a 5 Beton hos Se ane te is ei herthaeniins setoheen enena-». “nr reps~rap ee tet ae eee ms “ Rp eI as gen paneniMautcaes 5 — = é o Ne : g Nos. 15, 16 A oe) April 30, 1953 ATOLL RESEARCH BULLETIN 15. The Insect Life of Arno by R. L. UsINGER and TRA LA RIVERS 16. The Land Vegetation of Arno Atoll, Marshall Islands by WILLIAM H. HATHEWAY Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences—National Research Council Washington, D. C., U.S.A. | j } 4 5 | i ATOLL RESEARCH BULLETIN 15. The Insect Life of Arno by R. L. Usinger and Ira La Rivers 16, The Land Vegetation of Arno Atoll, Marshall Islands by William H, Hatheway Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences-—-National Research Council Washington, D.C. April 30, 1953 = b > * ei oo H i | evi, of wal Bae: ATOLL RESEARCH BULLETIN No. 15 The inecce Life of Arno by von ed ik Usinger and Ira La Rivers — ‘Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences—-National Research Council ‘Washington, D.C. Rorik 30) 1953 ACKNOWLEDGEMENT It is a pleasure to commend the far-sighted policy of the Of- fice of Naval Research, with its emphasis on basic research, as a result of which a grant has made possible the continuation of the Coral Atoll Program of the Pena Science Board, It is of interest to ete, idictoniedily: that much of the fundamental information on atolls of the Pacific was gathered by the U. S. Navy's South Pacific Exploring Expedition, over one hundred years ago, under.the command of Captain Charles Wilkes. The continuing nature of such scientific interest by the Navy is shown by the support for the Pacific Science Board's research programs, CIMA, SIM, and ICCP, during the past five years. The Coral Atoll ooo en is a part of SIM. The preparation and issuance of this Bulletin is assisted by funds from Contract No. N7-onr-291, Task Order IV. ee we we ee ee The sole responsibility for all statements made by authors of papers in the Atoll Research Bulletin rests with them, and do not necessarily represent the views of the Pacific Science Board or the editors of the Bulletin. a ne me ee ee es er oe Editorial Staff F. R. Fosberg, editor M. H. Sachet, assistant editor Correspondence concerning the Atoll Research Bulletin should be addressed to the above c/o Pacific Science Board National Research Council io 2101 Constitution Avenue, N. W. Washington 25, D.C., U.S.A. THE INSECT LIFE OF ARNO 1/ ue R. L. Usinger and Ira La Rivers On Arno, as elsewhere, insects are the oeet aonendie of land animals. This well-knowm phenomenon is a fumetion of their small size and their high reproductive potential. The total number of species of land anthropods on Arno is estimated at 500. This may be compared with 96 species reported from Canton Island (a somewhat drier coral atoll) and nearly 1000 recorded from piles (a much rereaisis high island). The inseets of Arno have to be abundant in order to maintain themselves, because most of the higher ter- restrial animals live on them. ‘hus, insects are the first link in many of the food chains, such forms as aphids, plant bugs, caterpillars and, in fresh water, midge larvae acting as primary converters of green plants into animal food. | Of no less importance than the "grazers" are the ever-present ants and cockroaches which act es s-avengers and the wood-boring beetles and termites which, with the help of bacteria and fungi, restore logs to hu- mus. Other katabolic agents include the flies and beeties which play a cominant role in the decay of fallen fruits and others which perform a Similar function in dung and carrion. Not ail insects are occupied in such vital jobs of a constructive or destructive mde Analysis of the fauna reveals that a considerable pro- portion of the insects in any community are busily engaged in eating each 1/ This report covers field work carried out in 1950 as part of the Coral Atoll Project of the National Research Council's Pacific Science Board. These studies were aided by a contract between the Office of Naval Re- search, Department of the Navy and the National Academy of Sciences (NR 160 175). i ee other. One-fourth of the insects of Amo fall directly in this category—- as predators or parasites end the Surkonhene Ee td enon higher when care- ful rearing reveals the true numbers of entomophagous parasites. Then there are the small ectoparasites—-the lice and mites, and one louse fly-- which infest birds and rats. Another vital though inconspicucus furiction of imissete on meee mainland areas is pollination. On Arno probably three-fifths of the higher plants are insect pollinated. Detailed studies on this are lecking but the single Arno bee, a Tek onttine Snes cs of Megachile, eaters pollen on its ab- « dominal hairs. It nests in the dead twigs of Scaevola and visite various flowers including Triumfetta, Scaevola and Wedelia. The above-mentioned pereeies Gee pacac to any community of plants and animals and are cuite independent of man. Tor this reason, the insects, with but few exceptions, are not much different on the uninhabited islets than in the villeges. Furthermore, many of the insects undoubtedly ar- ee before the first natives and tee Cerieece a state of equilibrium which may not have changed significantly with’ the advent of man. “There are a few important exceptions to this generalization—~such as the house fly, the house mosquito, domestic cockroaches, and lice. The role of these and other insects associated with man mals pe eeeneeed a detail later. | | | Animal Communities and Habitats It is a well-knom fact that plants end animals are discontinuously distributed over the earth's surface. Nevertheless it comes as a surprise to find that five Square miles of practically level lend of similar origin and relatively uniform composition should present such a variety of distinct Piast plant and animal associations. In spite of a superficial similarity there are at least four distinct terrestrial snide and numerous distinct but interrelated strata within each. in addition there is the — habi- tat including tide pools and the open ocean which is inhabited by marine water striders. : Marine Insects. The virtual absence of the largest group of animals, the insects, from the largest habitat in the world, the ocean, has been the subject of much speculation (Buxton, 1926). It has been pointed out that | insects of saline lakes can tolerate greater concentrations of salt than are found in the ape pnd insects a torrential streams and waterfalls are subjected to concitions which are comparable to wave action. That, ian, are the limiting factors that prevent insects a a group fron colonising the sea? The answer is to oe found in the meet of eens in the feck that, they evolved a special breathing mechanism (tracheae) wes was edmir- ably suited to terrestrial life and only sengndar thy. adaptable to Life in the a BET aa But, perhaps the ost qritigal. factor at alt is the depth Of the ocean, for a water is uninhabitable | even fe fresh water forms, ahe pro- fundal bottom habitat of our deepest lakes being beyand ee reach of insects. Then there is the fact that the ocean is | Piready ocoupied ty another group of Arthropods, the Crustacea. This group, which includes the ae crabs and their relatives, is so unten tie: adapted to all of the Ren eg niches in the sea that would-be invaders such as insects cannot compete with them. me Le One group of insects has overcome the oo difeicul ties by oveupying the surface of the ocean. The marine water apes attracted the recon of early voyagers who found small, gray, long—legged Peet ane coursing about on the high seas. These marine forms differ from their fresh-water relatives Bi el in that they have permanently lost their wings, there Udit no need to util- ise this means of dispersal in the ocean which, unlike fresh-water ponds, is in no immediate danger of drying up. | Halobates micans lives in the open ocean and in the deep parts of the lagoon at Arno. Nymphs and adults are blown up on the beach by strong winds. This happened on the beach of the lagoon at Ine Village one afternoon and evening. The bugs were awkward on land and sought shelter from the wind in the smallest depressions or in footprints on the beach. Here they doubtless fell easy prey to a golden plover which patrolled that section of beach and also to the ghost crabs which glided swiftly cver the sand after dark. i | In protected coves such as are found on the ocean side of Deere ena oe in Sie apsen ene RIE bes oP He AIT See eS are found. ‘One of these, Halobatés mariannarum, differs Ae STs eae from its relatives in the Marianas and Carolines, ae the fact that it differs at all suggests that it ig endemic ‘to the reefs of ne et ehalta! Baa has been isolated for a long time.” Nymphs add Adulte! of Gnye docat Halepeece swim in groups of hundreds or thousands in shallow abeae A search was made for the plankton which must: serve as their food, but oye eye pee. sumably, they feed on plankters which rise to the Moraes oe a dina a fact that Halobates are not eaten by fish suggests that the hs ee nal scent gland repels the ever-present marine predators or, eaters: aes a the bugs distasteful to the fish. “. Even more local, in small embayments, is Halovelia, a marine water strider whose body is less than one-twelfth of an inch in Tene, fre male is even smaller and rides on the back of the female to es a ces - sexes will not be separated by wind or tide. -The Strand. in: the daytime: the beaches of Arno .are clean. and: support very few insects. Robber flies flash hete:and there, capturing mostly. flies, but. this is:about the extent of the. activityiover, the uninterrupted stretches of sand. . ae a Ee aL: - o> At night, Armo: beaches take on a changed aspect. The Ghost.crabs. (Oc- cypcde ceratophthalma), common enough during daylight hours, literally. swarm over the sands after.dark. -They dart everywhere. At first, the in- sect’ fauna seems to be poorly represented, ‘few species: coming to lights,: and then usually only one or two individuais of any given kind. “As a ‘con- speuiane the discovery, one night, of a small, white cricket mar¢hing dabovt the beaches in great numbers ceme as somewhat of a surprise. They were . particulariv abundant at low tides when the greatest: expanse of beach was exposed, and seemed to be coming: from everywhere. They ranged in size from the quarter-inch aduits down to berely visible early instar nymphs, all. marching sedately. by the coleman iamp on the sand. They were rather weakly positively phototronic, and after the Light had -been on the sand"for ‘some ‘time, it could be noted that most of them were gradually orienting themselves toward it.. They leap vigorously at the slightest motion near .them:., Like the .storied. Mormon cricket of the intermontane West (Anabrus simplex)., these little beach crickets fell quickly upon an injured fellow and disposed: of it with the rapidity and facility denoting the confirmed omivore, which they were assumed.to ‘he, although no’other. data were obtained‘on their plant feeding habits, if any. Similarly, they were quite. agile. in catching the rether numerous small flying ants attracted to the light, and which were meandering about-the. sand. Bux,: judging solely by their numbers, it is unlikely that. predation was their usual way of making a living,. for there didn't seem to be a Jarge enough supply of available victims. for the horde-- by contrast, the lush plant growth would seemingly have had to supply the bulk of the food such a population of crickets would require. Here, cer- tainly, is one animal which can be reckoned an important factor in the | conversion of plant materials and smaller animals into food available for other animals greater in size, and conseauently occupying a more edvanced spot in the food-cycle schewe. One of the larger animals patiently en- gaged in such conversion was the above-mentioned ghost crab; these were ecbserved preying upon the small white cricket in large numbers. A rough count was made of the number of successrul catches out of the number of attempts that any one ghost crab made, with the resulting approximation of Leimgl.