=Plesionika*) serratlfrons C C C C Plesionika longirostris (=edwardsii*) C C C C C 0 Plesionika rostricrescentis R R Plesionika ensis 0 0 0 0 R R Plesionika ocellus R Plesionika sindoi (=P. ocellus*) R Plesionika martia 0 R 0 Periclimenes sp. R Eugonatus crassus R Heterocarpus ensifer 0 0 0 C C Heterocarpus sibogae C C C C R Heterocarpus gibbosus C 0 R R 0 Heterocarpus laevigatus C C C C C C Heterocarpus dorsalis 0 C 0 Heterocarpus lepidus 0 Heterocarpus longirostris C Heterocarpus tricarinatus R

*' Revised names (Chace, 1985). in catches of deep-water shrimps. Figure 4 shows the mean catch rates of three species, Heterocarpus sibogae, H. gibbosus and H. laevigatus, by depth during trapping surveys down the outer reef slope near Suva in Fiji (King, 1984). Analyses of variance in these data indicate that mean catch rates, from all depths combined, did not vary significantly between different survey times. In all species, there was significant interaction (P < 0.01) between the effects of time and depth. The depth distributions of these species may change seasonally; in H. gibbosus, at least, the data in Figure 4 suggest an annual cyclic migration up and down the sea-floor slope. Heterocarpus laevigatus, in Hawaiian waters, appears to migrate from depths of around 550 m to depths of 700 m during the egg-bearing season (Dailey and Ralston, in prep.2). Other cyclic migrations of shrimps may also affect their catchability. There is some evidence, for example, that Heterocarpus species make vertical migrations in the water column. Individuals have been found in the stomach contents of skipjack tuna (Katsuwonus pelamis) in Fiji (R. Stone, pers. comm. 1982) and in Hawaii, (R. Spencer, pers. comm. 1982). Several small « 10 mm c.L.) specimens of H. ensifer were found in samples caught in midwater trawls towed obliquely from about 110m to the surface in areas off Hawaii, where the depth was over 200 m (King, 1984). During 24-h sampling conducted at stations off Suva, H. sibogae and H. gibbosus were caught in greatly reduced numbers during the hours of surface daylight (King, 1983). Biological Aspects. - Unlike the majority of temperate-water, exploited carideans and contrary to earlier reports (Clarke, 1972; Wilder, 1977; King, 198Ia), more recent studies on tropical deep-water shrimps indicate no evidence of sex-reversal (King and Moffit, 1984). In all but the largest size-groups, sex ratios were ap- proximately 1:I. The eggs carried by deep-water pandalids are often brightly colored, and the intensity of color appears to depend on the developmental stage of the eggs. With increasing development, eggs become lighter in color with the formation of the 196 BULLETIN OF MARINE SCIENCE, VOL. 41, NO.2, 1987

Figure 2. The general distribution of the most commonly found deepwater shrimps at various depths (in meters) near tropical Pacific Islands. Island countries where surveys have been conducted are shown on the inset map. The vertical lines represent the approximate depth range of maximum abundance for each species. These depths differ to some degree in different areas and all species are not necessarily found in all of the islands mentioned in the text. Species illustrated, from shallow to deep are, (Ps) Parapandalus serratlfrons, (PI) Plesionika longirostris, (Pm) Plesionika martia and P. ensis. (He) Het- erocarpus ensifer. (Hg) Heterocarpus gibbosus. (Hs) Heterocarpus sibogae. (HI) Heterocarpus laevigatus and (Hd) Heterocarpus dorsalis. Sizes of each species generally increases with depth from a carapace length of up to 21 mm (about 120 mm total length) in P. serratifrons to up to 56 mm (about 217 mm total length) in H. laevigatus. transparent blastoderm and larval body. Finally, eggs become darker in color with the development of larval pigmentation and eye-spots. Within species, egg sizes tend to increase with increasing development (King and Butler, 1985). Between species, eggs are generally larger in the larger, deeper-water species (Table 2).