@. fr, additional characteristic of Arno night beaches is the great number of crab spiders hunting on the rocks and vegetation bordering the sand. Directing a flashlight at eye-level usually struck up myriads of diamond- bright points of light--they could be seen flashing on all sides as the light swept elong. Close inspection showed ‘them to be the eyeshine of hunt- ing spiders. Occasionally such a point of Light turned out to be a small ground gecko, but with some experience, it soon became neta to distin- guish these latter by virtue of their slightly larger size and more yellow- ish reflection. Except for the beaches the eats rocky, consisting of sharp, pitted elevated reefs. liere and in the strictly intertidal zone insects are vir- tually absent, contrary to the situation that exists in the Halophila tide pools of Samoa or the beaches of Hawaii. The one human element that enters into the picture in the intertidal: zone is the presence of human feces--a hygienic habit involving use of the tidal waters for sewage disposal which attracts numerous flies that would otherwise be entirely absent from the zone. Mangroves are a special association in a few places such as the lagoon at Bikariej. No insects were observed to inhabit the mangrove “1rees except marine water stricers. The Inner Beach is occupied by various plants such as Scaevola, which occurs singly or in pure stands and provides blossoms for the Marshallese head leis, Messerschmidia, Guettarda, Pipturus, Cordia, aay mostly on raised ledges, Pemphis. Each of these characteristic strand plants harbors a distinct fame, a ieaf mining agromyzid fly making conspicuous serpentine burrows in the thick Scaevola leaves, caterpillars of the Achaea moth on Cordia, the striking day-flying moth, Utetheisa, with red and black spots on a white background, on Messerschmidia and the beautiful blue Hypolimas butterfly on Sida. A particularly rich habitat in this zone is provided by the native custom of burning Scaevola and Messerschmidia thickets to provide space for coconuts. The dying or dead limbs with leaves browned but still attached, were attractive to two Scolytid beetles, one round-headed borer, and several small predaceous rove beetles. Also in the beetle galleries were two pre- daceous bugs of the family Anthocoridae and one curious little pieteeanruil the imposing scientific name, Ceratocombus. The only seemingly barren tree in this association was Pemphis. Re- peated search failed to reveal a single insect on this unusually hardwooded tree. Scaevola, onthe other hand, supports a considerable population in- cluding a small endemic bug, Campylomma, each island group in the Pacific having a species with distinctive spots on its hind legs. The Open Woodland. As one approaches Arno it is the erect trees that catch the eye. in profile, the tall trees give tne island its place on the | 1M A horizon. Coconuts and the screw-pine (Pandanus) comprise mast of the forest ang are the exclusive elements that rise above the strand’ scrub in the nar- row parts of Arno. The coconut-pandanus forest, whether geeded naturally or plented by man, is an open type with individual trees rather widely separ- ated, allowing sunlight to reach the ground during most of the day. Much has been written of the insect vests of the coconut in other parts of the tropics but Arno is singularly free fom such troubles. The Micro- nesian coconut scale, ‘an oval, reddish scale nearly one-fourth inch in dia- meter, is the most conspicuous pest on fronds and green husks. The boots, skirts and trunks, on the other hand, seen with insect life. Here one finds a whole community gathered largely for the protecticn afforded by this ubiquitous tree. Several kinds of cockroaches are found as well as small bark lice, Gaiwipan ante ltcte: Most of these fall easy prey to the skinks which frequent the same habitat. On Arno Pandanus is not so attractive to insects eee ne coconue per- “haps because it gees net provide as Bond pro see nioae Only at the tases of the fronds is a microhabitat found. Heré rain water accumulates, sometimes in amounts Bubeiesent to sustain dregonfly naiads and, in some parts of the tropics, mosquito larvae. On Arno during our stay moisture was sufficient only for earwigs paaledet poets: The fruit of the Pandanus is enormous and, when ti bal hign in sugar content--this providing a delicacy for the natives. Decaying Pandanus a fruits on the ground harbor an association of vinegar flies (Drosophila), fruit beetles (Nitidulidae)> and numerous predators (Staphylinidae, Anthocoridee, Ceratocombidae). In addition there is a small but very active, white, predaceous mite. gm The ground cover in the open woodl.and consists of various low plants such as Vigna marina, Wedelia, Fleurya and, in open areas, the sedge Fimbristylis, and associated grasses such as Lepturus. . Sometimes this growth is choked by the parasitic vine, Cassytha. This ground cover is rich in insects, the undersides of Vigna leaves hone heavily infested with the Micronesian red spider mite (Tetranychus), Vigna and Fleurya being spotted and withered by the hopping plant bug, Halticus, the grasses acting as hosts for the cosmopolitan green plant bug, Trigonotylus, and Fimbristylis harboring , eee a community of true bugs including the herbivorous sedge bugs, Ninus and Orthotyllelus, and Nysius picipes, Pachybrachius nigriceps and Pachybrachius pacificus, and the vredaceous Nabis capsiformis. That the lone predator is not sufficient to keep this association in check is attested by the enormous numbers of individuals present wherever the plants occur... The Nysius is of particular interest because it is found only on the islands remote from air bases in the Marshalls whereas Nysius pulchellus, from Guam, .. . is found exclusively on Majuro and Kwajalein. Presumably Nysius picipes is an old resident in the Marshalls, yet it is indistinguishable from typical Wake Island specimens. The Canopy Woodland. The richest zone on Arno occurs in the wider.. parts of the isiets where humus has accumulated and breadfruit grows. Here one finds a dense growth of trees 50 and 75 feet tall with foliage so dense that the ground is shaded most of the time. Before the time of the present inhabitants, taro pits were dug in such places, thus further augmenting the accumulation of humus and the maintenance of moist conditions. Here one finds the buttressed trunks of breadfruit with epiphytic bird's-nest ferns, a moss, Calymperes espe. a thick, low undergrowth of Polypodium fern, and decaying logs with Polyporus and Schizophyllum fungi. oo Here in decaying logs are wire worti larvae, stag beetle edults and larvae, termites, ants, Machilis and a large red Collembolan, wood roaches, earwigs, predaceous rove beetles, and other invertebrates such as scorpions, psevdoscorpions, centipedes, earthworms, isopods, and millipeces, as well as land snails of several species. On the fungi growing on the logs are fungus beetles (Ciidae and Fndomychidae), fungus gnats (Mycetophilidae), and various predators that take advantage of such situations. “Fallen logs are one of the most profitable foraging placés for insectivorous skinks and birds. Tree holes are frécuent in this part of the forest sna gece stands in them for considerable periods. In this microhabitat the endemic Marshallese mosquito, Aedes marshallensis, breeds. The eggs are laid at the water's edge and the on larvae feed on the very special type of organic matter at the bottom of such small bodies of water. That the mosquitoes thrive in spite of their rather restricted habitat is attested by the large numbers of individuals, yet these mosquitoes are of such retiring habits that they are seldom troublesome and only caused serious trouble during World War II when, on neighboring islands, dengue fever broke out. This mosquito, being an efficient vector, had to be controlled, but only in the vicinity of military installations because it has a short flight range. Taro pits provide one of the most intriguing problems for the biologist. Surrounded by ‘tusnetaee Evaploat vegetation, these bodies of shallow water should, by all mainland standards, be teeming with life. Yet they are al- most devoid of insects. A fresh water snail is always common but this is 2li that meets the eye at first glance. Mayflies, caddis flies, and all of the numerous water beetles and bugs are absent. Closer inspection reveals blood worms in the mud, from which a few midges emerge, the remainder hg forming the basic, and in some instances the only, diet of the dragonfly naiades. These fierce predators with tiie extensible labia and powerful jet propulsion are quite the fiercest animals in the water and at least on Arno, are overadapted to their particular niche. The only other arthropod of any consequence in fresh waters is a reddish shrimp which was found occasionally. The extremely depauperate fresh-water fauna is not surprising when one considers the nature of a coral atoll. Arno has a relatively heavy and uniform rainfall and therefore fares better than most atolls. Typically, fresh water is scant and seasonal on coral atolls and fresh-water life must be hazardous indeed. Even on Arno, the only endemic of aquatic habit is the tree-hole mosquito. The dragonflies are all strong-flying immigrants and the other forms may be accounted for by accidental transport of rela- tively recent date. Adult dragonflies are swift fliers and are seen more in the open forest where they hover or dart about capturing gnats and other small insects on the wing. The slower and more fragile damselflies were seen only in the vicinity of the taro pits on the widest part of Arno Island. Damselfly naiades were not seen but dragonfly naiades were common in wells, cisterns and taro pits where they undoubtedly eke out a precarious existence in the face of occasional hurricanes and the ever-present danger of drought. Herbivorous insects are not conspicuous in the canopy forests but nevertheless are abundant, a large red corizid bug occurring on Ailophyilus and the bird's-nest fern (Asplenium nidus), a small but spry spider mite on breadfruit leaves and a host of small saprophytic and parasitic insects in fallen rotting breadfruit, as in the rotting Pandanus fruit. The pro- ductivity of one rotten breadfruit in terms of individual Drosophila flies is enormous. i, Insects and Native Culture Certain insects such as lice and bed bugs havé accompaniea man in all of his wanderings. The island peoples were no exception to this end it is certain that numerous species even less intimated: associated with man than the above stowed away in native canoes. In this way, no doubt, the cosmo- politan house fly and its chief enemy, the predaceous Hermetia fly, came to Arno and also’ the Australasian and Oriental cockroaches, various ants, the rotten-fruit insects and the scale insects, mites and aphids which are attached to the coconut, breadfruit, Pandaanus, banana, taro, arrowroot, papaya and lime trees. Also to be classed as a silent hitchhiker is the notorious dengue~yellow fever mosquito whose larvae have’ traveled over the entire tropical world in drinking water. On Arno, as elsewhere, this species is so closely associated with man that it is found only in and around houses and breeds in the cisterns and other drinking water containers, leaving the tree holes of the forest for the closely related Marshallese mosquito. Termites attack man-made structures as well as native woods on Arno as elsewhere in the tropics. Four species, each belonging to a separate genus end each harboring distinctive Protozoa in its gut, were found on Arno. Without the Protozoa termites have been show to be incapable of utilizing the dry cellulose of dead wood as nutritive material. This is only a partial iist of the camp followers of man in the Pacific. Nevertheless, it is a formidable array and causes one to wonder how it is that life is so pleasant on Arno. The fact is that none of these pests, ex- cept possibly the house fly, is sufficieritly numerous to present a real problem in the daily life of the natives. In fact er is singularly free from such notorious scourges of the tropics as the coconut moth, the coconut oe beetle, the Oriental fruit fly, locusts, etc. Even the ubiquitous copra beetle was not seen. Agricultural Pests. At the time of our study there were no agricultural pests of any Significance on Arno. This is a situation almost without para- llel elsewhere in the world. _ The reasons for this happy state of affairs are conte, First, the number of crops is small a includes only those species best adapted to the conditions on a coral atoll. On Majuro, where beans and sweet potatoes are grom, ee plant bugs (Halticus) and mealy- bugs are very numerous and ds sed wi Perhaps of greater importance, these is no intensive agriculture on Amo, the staple coconuts, Pandanus, peoeaernit and taro growing with little ose af Sieawon and, except in ithe case oF bheadtratt. existing in amounts in excess of the actual needs of the apne on. The breadfruit being seasonal and especially favored as food, is more fully utilized, though no apparent, attempt, is made to plant new trees. Only bananas are planted in, dense stands and these are protected, ‘as young Souts. from pigs. mae 0 the tow level of insect pests must be svbri buted to the particular ecological tte onset that exists on Arno between herbivores and predators. Normality, herbivores far atEnAnEee. pre- aetors in an Enel community. On Arno, however, protected habitats are few and predaceous insects and vertebrates (mainly skinks) are exceedingly numerous isee Report by Marshell). Thus ee have a situation in which the population level (eqi1 brian) cf nacets of agricultural crops is below | the threshold of economic damage. Betoparasites. This are was 5 investigated by Dr. Marshall. _A list of the eeaeaeee. both external and a a of pigs, chickens, dogs and cats appear in his Report. Both the house rat and the Polynesian rat were Lao found on Arno and both were infested with lice (Hoplopl.eura) and two species of parasitic mites (Leelaps). Herons were infested with parasitic flies (Ormithoica) and mites, noddies with bird lice, Austromenopon, hipnoboscid flies, and mites. Insects in relation to health and sanitation. On Arno there are numer-~ — ous insects and other invertebrates that affect the health of man either directly or indirectly. Scorpions and centipedes are much feared by the natives. Although we cannot say from experience, the bite of a ent neoe or the sting of a scorpion are usually of the same order of severity as a bad bee sting. Certainly scorpions and coe ee are sufficiently common so that their poisonous effects are experienced rather frequentiy. It seems safe to assume, however, that the natives'! fears are exaggerated, on the basis of experiences elsewhere in the a ee | The most serious arthropod bite in Biot parts of the world is that of the black widow spider. This spider has een seis eet from nearby islands in the Marshalls -tut we found none on Arno. Likewise we found no ticks, chiggers or fleas. 3 i si 7 | A rather unusual pest of man in the Marshalls is the so-called "Bao in jekaro" (BIRD IN THE JEKARO), a yen, yellow edoneerd beetie with strong cantharitic sdeuga ins ae) Serie’ it Heyes esto to night lights on Arno and was found on several occasions flying An clouds weather. Every- where we heard the same story: that it was strongly attracted to the "jekaro", the sap from the inflorescence cf the coconut, while the latter was being tapped. If a "Bao" falls into the sap and is eaten or swallowed, it produces kidney disturbances, which are well-knowm ifretieens of can- tharidin poisoning. If crushed on the skin, it would produce a blister—- as we discovered from personal experience. Crushing an individual yg experimentally on the forearm produced a wheal about the size of a quarter and a blister which broke after a couple of days. The e¢toparasites of man on Arno include lice and, presumably, the tropical bed bug, although bed bugs were found ohly on Majuro. The appar- ent absence of bed bugs on Arno is remarkable, since the natives say that they were found during the Japanese occupation. Whatever their true status may be they were certainly not conspicuous on Arno during the period of our study. Lice, on the other hand, are rather common, especially head lice, and the natives engage in a louse-catching ceremony. There is no evidence of louse—borne typhus fever and this disease would not be expected in such a comparatively cleanly people in the tropics where excessively heavy, tight clothing is unknown. Three mosquitoes’ are commonly encountered on oceanic islands in the Pacific, the common house or night—biting mosquito of the tropics, Culex quinguefasciatus, and the two day biters, Aedes aegypti end Aedes marshal- lensis. The first of these, although found on most islands of the Pacific, was not found by us on Arno, much to our surprise and pleasure, since it was not necessary to use our bed nets. However, night-biting mosquitoes have been reported by others. The pantropic dengue-yellow fever mosquito, Aedes aegypti, is common, but only in the immediate vicinity of human habi- tations. As is the case elsewhere in the world, this species, orieinal ly a tree-hole breeder in Africa has adapted itself to the human race even more intimately than most of our domesticated animals. The eggs are laid at the water's edge in the concrete cisterns and other drinking water receptacles on Arno. The wary larvae are bottom feeders, ie swinging with S-shaped movements from surface to bottom to provide for their constant need for air on the one hand and food on the other. Adults have silver-banded legs and vegnshaped merce on ae thorax. We found = commonly in HORST» eee our laboratory, wibeane they lurked ‘be- neath tables and in dark corners biting ee ee around the wrists and = | ; Bani Sees Hes been reported from the Marshalls, yellow fever has not. Neither disease is pndeunie so the only danger lies in the introduction of the virus in a human carrier, an acai event since the carrier would have to egies from elastin where ine was infected to cee: within ae in- cubation period of the disease on or pusen! ie a widen mosquito on or atout the time of the onset of the disease. In a of this intricate ieee) dees bolts out on various ietande in the Morchanale and Gilberts during World War II and the possibility of reintroduction should not be overlooked in establishing quarantine penile ey | ‘The native Marshailese mosquito is eine a gay biter and a close rela- tive of the dengue-yellow fever mosquito (they. toth, belong to the same subgenus, Secenaaes This is a tree—-hoie rede auers larvae which — and act 13 ike aezypti larvae supaetatallys occurring in water-—filied holes in breadfruit arid other Use and in GogonuE haif shells in the canopy forest of Arno. Adults swarm about when one stands still in the ahede of a Cie: tree but they are not vicious biters. ia ee belongs to the so-cailed Nz Ne fe ee eee ee pseudoseutellaris being : vector of non-periodic Piiepindie in Fiji and Samca. We ‘in anak of fileriasis on Arno ee chronic form known as elephantiasis is con- spicuous in some parts of the Pacific). / wee Two groups of insects, the cockroaches and house flies, are of impor- tance in general sanitation on Arno. The cockroaches, although exceedingly abundant, are less conspicuous because of their nocturnal habits. The night-feeding skinks and geckoes eat large numbers of these but the popula- tion is still sufficient so that a certain amount of contaminetion undoubtedly takes place. Far more important hygienically, in fact the chief agent other than man himself in transmitting enteric disease organisms, is undoubtedly the house fly. This cosmopolitan pest occurs in enormous numbers on Arno. It is somewhat smaller and darker than mainland forms but is nonetheless abun- dant and bothersome. But for this pest Arno might well qualify aie the "paradise" so frequently extolled in tourist posters. The combination of unscreened privies, defecation in exposed intertidal areas, unscreened houses and unprotected food, constitutes an Nideal" situation for the particular kind of contamination to which the house fly is adapted. There is little doubt that the incidence of amoebiasis as given in Marshall's Report is attributable in part, at least, to contamination of this type. Aside from the obviously favorable breeding places, the large numbers of house flies may be attributable to the comparatively few natural enemies. Skinks cannot catch flies as readily as Sia rte insects. There is an ° efficient larval‘predator, the larva of the large Hermetia fly. This largest fly on the island was founda at all times in and Seeae privies but unfortunately does not reduce the population of house flies effectively. Interrelations of Insects with Plants and Animals of Arno The "food cycle" (Elton) or "food web" (Allee, Emerson, Park, Park, Schmidt) becomes exceedingly complex, even for a small association. However, eailyen an attempt has been made to show in the accompanying Tables (I-V) some of the food chains as they exist on Arno. Unfortunetely, quantitative data on insect populations are not available so calculations-of biomass are not possible. However, it may he accepted as a basic premise and this was con- firmed in a general way by field observations, that the herbivores of small size were most numerous in individuals. The role of each species listed in the food chains is indicated by appropriate letters as follows: herbivores {H), predators (Pred), parasites (Par), Saprophytes (Sapr), and — scavengers (Scav). In general, species are listed in order, starting with groups of herbivores, then proceeding to scavengers and then to successive groups of predators, members of each group preying on members of preceding groups. The departure from this scheme in Table V (The Human Community) is self-explanatory. It is noteworthy that most of the food chains end with 1Meas" This is a partial explenation for Marshall's statement that Viverds (the blue- tailed skink in particular) are the the most numerous land vertebrates on A the atoll. This does not mean that the skinks have no enemies ‘for they . were eaten on occasion by cuckoos, herons, chickens arid even by 4 land | crab, Geograpsus. However, none of these predators is Benerally distributed and none depends primarily on lizards for food. Therefore the lizards are relatively free from predators. They do have numerous endoparasites, in- cluding stomach nematodes and rectal flagellates. The significance of these data from the viewpoint of atoll ecology iat that the majority of the food chains are independent of man. ihe Only in the immediate vicinity of villages is the picture significantly ee and a such places man enters the basic atoli food chains mainly via antes and pigs. Aside from this, man has moved into the atoll biota and ae his ‘at niche. -18- Origin, Dispersal, and Speciation The origin of the fauna and flora of Arno is lost in antiquity but it is possible to piece together in a general way the probable history on the basis of present day observations and certain logical inferences. First, if one of the presently accepted theories of formation of coral atolls is true (see Wells' Report), then the insects and all other animals and plants must have reached Arno fortuitously, that is, by accidental trans- port. Various types of guandaads dispersal are knowm to occur (see Zimmerman, Ins. of Hawaii), among which are wind and convection currents of the air, birds, rafts of debris carried by ocean currents, native canoes and, more recently, trading boats and airplanes, Probably all but the last of these have contributed to the present biota of Arno. While it is difficult definitely to attribute the presence of a particular species: to a certain type of dispersal, here again we have certain facts and certain logical inferences to go on. Certainly the insects’ associated with man, such as lice, cockroaches, and the house fly, came with man and probably with early man in native: .: canoes. The dengue—yellow fever mosquito was’probably a more recent. im- migrant, since its dispersal elsewhere in the Pacific. (Hawaii, Usinger). is