------A ------c B .. , .. / ..... -....,. ---- .' \. ------0 ----1------E ~ ------F --~------H G

o 200 400 600 800 1000 DEPTH 1m) Figure 3. The depth distribution (light line) and depth of greatest abundance (heavy line) of Hetero- carpus laevigatus in (A) Hawaii; Gooding, 1984, (B) Marianas; Moffitt and Polovina, in prep.3, (C) Guam; Wilder, 1977, (D) New Caledonia; Intes, 1978, (E) Fiji; King, 1984, (F) Vanuatu; King, 1981a, (G) Western Samoa; King, 1980, and, (H) Tonga; King, 1981 b. Broken lines indicate the limits ofthe respective surveys. KING: DEEP-WATER SHRIMPS FROM PACIFIC ISLANDS 197

FEB. 80 MAY JUN AUG OCT DEC JAN. 81 MAR MAY SEP H.sibogae 300

j ] ] 500 ] ] ] ]

700

Hgibbosus 300

E ] i 500 ] Ii:: lJJ 0 j ! 700

H.laevigatus 300

500 ] ] ] ] 700

o [}5 1'0 CATCH per TRAP Figure 4. Mean catch-rates (kg per trap) by depth of Heterocarpus sibogae (top), H. gibbosus (middle), and H. laevigatus (bottom) caught off Laucala Bay, Fiji (King, 1984). Brackets enclose maximum catch rates which are not significantly different (P > 0.05) from adjacent catch rates (Student-Newman- Keuls test).

Within species, there appears to be a linear relationship between brood size and female weight. This relationship is likely to be a result of the area available for egg attachment as well as the female's capacity for obtaining, storing and con- verting resources for egg production. Mean brood sizes for several deep-water species caught off Fiji are given in Table 2; data for the shallow-water reef species, Saron marmoratus, are included in this table for comparison. The mean size of reaching sexual maturity in females (defined as the size at which 50% of the female population is ovigerous) is given for several pandalid shrimps in Table 3. At the time of reaching sexual maturity, females appear to acquire deeper abdominal pleura, particularly the second pleuron, which may protect developing eggs carried on the pleopods. The incidence of ovigerous females appears to vary with the time of year. In 198 BULLETIN OF MARINE SCIENCE, VOL. 41, NO.2, 1987

Table 2. Depths of maximum abundance, mean lengths of well-developed eggs, mean brood sizes (actual numbers), relative brood sizes (No. of eggs per "standard" 15 g individual), annual reproductive efforts (biomass of eggs per unit mass offemale) and lifespan reproductive efforts (annual reproductive effort multiplied by reproductive lifespan) for seven species of caridean shrimp (from King and Butler, 1985). Sample sizes ranged from II in S. marmoratus to 156 in H. sibogae. Standard deviations are in parentheses

Relative Depth Egg length brood Annual Lifespan Species (m) (mm) Brood size size RE RE Saron marmoratus 0 0.30 (0.03) 1,282 (132) 12,839 0.22 (0.08) 0.22* Parapanda/us serratifrons 200 0.62 (0.05) 1,344 (183) 8,685 0.10 (0.04) 0.10* P/esionika /ongirostris 275 0.70 (0.04) 13,721 (847) 16,875 0.13 (0.04) 0.13* Heterocarpus ensifer 425 0.69 (0.04) 2,864 (145) 6,509 0.08 (0.03) 0.08* Heterocarpus gibbosus 475 0.78 (0.06) 28,312 (1,195) 16,278 0.20 (0.03) 0.42 Heterocarpus sibogae 500 0.68 (0.04) 23,292 (2,266) 20,110 0.15 (0.03) 0.45 Heterocarpus /aevigatus 625 0.87 (0.09) 34,461 (1,352) 8,900 0.15 (0.03) 0.59 * Semel parous species-annual reproductive effort values used.

Fiji, over 50% offemale Heterocarpus laevigatus were carrying eggs in April 1979, June and July 1980 and May 1981 (King, 1983). In Hawaii, the corresponding time period for the same species was October to January (Dailey and Ralston, in prep.2). The spawning season of H. laevigatus, therefore, appears to be in the winter season of each hemisphere. The analysis oflength-frequency data is the only practical method available for estimating growth in deep-water shrimps. Length-frequency distributions may be arranged sequentially over an extended time period, allowing the progression of peaks or modes to be followed. This method has been used to estimate the growth of several species in Fiji (King, 1983; King and Butler, 1985) and of Heterocarpus laevigatus in Hawaii (Dailey and Ralston, in prep.2) and the Marianas (Moffitt and Polovina, in prep.3). Growth curves, for four species (sexes combined) are shown in Figure 5. Data from the Marianas indicate that male H. laevigatus grow more quickly than females but reach a smaller ultimate size. Growth data for Heterocarpus laevigatus suggest that females become sexually mature at approximately 4 to 5 years (40.5 mm carapace length), with the age of the largest size groups in the samples over 8 years. Note that it is not possible to age deep-water shrimps precisely, and the estimated ages given above assume that growth throughout life follows the von Bertalanffy curves shown in Figure 5; this may not be so, particularly if larval growth differs markedly from that of adults.