knom to be associated with the voyages of exploration and profiteering during the 19th century. Insects which are attached to breadfruit, coconut, Pandanus, arrow- root, banana, and lime (e.g., scale insects, aphids, etc.) surely came with these crops when they were introduced by the natives. Many insects of small size have been collected at high elevations (a mite at 30,000 feet, for example) in a study of several years! duration walle es (Glick). Especially during the infrequent but violent hurricanes, small forms, whether winged or not, must have been blow to Arno. Stronger fly- ing insects such as the dragonflies and the monarch butterfly, we know are capable of long distance flights, aided no doubt by the wind. Thus Anax has flown to ships at sea and readily establishes itself on remote islands, wherever fresh water is to be found. Similarly, the monarch butterfly, though absent when its milkweed host is wanting, arrives soon after the ac- cidental or deliberate eee ieoneae oe Asclepias curassavica. Though rafts may be necessary to account for endemic skinks, many of the animals which present apparently insuperable preblems of dispersal (e.g:, land snails, wingless weevils, etc.) could have been transported on the feet of migratory birds (Zimmerman). Beyond @ certain point this type of reasoning degenerates into idle speculation aud is of little or no use to anyone. However, it is instruc- tive to carry the analysis as far as possible in order to piece together the separate parts that contributed to the present biota of Arno. To a certain extent the same sequence of events can be traced when a new islet is built up from reef to sand bar and thence, through accumulation of vege- tation, to a cemented, and subsequently decomposed substrate capable of supporting life as we find it. on the densely covered islets. Literally all stages in this succession can be seen today. But this is not the end. Acultural processes are proceeding at an accelerated rate on Amo as else- where in the world (see Mason's Report) and the biota is bound to change too. In the long run this may be regarded as inevitable and therefore of no particular concern. However, the depauperate nature of the biota (e.g., the small number of different kinds. of plants and animals as compared with an equivalent mainland area) and of even greater importance, the disharmonic _20- nature of the biota (e.g., the complete absence of certain types with con- sequent gaps or ecologic vacua) makes Arno soup eae See ie to dis- miptivs influences. Furthermore, the limited economy of the people, based as it is on so few crops, could be brought to a standstill by failure of one of the staples. Hence, it is important to look to the future and try to foresee immediate and long-term dangers which might ensue from the dis- persal and establishment of noxious piants and animals from nearby or farther distant islands and mainiand areas. The most imminent cases of this kind are on the neighboring atoll of Majuro. _Due to wartime activity and air transport, Majuro, and to an even greater extent, Kwajalein and Iniwetok, now support several immigrant in- sects which, if introduced to Arno via the native canoes or the copra boat, would prove to be detrimental to the natives. Foremost among these poten- tial immigrants are the infamous migratory locust and a second, slightly smaller grasshopper, the black widow spider, the night mosquito, an Odynerus wasp and a tick. In addition, bed bugs, fleas, the coconut Brontispa beetle and the copra beetle are either absent from Arno or are so scarce that we did not find them. Obviously an effort should be made to prevent or delay their establishment or increase on Arno. Serious pests from more distant islands of Micronesia include the giant African snail, the Rhinoceros beetle (Oryctes), and a malaria mosquito, any of which could reach the Marshalls from Guam or Saipan unless quarantine regulations are strictly enforced. epee We Lage @ “was “ai odd eon oun q ha vides Lal siuoit aye oe ena ste ‘oe tel ‘ean to. sheetei ge sal Uj east Pia Table I Strand Community A. Open Beach Stratum - Mostly Nocturnal Sand crickets - H. Sand fleas - S&S. House flies - (human feces). Sapr. (daylight). Robber flies - Pred. (Daylight). Marine water striders {blown onto beach) - Pred. Crab spiders ~ Pred. Ghost crabs — Pred. Rock geckoes - Pred. oe re ee Plover — Pred. Curlew - Pred. Tattler — Pred. Marine Insect "Community" A. Surface Stratum Pelagic plankters - Pred. H. (Large - Open ocean ) Marine water striders (Large - Tidepool species) Pred. (Small - Protected coves ) Seo s) i. eerie culm eeauen. | f yen ' 2 y, > 2 A ‘ ‘ F a ‘ oe nd . eigued) - : i ALT ~ seta! 1), wreaks netoni - Slesi) Rania Table II Inner Beach Community A. Foliage Stratum Leaf-mining fly - H. Scaevola Plant bugs - H. Scaevola Caterpillars - H. Cordia, Messerschmidia, Sida Leaf-—cutting bee - H,. Aphis - H. Ants - Scav. eee ee ee Ladybird beetles - Pred. Green iacewings - Pred. -Syrphid flies - Pred. Spiders - Pred. Geckoes — Pred. Skinks - Pred. is Bark and Dead Wood Stratum Bark beetles -— Wood borers Round-headed borers — Wood borers Rove beetles -— Pred. Predaceous bugs —- Pred. Spiders - Pred. Geckoes —- Pred. - Skinks ~ Pred. C. Ground Stratum Sand crickets - H. Crab spiders - Pred. Ghost crabs — Pred. Rock geckoes - Pred. Golden plover —- Pred. Heron — Pred. -23- a R . 1 ; v aj 7 d 1 7 3 ‘Pee, ~ 7 Va , i SS 5 i $y v Ue ei ag i j Y Fi ores onnmntgiis 7% ” J, ‘ N 7 ae ae ~<" oa 4 4 a 4 \ Q : } E ie A iO \ " ] % 7 : ; 4 ae oe i BAY \ = at A ze j ih i ¥ To. a wi y ‘] a > ; arenes bene oe ae Table III Open Woodland Community A. Coconut - Pandanus Stratum Coconut scale - H. Sugar-cane weevil - H. Bark lice -— Scav. Ants - Scav. Cockroaches - Scav. Earwigs — Pred. Luteva — Pred. Spiders — Pred. Geckoes — Pred. Night Skinks - Pred. Day and Night B. Ground Cover - Vigna, Wedelia, Fleurya, Fimbristylis, Lepturus, Cassytha Red spider mite - H. Vigna leaves Hopping plant bug - H. Vigna and Fleurya Green grass bug - H. Grasses Leafhoppers — H. Sedge bugs - H. Fimbristylis False chinch bugs - H. Fimbristylis Thrips - H. Bees - H. Caterpillars -_H. Mealybugs - H. Aphids - H. Ants - Scav. Damsel bugs, Nabis - Pred. Spiders — Pred. Thrips ~ Pred. Stink bugs - Pred. Ladybird beetles - Pred. Green katydids - Pred. Geckoes — Pred. Night Skinks -— Pred. Day and Night Table III (continved) C. Fallen Green Coconuts Rats - H. Coconut crabs - H. Mosquitoes - Larvae-detritus. . D. Rotten Pandanus Fruit Stratum Vinegar flies - Sapr. Fruit beetles - Sapr. Rove beetles -— Pred. Spiders — Pred. Predaceous bugs - Pred. Predaceous mites - Pred. Geckoes — Prec. Skinks - Pred. E. Soil Stratum Earthworms - Sapr. Miliipedes - Sapr. Sowbugs - Sapr. take Garden centipnedes - Sean H. Silverfish - Scav. Ants - Scav. Hermit crabs — Pred. Scav. Rock Geckoes - Pred. Night Ground skinks —- Pred. AGE Table IV Canopy Woodland Community A. Breadfruit Stratum Corizid bug - H. Bird's nest fern Leafhoppers — H. Spider mite - H. Breadfruit leaves Mosquito larvae - Detritus. Tree holes Ants - Scav. Spiders - Pred. Geckoes - Pred. Night _Skinks - Pred. Day and Night B. Underbrush Stratum - Alophyllus, Pipturus, Terminalia, Pisonia, Cordia Aphids - H. Mealybugs - H. Leafhoppers - H. Corizid bugs - H. Fruit flies - H. Ants -— Scav. Spiders — Pred. Ladybird beetles - Pred. Syrphid flies - Pred. Geckoes - Pred. Night Skinks - Pred. Day and Night C. Fallen Log Stratum Fungous flies — Sapr. Fungous beetles —- Sapr. Weevils — Woodborers Termites — Wood feeders Wireworms - Sapr. False wireworms -— Sapr. Stag beetles - Sapr. Ants - Scav. —~26— Table IV (continued) G.. (Ceonts) Earwigs — Prec. Spiders - Pred. Pseudoscorpions —- Pred. Scorpions -— Pred. D. Rotten Breadfruit Stratum Vinegar flies - Sapr.. Fruit oeetles - Sapr. Rove beetles - Pred. Spiders - Pred. Predaceous bugs —- Pred. Predaceous mites — Prec. Geckoes -— Pred. Skinks - Pred, BE. Soil Stratum Earthworms - Sapr. Millipedes - Sapr. Sowbugs - Sapr. Garden centipnedes - Sapr. H. Silverfish - Scav. Springtaiis - Sapr. énts - Scav. Rock geckoes — Pred. Ground skinks —- Pred. F, Taro Pit Aquatic Stratum Midge larvae - Detritus Shrimp - Pred. Snails - H. Dragonfly naiades - Pred. - Sys Table V Human Community Man - Omnivorous. Eats fish, coconut, breadfruit, taro, pandanus, bananas, papaya, and various imported foods. Ectoparasites - Lice, bedbugs Endoparasites —- Amoebae, hookworms, etc. Domesticated Animals Dogs - scavenger - copra — garbage Pigs — scavenger - soli insects, roots Chickens - Omnivorous. Copra, ground insects, skinks Cats - Omnivorous. rats Household Vermin Rats - Eaten by cats Ectoparasites - mites and lice Cockroaches — Evania parasites Termites Mosquitoes - cisterns - Dragonfly naiad predators Houseflies — Hermetia parasites — Privies House gecko eats all household insects ap ae agltcke tbat eS Bled ase hin ce eh Ki" aap uf ATOLL RESEARCH BULLEZTIN No. 16 The Land Vegetation of Arno Atoll, Marshall Islands by William H, Hatheway Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences——National Research Council Washington, D.C. Boral (30), 1953 ‘ i ei “ft . » ‘i f 5 - z m 4 " , < ; 7 aa ‘< wy 7 im, i ; . : ad ay \ Us i 2 j Gd 7 Jy 7 x j . 2 a tama: 2) aes ; : | 2 z " Dh : bi " ‘ Mee ot AMINE: BOVE IO? OLTIOM Se tt y e iN ) a: Pomiod sloredins Lace eiett—-esomne tol 2 hae hae | reoaahiteal } a i + ' | 4 i} The Land Vegetation of Arno Atoll, Marshall Islands SCIENTIFIC INVESTIGATIONS IN MICRONESIA Pacific Science Board National Research Council Willism H,. Hatheway Harvard University Canor.dze; Massachusetts January 1953 ACKNOWLEDGMENTS This report presents the results of botanical field work done in the summer of 1952 under the Coral Atoll Project of the Pacific Seience Board of the National Research Council. The work was supported by funds granted to the National Academy of Sciences by the Office of Naval Research, The field work was carried out with the active assistance of the Ua S. Navy Department, the Military Air Transport Service, and the Trust Territory of the Pacific Islands, The writer wishes to express his thanks for help received from lir, Harold J». Coolidge, Mrs. Lenore Smith, and Miss Ernestine Akerse Mr. D. W. Gilfillan, District Administrator of the Trust Territory at Majuro, and Mr. J. E. Tobin, staff anthropologist, took especial in~ terest in this study and assisted greatly. The heip of Lt. Cmdre Ce Ke Brust, who arranged to have put at the dieses of the author certain re~ sources of the Ue S. Navy, is gratefully acknowledged, Special thanks are due the people of Arno, especially Kilmej le; Felix Je, Tobu, Lajibili, Jemelik, Jiblok J.s, Abijai, Lakarim, Jona, Nemot Le, Jajin Fe, Bingel, Jakalu, Saburo, Jon, Lauken, Overton, and Likotak.e At Majuro Raymond DeBrum, Konto Se, and Dwight Heine helped me in many waySe I am indebted to the following for stimulating discussions and ad- vice: F. Rs Fosberg, He St. John, E. Le. Stone, Jr., J» Ws Wells, D. C. Cox, RoW. Hiatt, De Strasburg, Le Mason, P. EH. Cloud, H. Uyehara, E. H. Moul, Mary Murai, Me Ss Doty, N. De. Newell, J. Pe Morrison, G. D. Sherman, I. Es Lane, and EF. H. Bryan, Jr. F. Re Fosberg has made some adjustments in the botanical names of the plant species, CONTENTS Introduction - ------}j-j--= Res al) rae Vac ear | (ee) nes Ge ek fen Coconut type - ----=---+-+-+-+--+8-- ee tee ha et Breadfruit type -------+++-+-+-+-24e48-4-+-6- --- Scrub Forest --- -------+-+-+-+-2++-+-+-+-+-+-6 Boulder Ramparts ------+-e2+-e+-<8-++-+--+-+e5+--- Stony Interiors « -«<«<<=62 sw =2-<8-=-s=8 =e -2 == 5 Sandy Interiors =—=-<« =| <= 8 es eee ee ee ee ee Lagoon Shores and Dynes -=------=-- ------ Sehgal Mangrove Swamps ----+---+-+--++-+-+-++#-+s4--- Fresh-water Swamps -----.-------+--+--+-+-+-+--4 Secondary Forests of the Interior of Arno Island - - - - Vegetational Change ---+---+-+--+-+-2+-+-+--- Phosphorus Relations - ~~ ----++<<-2<.