Table 3. Depths of maximum abundance, carapace lengths (mm) at female maturity, total and reproductive lifespans (years)-from Figure 3, and maximum carapace lengths of seven species of caridean shrimps (King, 1983)

Length at Total Reproductive Depth maturity lifespan lifespan Lmax Species (m) (mm) (years) (years) (mm)

Saron marmoratus 0 8.5 nd nd 12 Parapanda/us serratifrons 200 11.5 nd nd 21 P/esionika /ongirostris 275 25.0 3.5 0.6 28 Heterocarpus ensifer 425 20.5 nd nd 29 Heterocarpus gibbosus 475 28.5 5.2 2.1 42 Heterocarpus sibogae 500 33.5 6.1 3.0 41 Heterocarpus /aevigatus 625 40.5 8.5 3.9 56 nd = no data. KlNG: DEEP·WATER SHRIMPS FROM PACIFIC ISLANDS ]99

60

HI

40

:r l- e> Z ~ 20 «ll.: a: <5

2 4 6 8 TIMElyr)

Figure 5. Growth curves for (PI) Plesionika longirostris. (Hs) Heterocarpus sibogae. (Hg) H. gibbosus and (HI) H. laevigatus. The curves have been terminated at 90% of the theoretical maximum length. The heavy portion of each curve shows the estimated reproductive lifespan of each species.

Mortality rates have been estimated for H. laevigatus by examining the decrease in relative numbers with age in the samples. In Fiji instantaneous natural mortality rates were estimated to be 0.66 or about 48% per year (King, 1986); in the Marianas an estimate of 0.75 or about 53% was obtained (Moffitt and Polovina, in prep.3).

DISCUSSION Caridean shrimps have been caught near many Pacific Islands in numbers sufficient to create interest in their commercial exploitation. Most of the surveys reported in this paper were initiated from a potential resource point of view. The biological parameters, particularly the combination of slow growth rates with high natural mortality rates, suggest that the potential resource would be highly sus- ceptible to overfishing. The main constraints to the development of fisheries on deep-water shrimps, however, are the high costs of fishing in deep-water and the securing of suitable markets (King, 1986). Apart from commercial considerations, the various aspects of deep-water shrimp biology and ecology have been of theoretical interest in the study of the evolution of life-history patterns at various depths (King and Butler, 1985). Several life-history variables, including maximum size, longevity, reproductive lifespan, egg and brood size, and reproductive effort have been estimated and are summarized in Figure 6. Regression lines are fitted through all data except those for lifespan reproductive effort, which are obviously non-linear. Maximum length, mean reproductive lifespan and mean egg length of each species are all signifi- cantly, positively correlated with depth of distribution. Relative brood size and annual reproductive effort are poorly correlated with depth, and slopes are not significantly different from zero. A notable aspect of these results is that annual reproductive effort does not appear to differ with the depth distribution of the different species. It is possible 200 BULLETIN OF MARINE SCIENCE, VOL. 41, NO.2, 1987

Sm Ps Pl He HgHs Hl-SPECIES I 60 •

40 (AI Lmax slope=0.07** Imm) r~= 0.91** 20

4 (BI 3 REPROD. slope= 0.01 •. LIFESPAN 2 r~= 0.98** (yr)

0

slope=0.0008* leI r~= 0.77* EGG 0'6 LENGTH O'~ lmm) 0·4 •

(OJ REL 20~ • slope=1.04 NS • • r2<0.1 NS BROOD 10 • SIZE • • Ix10-3) • 1 lEI I 0·2 • I I ANNUAL O'j • • • slope= -0.0005 NS RE. • 2 0·1 • • r = 0.05 NS 0'6 • IF) LIFETIME R.E. 0'4

0·2 • •

I 0 200 400 600 -DEPTH(m)

Figure 6. Life-history variables in caridean shrimp. (A) maximum carapace length; (B) reproductive lifespan; (C) egg length; (D) relative brood size; (E) annual reproductive effort; (F) lifetime reproductive effort. Reproductive efforts are measured as biomass of eggs per unit biomass of female. Species labelled as in Fig. I, and values are plotted at mean depths of maximum abundance for each species. KING: DEEP-WATER SHRIMPS FROM PACIFIC ISLANDS 201 that the similar annual reproductive efforts found in the shrimp species in this study may represent a ceiling on parental investment. Above this ceiling, adult mortality may be prohibitively high. That is, there is a survival cost in the form of an increased risk of parental mortality (due to the high energetic cost of re- production) with any increase in reproductive output. Several models have been proposed to explain how different environmental factors affect the selection of life history patterns. Many predictions from these models appear contradictory. The application of concepts such as r-K selection (Pianka, 1970; Steams, 1976) to the present, and many similar, data appears fruitless, since nothing can be known of the intensity of competition in the different environments. Any discussion of the present results in terms ofr-correlates (early maturity, small body size and semelparity in shallower water shrimps) and K-cor- relates (late maturity, large body size and iteroparity in deeper water shrimps) adds nothing more than a qualitative description of these findings. Some authors have argued that competition may be of less importance than density-independent mortality in influencing life history patterns. Reproductive effort may, therefore, differ in ways which conflicts with competition-based theory (Christiansen and Fenchel, 1979). Although age-specific mortality in deep-water shrimps is not known, there is evidence that adult predation is higher on the shallower outer reef slopes. Extensive trapping suggests that the number of species and total biomass of potential, and confirmed, fish predators decreases with increasing depth and distance from the reef(King, 1983). It is suggested that predation is the main cause of adult mortality in deep-water shrimps and the chances of survival increase with increasing depth (King and Butler, 1985). Species of shrimp in deeper water appear to have greater longevity, to reach a larger maximum size, and to be iteroparous. These factors allow deeper water species to have a higher total lifetime reproductive effort. Figure 6F shows that, except for the reef species Saran marmoratus, lifetime reproductive effort increases with the increasing depth distribution of the different species. There is some evidence (King and Butler, 1985), that deepwater pandalids produce planktonic larvae which migrate to surface waters. If the planktotrophic larvae move through the height of the water column, the probability of larval survival is likely to decrease with increasing distance traversed and with increasing time spent in the water body. Larvae in the water body for an extended period are, presumably, subject to food fluctuations and predation for longer periods before settling. In addition, the larvae of deeper water species are subjected to a greater range of water temperatures and other environmental factors during their vertical migration; the differences between bottom and surface temperatures were approximately 5°C at 600 m, increasing to about 17°Cat 200 m. Such variations may reduce the probability of larval survival. The decrease in larval survival probabilities with increasing depth may be offset by the production of larger eggs by shrimp species in deeper water; larger eggs generally produce larger, or more advanced, larvae which have greater survival prospects than small ones. Figure 6C shows that egg size significantly increases (P < 0.05) with the increasing depth distribution of the different species. In summary, it is proposed that the adults of species in deeper water are exposed to less predation, which allows a long lifespan and iteroparity. This results in an increase in total lifetime reproductive investment with the increasing depth dis- tribution of the different species. It is also suggested that the probability of larval survival decreases with in- creasing depth. These risks are partly reduced by the production of larger eggs 202 BULLETIN OF MARINE SCIENCE, VOL. 41, NO.2, 1987

and partly by the greater lifetime reproductive effort in deeper water species. From this, it appears that the production of a relatively large number of eggs is of less importance than the production of larger eggs with a greater chance of survival.

ACKNOWLEDGMENTS

Field work in Pacific Islands was carried out while the author was at the Institute of Marine Resources of the University of the South Pacific, Fiji. Thanks are due to the Director of this Institute, Dr. U. Raj. Participation in the Second International Symposium on Indo-Pacific Marine Biology was made possible by partial funding from the U.S. National Science Foundation (through Dr. M. L. Reaka) and the U.S. Western Society of Naturalists (through Dr. A. Wenner and Dr. D. Montgomery). Dr. Wenner and A. Carver kindly provided comments on the manuscript.