+-+-4--:5 fasvorical Resime -——<= = 5 = el—_ 44 2 ae aoe ee Mpa teresa Cl ee i ea a a a ah ae a ea Ld INTRODUCTICN In 1952 78 percent of the land surface of Amo atoll was wider coconut and breadfruit forests, both of which types were maintained as such only by periodic clearing of umiwanted wild plants. With the ees of recently deereree weeds, virtually every species of plant on Arno found a specific use -- in fearon construction of houses, canoes, and im- plements, or in the weaving and coloring af meee, to mention only a fewe The possibility that even wild native species had been planted on some sites re ee could never be overlooked. Obviously, Arno's vegetation was inseparably Linked with its people. To the student who regards vegeta- tion as simply the result of the interaction of flora and habdtat, inhabited atolls such as Aro can be extremely confusing. Qn the other hand, a study of the interrelations of the eee the plants of an atoll is ie esting not only in the iaetahe into the life of the atoll people which it provides, but also in its illumination of certain important processes of vegetational and edaphic change which might otherwise not be understood, About 1000 persons were resident at Arno in 1952, most of whom lived in villages along the lagoon shores of Pie widen islands. Islands less than 600 feet wide and stony lands were fae the most part wuinhabited, Dwellings were concentrated near the lagoon shores of the wider islands, for these sites, besides providing easy access to the lagoon, were con= veniently near those areas best suited for Laptein food plants -- the breadfruit and the yaraj (cyrtosperma). In the ieee of the wider islands the salinity of the ground water was sufficiently low to enn their growth. The villages, however, were not always conveniently near coconut groves. Copra-making parties often had to travel considerable distances to their work, At Arno right to the use of the land was held not by individuals but instead by exteneed ie eroups (bwij), the heads of which were termed alabs. The land pete (wato) was commonly a narrow strip of ground one to five acres in pea extending from lagoon to ocean. In many cases a bwij held rights on more than one wate. From its watos. . the bwij parvected mien of its food and wood and all of its copra. The latter produce constituted the chief source of cash income to the bwij, from which its members purchased imported foods, clothing, soap, kero- sene, and aihee trade cee Land could not be bought or sold; the ane to the use of the land was passed in strict hereditary fashion et one det: dha Pei webs the next. Aside from a very small cr by the ee eee, ere was no public land on Amo, Right to ihe wee GF each piece of land on the atoll was claimed by at least ag ‘eee renee. it is obvious that the vegetation found on any piece of land was dependent upon the previous use of that _ land by its eee Intensity of land use depended not only on the suitability of the land holding for the growth of economic plants, but also on such factors as its distance from the places of residence of the bwij, the number of acres per bwij member, and very probably on the industriousness of the members of the bwij, especially the alab. My assistant informed me that the bwij of which his father was alab, was wmable properly to take care of all the land which historically it had been alotted. "Some have too much land. Some don't have enough." At Majuro atoll brush had grown up on land occupied by the United States Government during the war and a 2 ue since abandoned. The former Marshallese enmeesl had by no means relin- quished claims to this land, but were not certain of their legal rights to its use. Land disputes were all too common at Aro, Since the author was advised not to become involved in sisi seine he did not Snvestipate their relationships +o the intensity of use of the land. In 1952 the flora or eno eet ner eeed or abone 125 species of vascular plants (ef Anderson, 1951). Of these nearly half had been intro- duced within ae past 100 years, roughly J Suctaltsic wcalaxe as cultivated plants and 19 accidentally ag weeds. That is to say, the Amo flora of ~ | 1850 consisted of pore een ee species of higher plants. of these, however, probably about 2h, had been introduced by the Marshallese as ceeneiie ei onee or weeds. Although a native flora of approximately bid species may seem very small, that of Amo is relatively rich compared to the floras of other “low islands situated at greater distances from the rich source of seeds of Malaysia and Melanesia, or eee ene tse rainfall. Canton Island in the Phoenix group, relatively near the rich flora of Fiji, has a native flora of only 16 species, but it is nearly a desert. Palmyra Island, lying to the east of Arno, receives ‘about’ as Puen rainfall as the latter atoll, eet is much more remote from major sources of seed. Its flora consists of only 16 species. | Ps Although the flora of Arno is larger than that of many atolls, it is small and relatively wninteresting compared to the rich floras of high islands like Fiji, Hawaii, Tahiti, or even Ponape. Not a single vascular plant species is restricted £0 Arno atoll, ay of the native wild species are wide-ranging plants, most occurring on the eneee of islands in both the Pacific and Indian Oceans, and at. least 7 -- 16 percent of the native flora -- occur naturally in the tropics of both Rereeneres Except for a A Pe: ya single species, Pisonia grandis, all of the native plants of Arno possess seeds or fruits which are distributed by wind or ocean currents. The fruits of the Pisonia are glutinous and adhere to the feathers of the wide-ranging seabirds which roost and nest in its branches. It is not surprising, then, that this tree occurs on nearly every high and low island in the Pacific and Indian Oceans which is sufficiently moist to permit its growth, as do, indeed, a large proportion of the other native species of the Arno flora. The climate of Amo is probably not very different from that of nearby Majuro, where annual rainfall is of the order of 120 inches, The winter is marked by stronger, more constant northeast trades and by somewhat less rain than the summer's Mean ssaiaaal temperature is Slee F.3 mean monthly tenperatures vary only within 2° of the yearly average. The diumal variation is about 10°; nights are pleasantly cool. | Arno atoll contained 133 islands and islets in 1952, which formed e square miles (3200 acres) of ‘ary lend, Almost all these islands had the same ee Their seaward sides consisted Of mide of ae _ cobbles, and stones throw up on he reef flats by storm waves. Sand ac- cumulated on the lagoon aides of these Pega ee dimes formed a where a sandy bench eed te prevailing ripaethiena’s badge at aes stony ridge of the Beean side and the dune ridge ot the ieeeie shores, was A often an interior depression. The islands tended to increase in width by additions of axial on the lagoon side, and ges material on the ocean shores. The ground water of an atoll islet is characterized by che presence of a Ghyben-Herzberg lens of fresh water. The depth of this lens and iis salinity depend on the distance to the beaches and the permeability of the soils, Stony land is much more permeable to the salt water of the ocean o~ | than is dune sand, and, with the:rise and fall of the tides appreciable mixing of fresh and salt water, occurs. Thus the stony land of the ocean side of islands is underlain .by much more saline ground water than lagoon= shore dunes. The ground water, in the interiors of the wider islands is remarkably free from salte Cox (1951) found only 8 ppm. chlorides in a _ well at Ine about. 500 feet from the lagoon; in an interior. pool on Tatu Island only 200 feet from the ocean I found only 15 ppm. 150 feet from this pool the growmd water in the boulder rampart had a chloride concentrase’:. tion of 7100 ppm, about 70 times as great. Fosberg (199) and Cox re- marked that ground-water salinity apparently acts as an effective control on the iedeetian of the breadfruit, an economic plant of aboriginal introduction. Variations in salinity affect the distributions of species of the wild native flora as well; the tolerance to salt of species of _ the Rhizophoraceae (mangroves), Pemphis acidula, and others is, of course, well lnow.. . As Stone (1950) and Wells (1951) pointed. out, typhoons occur about ) times per century at Arno,.and have left their mark on the islands in many wayse The washing away of entire islets is a spectacular consequence of typhoons, but windthrow and partial inmdation have affected soil and vegetation. In certain local areas on Kilange Island, for example, nearly every large breadfruit tree was blown down in the typhoon of 1918 and large quantities. of soil were thus sctnnbatiet In other places considerable amounts of sand and rock had been suddenly dumped on low land, or the surface layers a been washed out, Such areas were often characterized by low organic and nutrient content of the surface horizons, The more superstitious believed such land to be inhabited by malignant spirits which hinder the growth of plants, Stories of hallucinations caused by these demons were common. vee Relatively minor changes were constantly occurring along beaches. Between Ine and Matolen, for examole, the shoreline was eroding in places, and plants which formerly grew along low dune ridges were stranded on the beach. A Pemphis tree with several pebbles embedded in its wood 3.2 feet above the present level of the ground was discovered growing on a sandy beach 15 feet from an eroding due containing similar pebbles near its surface, The tree had in effect moved 15 feet toward the lagoon during its lifetime, Near Lukw8j coconut trees, which had previously toppled onto the beach as the dune upon which they were growing was undercut, were being covered with sand as the beach again built up. Entire trwmks were buried save for the erect terminal parts, about 5 feet tall. The leaves were nearly twice as long, and some trees were bearing heavily. The vegetation of Arno atoll is here classified into 9 types (Table I). These are merely arbitrary groupings of agerepates of plants into wits con- venient for description, analysis, and mapping, This, however, is not to deny that interrelations may be found between plants erowing near one another, The vegetational map has been drawm from 19-1945 aerial photo- graphs taken by the United States Navy, supplemented by ground observations and interpretation of oblique photographs taken by the Navy in 1952, In general, the vegetational types listed in Table I are only those which could be distinguished clearly on the aerial photographs, Thus "scrub forest" is mapped as a single mit, but is divided into h subtypes in the descrip- tive text. Bowmdaries between the types are, of course, no less arbitrary than the types themselves. Even on the growmd it is not always possible to classify with certainty as "coconut type" or "scrub forest" a stand of small, wild native trees through which coconut trees. projected. Bread- fruit groves were usually characterized by a rank undergrowth of native -~ 6 mm trees and vines. extensive "breadfruit groves", "Secondary forest", on Aro island contiguous with usually contained numerous breadfruit trees, Obviously a line separating "secondary forest" from "vreadfruit grove" must be based to some degree on a subjective estimate of inten- sity of use, illustrating the fact that on Amo vegetational study was inseparable from cultural anthropology. TABLE I Type Number of stands Acres Average size of stand a. a on ee Y veawi Ciceie a Coconut ellie OeeliOy . ” 1553) Loo etnat al 137 NEA OS 7 1525 Laora i: 65.5 6505 Mellal 9 Belg) | 3 oh Breadfruit 53 Zl feels, alla 542 Scrub. Forest 13 EOpaas 60a a. deen | Te Saline Flat . 4S BOLD rd ye D9 ilangrove Swamp et. 2bg8 10; ‘a ara Fresh=water Swamp. Bu: 6.6 202 Secondary Forest . 