LITERATURE CITED

Anon. 1977. Report of a survey by the French R.V. "Coriolis" in Western Samoan waters. Photostat copy. ---. 1984. Status of fisheries and assessment of development and management needs for selected crustacean species in the western Pacific region. Western Pac. Reg. Fish. Mgt. Council, Honolulu, Hawaii. 61 pp. Brown, I. W. and M. G. King. 1979. Deep-water shrimp trapping project: report on Phase I. Fish Div. Fiji. Tech. Rep. I. 30 pp. Chace, F. A. 1985. The caridean shrimp (Crustacea: ) of the" Albatross" Philippine ex- pedition 1907-1910, Part 3: Families Thalassocaridae and Pandalidae. Smiths. Contrib. Zool. 411 pp. Christiansen, F. B. and T. B. Fenchel. 1979. Evolution of marine invertebrate reproductive patterns. Theor. Popul. BioI. 16: 267-282. Clarke, T. A. 1972. Exploration for deep benthic fish and crustacean resources in Hawaii. Hawaii Inst. Mar. BioI. Tech. Rept. 29: 18 pp. De Reviers, X., M. Theriault, R. Grandperrin and J. Crossland. 1982. Essais de peche de crevettes profondes aux casiers a Vanuatu. ORSTOM, Notes et Documents d'Oceanographie 4: 42 pp. Gooding, R. M. 1984. Trapping surveys for the deepwater caridean shrimps, Heterocarpus laevigatus and H. ensifer, in the northwestern Hawaiian Islands. Mar. Fish. Rev. 46(2): 18-26. Intes, A. 1978. Peche profonde aux casiers en Nouvelle-Caledonie et iles adjacentes. Essais Prelim- inaires ORSTOM. (Noumea) Rapports Scient. Tech. 2: 10 pp. King, M. G. 1980. A trapping survey for deepwater shrimp (Decapoda: Natantia) in Western Samoa. Report of the Institute of Marine Resources, University of the South Pacific, Fiji. 26 pp. ---, 1981a. Deepwater shrimp resources in Vanuatu: a preliminary survey off Port Vila. Mar. Fish. Rev. 43(12): 10-17. ---. 1981 b. The deepwater shrimps of Tonga: a preliminary survey near Nuku'alofa. Report of the Institute of Marine Resources, The University of the South Pacific, Fiji. 29 pp. ---. 1981c. Increasing interest in the tropical Pacific's deepwater shrimps. Australian Fisheries. 40(6): 33-41. ---. 1982. Report on the South Pacific Commission deepwater shrimp assessment consultancy in Papua New Guinea. South Pacific Commission, Noumea, New Caledonia. 25 pp. ---. 1983. The ecology of deepwater caridean shrimps (Crustacea: Decapoda: ) near tropical Pacific islands with particular emphasis on the relationship of life-history patterns to depth. PhD thesis, University of the South Pacific. 258 pp. --. 1984. The species and depth distribution of deepwater caridean shrimps (Decapoda, Caridea) near some southwest Pacific Islands. Crustaceana. 47: 174-191. ---. 1986. The fishery resources of Pacific Island countries. Part 1. F.A.O. Fisheries Technical Paper, 272.1: 45 pp. --- and A. 1. Butler. 1985. Relationship oflife-history patterns to depth in deep-water caridean shrimps (Crustacea: Natantia). Mar. BioI. 86: 129-138. -- and R. B. Moffitt. 1984. The sexuality of tropical deepwater shrimps (Decapoda; Pandalidae). J. Crustacean BioI. 4: 567-571. Methot, R. 1984. Analysis of the potential yield of Hawaiian deepwater shrimp. Unpublished report to the National Marine Fisheries Service, Hawaii. 5 pp. Moffitt, R. B. 1983. Heterocarpus longirostris MacGilchrist from the northern Marianas Islands. Fish. Bull. 81: 434-436. Oishi, F. 1983. Shrimp industry development project. Division of Aquatic Resources. Department of Land and Natural Resources, State of Hawaii. 22 pp. Pianka, E. R. 1970. On r- and K-selection. Am. Nat. 104: 592-597. KING: DEEP-WATER SHRIMPS FROM PACIFIC ISLANDS 203

Steams, S. C. 1976. Life-history tactics: a review of the ideas. Q. Rev. BioI. 51: 3-47. Struhsaker, P. and D. C. Aasted. 1974. Deepwater shrimp trapping in the Hawaiian Islands. Mar. Fish. Rev. 36(10): 24-30. --- and H. O. Yoshida. 1975. Exploratory shrimp trawling in the Hawaiian Islands. Mar. Fish. Rev. 37(12): 13-21. Wilder, M. J. 1977. Biological aspects and fisheries potential oftwo deepwater shrimps, Heterocarpus ensifer and Heterocarpus laevigatus in waters surrounding Guam. Master of Science Thesis, Univ. of Guam. 79 pp.

DATE ACCEPTED: December I, 1986.

ADDRESS: School of Fisheries, Australian Maritime College. P.O. Box 986. Launceston, Tasmania 7250. Australia.