8 Mie D3 Fol Total 372 3205.3 8.6 COCONUT: TYPE: =~ Primary species: Cocos nucifera Height of trees: 0 to 50 feet, exceptionally to 80 fect Density: about 95 trees ver puoiiich exceptionally « as dow as Ho or as high as 1,0 trees cee acres = Ti Pereens 69.38 66.37 2.0); 0.97 8.6), 17267 0.92 0,8) On21 23h 100,00 69 percent of the land area of Amo atoll was mder coconut forest in 1952, for copra formed the main article of export of the atoll. Copra production was feasible on practically the entire land surface of the atoll, saline flats and swamps and certain parts of the interiors of wider islands forming the only significant exceptions to this rule. The wide interiors of islands, however, were devoted in large measure to food production for local consumption, and the exceptionally rich phosphatic soils of Takleb and Namwi had been given over to some degree to such special crops as Kapok, Papaya, and Banana. Remote, narrow, or stony lands, areas in which the ownership of the land was in dispute, or lands the workers of which were less energetic than others, in some cases supported stands of wild native trees « The coconut thrived over an extremely wide range of environmental conditions. Tt was productive over a 1,000-fold range of ground-water salinity and over at least a 50«fold range of concentration of available phosphorus. Temas found on all habitats from freshewater swamps to dry, windswept dunes, on organic mucks, fine or coarse-textured sands, or among fragments of coral rock. Except on those few areas in which it e was not productive, the coconut was utterly worthless as an indicator of environmental differences, "bh ‘ta The plants which grew wmder the coconuts, needee: were not always the same in the different habitats on which the coconut was planted. Consequently, it is convenient to recognize three environmental subtypes within the coconut type, corresponding to the 3 major physiographic divisions of an atoll islet. Secondary species: Pemphis acidula, Scaevola frutescens, ~ Ba | Hernandi.a sonora, Calophyllum inophyllum, Barringtonia asiatica, Suriana maritima, ‘Pandanus tectorius, Sophora tomentosa, Canavalia microcarpa, [pomoea tuba Ground layer: Lepturus repens, Fimbristylis atollensis, Thuarea involuta, Cassytha filiformis, Triumfetta procumbens, Vigna marina, Tacca leontopetaloides, Polypodium scolopendria, Euphorbia chamissonis, Canavalia sericea The coconut trees usually overhung the lagoon, sometimes toppling onto the lagoon beach where the shoreline was eroding. Along the lagoon shores of the leeward islands Pemphis acidula, a much-branched low tree with exceedingly heavy, hard wood, was often abundant on exposed beach- rock. In or near villages Hernandia sonora and Calophyllum inophyllum.- were observed, the latter commonly being used for rain catchment. (On dunes Lepturus, Fimbristylis, and Thuarea formed a low, dense cover under the coconut trees». 2. Low Interiors . Secondary species: Morinda citrifolia, Pandanus tectorius, Saskia and other trees of the scrub forest type. Ground layer: Lepturus repens, Fimbristylis atollensis, Thuarea involuta, Wedelia biflora, Cassytha filiformis, Triumfetta procumbens, Vigna marina, Polypodium scolovandria, Euphorbia hetero- phylla, Tacca leontopetaloides, Asplenium nidus, Nephroiepis sp». Boerhavia diffusa. Fimbristylis and the grasses grew taller here, and Pandanusy sprouts of liorinda, Guettarda, and other trees of the scrub forest formed jungles in some poorly tended groves. Epiphytic mosses, Nephrolepis, and Aspléniun appeared on stumps, fallen logs, and rocks. The velocity of the wind was much reduced and the habitat seemed moister. Boulder ridges and stony land. Secondary species: Trees of the scrub forest type Ground layer: Aspleniun nidus, Nephrolepis sp», Wedelia biflora, Polypodium phymatodes, Fimbristylis atoliensis, Lepturus repens, Triumfetta aa Vigna marina, Thuarea involuta, Ground cover tended to be spotty among the rock fragments of boulder ridges; less than 5 percent of the ground might be covered with vegetation, The ferns Asplenium, Nephrolepis, and Polypodium, together with patches of Wedelia in openings, were perhaps the most abundant plants, Invasion by wild trees from stands of native forest along the ocean necessitated nearly constant clearing, and in many places thickets had grotm up to such an extent that collection of fallen coconuts was impracticable. Types of Poor Coconuts About , percent of the acreage umder coconuts on Arno Atoll was relatively mproductive, The Marshallese recognized two types of "poor coconuts:" "Laora" and "Mellal." Each had a fairly ééndveeane see of characteristics which, with practice, could be recognized saat, 1, "Laora" on Arno atoli occurred on Amo island and in the breadfruit belt in Ine village. In its extreme form on Arno island "Laora'' was charaterized by: Yellowing and scorching of the leaflets of the coconuts, most severely at the tips, Areas adjacent to the midribs were usually green. The leaflets were sometimes twisted and more or less knotted together, - Dead leaves tended to hang on the tree, The trunks of the coconut trees were sometimes blackened, « Density of stocking was low -- about 0 trees per acre in the worst affected areas of Arno ae Dead trees were conspicuous, About 35 trees per acre were dead in the worst affected areas of Arno. The older trees were mostly barren of fruit, although they produced - 10 = inflorescences,. Apparentlythe fruits.aborted-at an‘early stage of development. The. younger trees. bore fairly well; their water was exceptionally sweet. | uF In: the worst affected areas. Qf Amo island the .ground cover con-= sisted of clumps of Fimbristylis atollensis, covering the ground between which was a leathery black blue-green alga resembling a crustose lichen, Tacca leontopetaloides was common, .Its leaves were yellow except along the veins and were characterized by the presence of circular brown spots 1/8 to 1/h. inch in diameter. __Laora was restricted to the interiors of the: wider islands, where it occurred on soils of the-Arno series. oe The Arno people attributed "laora" to a "Lack of. salt." It.is true that areas on which the malady..was:present.contained very little sodium chloride in the ground. water »,..Productive coconuts, however y occurred on sandy soils no less distant, from. lagoon and ocean and thus probably con- -taining no more available sodium... It will be noticed (Fig. 13) that the _area.of poor coconuts. on Arno formed a semi-circle-around an area of. _ secondary forest containing numerous abandoned: house-sites and yaraj (Cyrtosperma): pits. On small clearings within the secondary forest: the ' - coconuts invariably exhibited severe “laora!" symptoms. As one’ proceeded radially from the. former center of habitation immediately on the lagoon side of the yaraj pits, the malady appeared: to become progressively less - severe... It.is thus possible. that the.poor growth of coconuts in part. of the. interior of: Arno island was. related; to the former use of the land for human habitation, .... Pa oo u 2. "NHellal", wmlike "laora," occurred at or near the shores of ‘gilda, Its distinguishing features were: General yellowing of the leaves of the coconuts. New leaves were slow to twm green. The palms appeared stunted and probably died early (trees in laora areas were often very rez UIE) Dead leaves did not fall phone The trees bore only a few nuts. Ground Besos was invariably depauperate, consisting chiefly of an open stand of bunchy Lepturus repense | | The parasitic vine, Peseta filiformis, was usually abundant, Intsia bijuga grew as a straggling shrub; athoueh on better ates it became a tree 60 or more feet tall, | Polypodium scolopendria ee eran COMMON» ni ban Mellal occurred at or near the Jeecen or ocean shores of islands, on immature soils of the Shioya series. i i Stone suggested that the poor eeute of these a ol be at- tributed to sodium toxicity. Analysis by the present author of the ground water in an area of "mellal" on Langar island Senedd only 135 PPM chlorides. Judging from the relatively slight ee ee of organic matter in their upper layers, the "mellal" soils appear to be young, and may have resulted from washouts or heavy deposition of a during ty- phoons. It is here suggested that many mellal soils may be deficient in nitrates, but that eee deficiencies probably do not become striking wmtil more than 30 years after the original deposition of a soil or the exposure of a fresh surface in a typhoon, Areas washed out during the nes HEN Ae st “1918 +; typhoon supported “apparently vigorous, stands of young ec coconut tre 86 A legend dated thet the "mela" area of Langar island was buried wider sand deposited during. an “ancient typhoon. Tne e2 sxcavating a ‘pit to the water table, however, the writer encountered no vee inate’ Begnonies of Copra Production on Arno The prac of copra has fruditidatiod widely. since nee Mrs He Be Blodgett, manager of wi Island ‘Trading Company Majuro, paid the follow i ing prices per ton ae maps delivered | (on puxe trucks) at Majuro: Fatty OBO! ioe 148100 i Jan, 1, 1952 we f Feb. 10, 1951 130 Fone AmgNEEoen iheath: iyGlEleitus ‘san Oct. 1, 1951 105 Jan. 1, 1952 96 April 1, 1952 5. Re ae i July Ly 1952 70 meh Sept. 1, 1952 10 The 1952 price of 33¢ per pound, after fixed costs were deducted, a left very little for the "ri jerbal" (workers), The chief costs vere: Bina of the ned) 1¢ : oie Share of the alab 0. bh shipping costs (Ammo-lajuro) 0 am Total 2.1¢ That is, at the price of 33¢ per pound, the ri jerbal netted only ‘1.4¢ per pound ($28. 00 per ton). : . The share of the iroiy was fixed; it ‘did not saviscvsnsily with, the price paid by ITC Majuro,” © believe 2 the same was true of oie a of the alabs 5 16) es At the rate of 1¢ per pommd of the 622. tons of copra sold at Majuro, income of the iroij of Amo in 1951 was about $12,100, or $5.5) per acre of coconut land. Although the political and social prestige of the royal classes of the Marshall Islands is said to iiehreinte sane ASpoehr, 199), their economic position has probably improved with the introduction of the copra trade. The relatively large income derived from "rent" of copra land may have been a primary consideration in the revolt of the three iroij erik of Liwaito (Mason, 1952). In 1952 these men in effect con- sidered themselves iroij lablab on at least part of their land -- that is, they demanded (but did not always receive) the share of the iroij on that land, In 1951 ITC Majuro purchased 622 tons of copra from Arno atoll. This copra was produced by about 216,000 trees TABLE II Type Estimated coconut trees Acres Total Trees per acre : Coconut 3 95 Pa ee 202,000 Laora 60 65 35900 Melial 70 Fi Sat 2,200 Breadfruit 15 ‘QNT h,150 Scrub Forest 5 566 2,830 Pandanus swamp 5 6 270 Secondary forest 10 75 750 Total | 216,100 * f, considerable :proportion of the copra share of the iroij was set aside by the council of alabs for the payment of extraordinary medical expenses incurred by Arno residents. The prestige of the iroij, however, was nailer) not lessened by such philanthropies, however involuntary they may ve been. we HY “Th shipping copra the dried coconut meat was Ment tee into coffee bagse The contents of about 350 nuts were sufficient to fil ik one bag, which then pela en Jeon 106 pounds in fetch That is, the contents of a single sd ee dried, merenee about 0,303 pound on ies averages It follows that ene, contents of Aa me 10° wie produced on YBa x 10° trees were i ekeeveyt to Yajuro | in i 1951s Thus, the average number of ripe nuts harvested for “copra: was 196 O3 0428 oe were oroduced per acre of coconut ‘Lande : | Local. consumption of cocontne as food is here estimated from the Majuro data of Spoehr (19195 peis3). in 19h7 58 3 persons belonging to 7 households consumed in 21 dans 1083 nuts and 607 querts of coconut | toddy (jukaru). It is ‘aifPicult peor to estimate the equivalent of tHe toddy in fa Copeland (1931) stated oe 5765 coconut trees in ‘the Philippine Islands produced an average of 0 065 liter per tree per day in the year 1909-1910 a aking jukaru an inflorescence of the coconut tree. ‘is cut off and the sap dripping from the cut stalk is collected over a ‘period of 2 Ye 3 weeks. Assuning, conservatively, eh that the inflorescence would have produced 5 wong had it not been out, | one quart of jukaru is approxinately equivalent to 0. old nut. Thus the total cotisumption: of coconut products as food reported by Spoehr was approximately the equix ralent of 1355 mutts, or Lei coconut per person per day. I suspect this ees is conservative if applied to Amo. Majuro village was erouded with wer refugees in a when Spoehr made his study, and coconuts were gn saad short wie Accepting this figure as being of the pore ees order of magnitude, however, the total local yearly cone taon ‘Of. coconuts as food at Amo is about 1,00, 000 -15 =, nuts, or about 1.85 nuts per bearing tree. Nuts processed locally for coconut hair oil and massage would probably bring the figure to 2 nuts. per bearing tree. This total is the equivalent of about 60 tons of © copra. ; Summarizing these data for local and export production, on Arno atoll about 21 nuts were harvested per bearing tree in 1951, the equivalent of a production of 0.31 oe of ba per acre of coconut land. How do these figures compare ‘itt those of other copeersinedieine areas? Webster (1920: p.l3) stated that "owing to close spacing and no cultivation, the beech annual yield per tree in the Philippines is estimated at 25 nuts," but that on well managed plantations an average of 60 nuts per tree enti be expected. Cooke (1932) reported that in well managed plantations in Ceylon on good sites where the number of trees per acre was h8 to 60, yields of 66 to 86 nuts per tree were abe tained and that 15 piculs (0.9 tons) of copra per acre were harvested. Child (1950) estimated” that world production of copra in 1938 was 3,906,113 metric tons, produced on about 9,616,000 acres, or about 0.45 tons per acre (a aweehe ton equals 220.6 pounds). The estimated production in the "Pacific feeeaneat was 0.333 tons per acres It is apparent that Arno's per-tree and per-acre productivity is Low compared with that of the world, and even with that of the average of the "Pacific Territories." This low productivity is probably in part due to inefficient harvesting, although probably no more than 25 nuts per tree could be expected, given the crowded conditions of the . existing groves, Total production of the atoll could no aeube be in- creased by converting some of the land under scrub forest to coconut %* "Based on Schnurmacher's Review-ofCeconut Products for 1938, lanila, 1939 0 Table IV, Pe2e,!! a) 16 wo eer era Much of this land, however, is not well suited to copra production, as most of it consists of boulder ramparts and stony land exposed to wind and salt spray. Assuming that all this land Ge planted to coconuts, the acreage under coconut plantations would be increased only 25 percent over that of 19526 The gains in copra production to be obtained by iereans in acreage = broke efficient hapresteho eas: probably slight, however, compared to those which might result from more careful management than was practiced in 1952, If the groves were thinned, competing nostcbr cp kept more efficiently in check, the practice of applying phosphate and perhaps potassium fertilizers initiated, and a program of introduction of hie eemione races established, pereacre productivity might well be doubled. The eee tad on of a variety such as the Philippine Romano would save considerable labor in making copra, Only 3,270 Romano nuts ere Pores ta make one metric ton of copra on Mindanao (Webster, 1920), whereas about 7,280 nuts were necessary at Arno in 1952. It is obvious, ong that programs designed to increase copra production at Arno and elsewhere in the southern Marshalls would require changes in social patterns which might be wdesirable or wmpopular. Although complaints about the low price of copra and the resulting low income of the "ri jerbal" were frequent in 1952, it was the author's “impression that the people of Ammo were not eager to-acquire wealth and property. Regular working hours were mlmown. -To produce enough copra to buy cigarettes, needles, thread, soap, fishing line, hooks, and other ‘trade goods to last his inmediate family one year, a man had only to ‘work tyo to three weeks ‘at making copra. Much time was spent fishing, visiting, and talking. More efficient management of the Arno coconut ee groves would possibly doublé the ash income. of the tri jerbal", but only at the sacrivice of leisure hours. Abundant .leisure:.tdme: is: perhaps.-one . of the host pleasant aspects ‘of Marshallese life, Programs that .might dew crease it in the namie’ of increased efficiency :should be recommerided .only “after careful consideration has been given td. their- sociological implications. "-BREADPRUIT PE Primary species: | Artocarpus altilis 5 to 30 trees Ber es . Secondary apeptebe re nucifera (10 +o 15 trees per acre) pa ‘“ “Pandanus ‘ teetorius : nh _ Prema obtusifolia “-. Allophylus timorensis: ° Pipturus argenteus Guettarda speciosa: ::.: _ Morinda ae ae Ground layer: Wedelia biflora Vigna marina Tacca leontopetaloides . .. Gentella asiatica Lepturus repens Tpomoea tube Thuarea'involuta §-+.:-» 4, :littoralis. Fleurya ruderalis Oplismenus spe “Fimbristylis atollensis ~- Triumfetta procumbens Polypodium scolopendria Hedyotis biflora Asplenium nidus Alocasia macrorrhize _ Nephrolepis sp. Crinum asiaticum - 2. .-.§tenotaphrum subulatum _. Yaraj pits: Cyrtosperma chamissonis . --Vfusa paradisiaca — Hibiscus tiliaceus Colocasia estulenta9 -'. | - . Clerodendrum inerme me, considerable proprotion of the centers of the wider islands was ocoupied by breadfruit a ea A ‘common ‘spprexinate pes of “his ~ were, on the Lagoon, sides of islands was “the main village path parallel to ERE lagoon. On relatively narrow islands the boundary on the ocean sides was eel! the beginning of the stony Land complex or boulder a _ On. wider islands, however the breadfruit : zone > usually did ns extend ree this far, for potential .breadfruit land not needed for food production was given over, where feasible, to copra culture. The breadfruit type was best developed at Ine, where pigs had not been allowed to run wild and devastate the yaraj. (Cyrtosperma) pits, as they had at Amo, Bikarej, and Tutu. At Ine these pits were about 15 to 20 feet wide, 30 to 80 feet long, and 5 to 10 feet deep; earth heaped up around their sides formed the soil upon which the breadfruit and plants associated with it grew. The breadfruits grew as widely _ scattered trees 50 to 70 feet tall, with spreading crams, 25 to 75 feet in diameter, A striking feature of this type was the nearly complete absence of small diameter classes, Breadfruit trees between one and eight inches in diameter were, definitely uncommon 5 most were 18 to 36 inches in diameter above -the buttresses, Coconuts were common in the forest, but occupied only a small proportion of the canopy. Sprouts of Allophylus, Prema, Guettarda, and Morinda were usually abundant along with small breadfruit seedlings, and formed secondary forests in .-abandoned breadfruit groves. Openings in the canopy permitted rank — -growths of Wedelia biflora, the. stems of which at Ine were often inter- laced .by the purple Semered morning glory, Ipomoea littoralis. In more completely shaded places the. grasses Oplismenus, Thuareas or Steno» _ taphrum covered the ground. In fact, the nearly complete ground cover _- in both breadfruit and coconut types -- except where the latter occurred 'on ‘stony land near the ocean -=- was a characteristic of these artificial forests. not shared by types. in which wild native trees were predominant. At Ine the bottoms. of the yaraj pits were usually occupied by the gigantic Cyrtosperma, but abandoned pits supported rank growths of Clerodendrum or thickets of Hibiscus tiliaceus. A fresh-water swamp at Tutu island formerly was planted to yaraj, but since pigs made its oo 1D te culture impossible Bruguiera had been introduced. According to Fosbere (199) and Cox (1951), ground-water salinity is an important factor of the physical in tae aNe liniting the distri- bution of the breadfruits The typhoon which swept through the eastern Marshalls in 1951 provided striking confirmation of these macet Al- though most of nee atoll was relatively maffected, the preadfruit groves of Bikeres were hard=hit, There many trees ere relatively ae: to the ocean, along the main eos which extended in a north-south direction along the west side of the joven High storm waves from the west washed over the dand and extended perhaps 200 feet inland. All breadfruit trees in the path of these waves were killed or severely eee but coconuts and cultivated varieties of pandanus were maffected. Tn 1952 patches of dead breadfruits sti alias with Beata bearing bvisees so that it was still possible to SetéMand at which points and hot far the storm waves had washed over the land, Agricultural land throughout much of the tropics is characterized by the presence of a relatively uniform weedy flora. These weeds owe their present wide distributions to the agency of man who, with modern, rapid methods of transportation, has unintentionally carried sone of them to even the most remote places. Amo atoll in 1952 had about 26 species of unwanted alien weeds, most of them without well established native names or uses. Most of these had probably been introduced after 1873, when an Havaiian mission was beet eka at Ine, It is of interest | to observe that on Arno Si orhigg Roe weeds were virtually restricted to village paths and houselots. Although the breadfruit and coconut groves were maintained as such only by periodic disturbance in the form of - 20 » clearing and burning underbrush, their floras consisted almost exclusively of native species or plants of ancient aboriginal cteneiva ton Moreover, ‘it is a curious fact that the native herbs of coconut and breadfruit forest, with BeTTy few, sxceptionss, did not occur’ in He forests of wild native trees. Thus arises the problem of ‘the nature of their ancient, pre=aboriginal habitats, for the coconut and breadfruit groves, their present habitats, became established on Ammo only after the arrival of the Marshalleses naiecutoen ae these species were found in 1952 on open, newly formed "eal gree Mite Langar and Matolen, islands nearly obliterated in the 1905 typhoon were reforminge The following species of herbs were observed on sand freshly deposited along the lagoon shores of these islands: Lepturus repens 8 Tena ‘marina Fimbristylia atollensis | bbe WedeHia bitoni. Triumfetta procumbens ~~ ~ Cassytha filiformis '‘Siape ainare is no reason to believe that typhoons were formerly any less Greamerts than they are now, it is extremely probable that this habitat has always been available to these species. A more extensive search would probably reveal a few other species of the ground flora of bread- fruit and. eheonie soaa =~ Gee, Euphorbia chamissonis and Thuarea alee! selva growing in similar situations. Certain species which were restricted to relatively moist, partially shaded situations ‘may never have grown on open sandy shores. Could such plants as Centella asiatica, “Hedyotis Baereros Tpomoea littoralis, and ‘Oplisinenus spe be weeds of anesene aboriginal introduction? It would ‘be surprising indeed if such dabeetnet cons did not take place in the past as they do today. The seedless varieties of breadfruit, for example, are propagated chiefly by root suckers...In transporting small trees from island to island the | wandering Marshallese. could scarcely have avoided accidental introduction of unwanted seeds.in the soil around the roots. Other weeds of possible both restricted to village paths and houselots, but possessing well established native names, SCRUB FOREST Since the introduction of the copra trade in the latter ee of the 19th century, about two thirds of the land surface of Arno atoll has been cleared and planted to coconuts. Probably at Least 80 percent of the 222); acres of Amo under coconuts in 1952 must have supported stands of wild native trees and shrubs iess than 100 years, before this sale ee made In 1952 native forest and scrub eared enaeel in the form of numerous small stands which for the most part were restricted to situations in ees copra production would probably have been relatively uneconomic. Native forest and scrub was thus characteristic of small or narrow islands remote fron villages and of stony and and ramparts, especially those exposed to salt spray and the drying effect of the northeast trades. In such situations the native brush acted as an effective wind- break, protecting the coconut and breadfruit trees planted farther inland. | This wild vegetation was composed of. woody plants ranging from 2 to 80 feet in height, and. varying bees iu from nearly impenetrable thickets of shrubs along ocean shores to closed-canopied forests of large trees through which one walked with ease, Furthermore, the type included 17 woody species, few of which were distributed over the entire area of the type as it is mapped. In view of the great range of variation of physiognomy and floristic composition of this vegetation, it is perhaps unrealistic to consider.it a single type. Largely for eomnentened of Mapping, however, the scrub forest is rere, preated as awit, It was not possible in the field to examine each #2 ihe hs, atonal mapped as serub forest, and in general the component species cannot be distinguished on . aerial photographs of a scale of less than 1:5,000, Subdivision of the type based on floristic composition was thus impracticable. Furthermore, the average size of the stands of scrub forest as here preseneea is only h acres; many of these are 200 feet or less wide. Splitting of the type would, have Jed. to obvious difficulties in drafting the map, Although the scrub forest of Arno was reduced in 1952 to numerous small stands occurring chiefly on the ocean sides of windward lies it is perhaps possible conceptually to reconstruct the appearance of the vegetation of the atoll of the early 19th century, when perhaps only about 15 percent of the land surface of the atoll was appropriated for agriculture. Certain regularities were peered, fey example, in the distributions of the native trees as one proceeded across the islands of the atoll from lagoon ie oceane Thus, large individuals of Pisonia _ grandis occurred on several of the windward islands from sitions to Takleb, It is probable that these trees were relics of former forests in which Pisonia was an ieeneenh species. In 1952 Pisonia also formed forests along the boulder ramparts of a few islands, but the faeee aes nae ‘smaller. than the scattered individuals still standing inland. Qn the basis of such observations it is possible to construct a diagram (Fig. 12) on which are charted the distributions of the more important species of tha nahirs woody flora across an hypothetical, relatively mdisturbed atoll islands For each Bpecies une ordinate represents the author ts eninitajmasagh of the importance of that species in the vegetation, Those parts of the figure in solid black represent trees actually observed at Amo atoll; ea is: areas aad the author's extrapolations 5” For example, a few large Pisonia trees were observed about 300 feet from the ieaties shone of Takleb island; surrounded by young breadfruit and kapok eeagieke These Pisonia trees, together with other individuals observed ms similar situations on other islands of the atoll, are represented in , the soild black area aed the lagoon shore of the hypothetical islet. The chart thus constitutes a diagrammatic coriceptual ‘synthesis of the author fs BS nell Bas: of many stands and individual trees on several eeepc | . H In many cases it was impossible a impressions of "importance! with numerica al’ sane based on sample plots, for they are ergs in part on the observations of the heights and trunk and crow diameters of Psa sit ora Despite the highly subjective manner in which Fig. joess ceeerrnetct the author believes that it is in a sense Séproducibie nt is@o, that an independent investigator on Arno could construct a similar series of diagrams trithout having first con- sulted the ones here presented. Essentially, the diagrams constitute an attempt to avoid verbal descriptions of the field occurrences of trees (Cefn, "Hernandia sonora: commonly encountered slates lagoon shores, especially along iain paths, but on Langar observed on stony Jand 200 to 00 feet from the ocean, where the trees attained maximum heights of approximately 60 feet."), ~ 2) It may be noted from the chart that the author noted a general zonation in the flora and in, the vegetation us expressed in the "impor tance" of the different species, Certain species, such as Scaevola frutescens and Tournefortia argentea ranged completely across the islands of the atoll, but were much more abundant and formed a larger proportion of the vegetation near.iagoon and especially ocean shores. Other species, such as Pisonia grandisi,.Allophylus timorensis, and Pipturus argenteus, -were far more characteristic of the interiors of islands, where, indeed, _ they attained their greatest sizes. Guettarda speciosa and Pandanus tectorius, on the aber hand, were more or less ubiquitous. In general, __ the distributions of the species overlapped, but did not coincide. Furthermore, as is discussed below in more detail, the forest and scrub as it existed on Arno in 1952 was commonly characterized by the presence of small stands consisting of a single species of tree. It would seem wirealistic, therefore, to attempt to distinguish within the broad scrub forest type plant, commmities based on floristic composition, Not only did the distributions of the species making up almost any conceivable plant commmity fail to coincide, but very commonly the component species did not even grow.together. Indeed, the author found no evidence on. Arno that the native plants were distributed in such a fashion as to suggest that any of the species might be mutually dependent. Competition for light, nutrients, and eS HeReaNS occurred, but such inter- actions did not result in apparently integrated plant communities the composition and appearance of which were relatively uniform from place to place, | Since it is mrealistic to attempt to classify the native scrub forest of Arno atoll into conmmities based on differences in floristic compositions = 25 = it is convenient instead to consider the variation of the scrub forest in relation to the different habitats on which it was found. The major physiographic divisions of an atoll island (at Arno) were boulder ram~ parts, low interiors, and Lagoon shores and dunes. The low interiors. . oar conveniently be cl.assed into those vossessing stony or sandy soilsy since it is lmowm that the salinity of the growmd water of an atoll islet depends on the permeability of the materials making up its soils. ' When one compares such a classification of habitats with the distribution of fers as presented in Figs 12, it is evident that a general corres pandence exists, although, of course, it may be objected that this cor- respondence is manufactured, since the diagrams are based to a large ex-= tent upon subjective impressions. Nevertheless, it is true that Scaevola and nournerartie se important species were largely restricted to boulder ramparts and lagoon shores, Ochrosia, Hernandia, Intsia, Barringtonia, ' and Cordia occurred as iinportant species only on stony land on the ocean sides of fetnde (ages tee ees on ere fom the ocean, Pisonia and Allophylus pened ie Gece ee tosnene on tle sands of island interiors. "The following descriptions of the native forest’ and serub, then, are basically descriptions of the "scrub forest" vegetation w: it occurred on the different "natural areas" of the ‘atoll, _ BOULDER RAMPARTS | Primary species: Scaevola frutescens _ Secondary species: Tournefortia argentea Guettarda speciosa Terminalia samoensis Pandanus tectorius Cocos nucifera Pemphis acidula - 26 » At Arno coconut plantations. very commonly extended to the tops of boulder ramparts of the ocean shores of leeward islands. The proportion of wild woody shrubs in the understories of these groves commonly in-" creased toward the ocean shores, and thickets occurred in neglected plantings. On ocean shores exposed to the full force of the prevailing northeast trades -= e.fe; Bitarey, Mahwat, land ine chain of islets ex- tending between them; Ijoens; and the northern tip of Matolen ~~ native shrubs and trees formed belts of vegetation on the tops of the boulder ramparts and extending inland in places 200’ feet or mores ne In most such situations the primary species of the vegetation of the tops of boulder ramparts was Scaevola frutescens. Thickets of this species |in places extended about 75 feet inland from the ramparts, increasing in height from one foot at their outer margins along the stony beaches to fifteen ‘feet where they merged with the forests of larger native trees in the interiors of the islands, These thickets were in effect sloping | - hedges, consisting of an unbroken layer of large, fleshy leaves supported by innumerable upright. branches, Although such thickets. were not Bok literally impenetrable, they were exceedingly difficult to PRS and have ‘been aptly termed "beach: barriers" by Mr. E. H. Bryans Tournefortia. argentea, Guettarda speciosa, and Terminalia samoensis usually grew. among the shrubs of Scaevola, in places partly replacing them over small areas. STONY INTERIORS Primary species: Pandanus tectorius Ochrosia oppositifolia Guettarda speciosa Tournefortia argentea Cordia subcordata Intsia bijuga Allophylus timorensis Pisonia grandis ae Hernandia sonora ie Barringtonia asiatica Cocos nucifera Secondary species: Terminalia samoensis Scaevola frutescens Pipturus argenteus Soulamea amara Ground. layer: Asplenium nidus --Polypodium scolopendria Feperomia Spo Lianas: Ipomoea tuba Wedelia biflora Epiphytes: Aspleniwn nidus Nephrolepis spe Height of Canopy: 15 to 60 feet; mostly 25 to 35 feet, Density: 100 bo 300 SE bre teal ner acre Ljos ach are, = A fore in eed 1h species of native lente occurred was commonly encountered on the on soils of windward aaa immediately iiiand from the beach-barrier scrub of Scaevola frutescens. The width and development of this forest appeared to be largely dependent upon the extent of human activity. Where the land was remote from human habita-= tion, too narrow Per Brroctive crop production, or possibly in areas: of disputed land omership, the gone of native forest was in places” 300: or more feet wide. In such situations as one proceeded inland the size of the trees PHereseed wuitil individuals of Pisonia, - Tite Parringtonia, nll or Ochrosia might attain heights of 50 to 60 feet and trumk diameters of 20 inches or more, The average diameter of the trees over 1 inch deb.h. in one Ochrosia stand, was Lal inches. * diameter at breast height, me Ole (Basal area: 16h, Sq. ft/acre; density: 150 trees/acre). To walk into such a forest from the ocean reef flat was to step into a different world, The light intensity fell to > only ub, percent of the glare obtaining on . the reef flat. One was inmediately conscious fon ies numerous: coconut and large hermit crabs; the former usually scuttled rapidly to vee excavated in the purple-=brown phosphatic sand between the stones ecole of the forest floor, At the approach oo an aden, the terns meggting and nesting in the trees set up a hoarse clatter, Where the bird's nest fern, Aspleniun nidus, - grew among the ‘fragments of coral rock, its enormous leaves -- to five feet long -=- were often eonspieuousty spattered Meee the white droppings of these birds. cane ane there patches of Posoecrias formed small "rock gardens" under the trees. Elsewhere, DOWBYER» ae | _ ground, was quite bare of ee) other than ee Si ABpELY. algae coustsne the rock TeReD DEG Tree seedlings were usually not conspicuous. Qn the windward islands these eae See in their growth the _ prevailing direction of the winds es ena of Che forest and of tie _ Seaevola scrub to windward was in ffect a plane of leaves dipping toward the ocean beach at angles of 9 to 13 degrees. On Langar, Barringtonia _ and Hernandia were predominant in ey anat portion of the native foreste These trees appeared. jbo, have, been ee ae in vheir ee ere by ris the salt-laden winds, for aithough os were massive, cae trunks over 2 feet in diameter, many were scarcely 25 feet tall, Their trun'ss were inclined at dangerous angles or actually prostrate on the ground, ine branches projecting aes the canopy . of the spores were mostly. ale The grotesque shapes of ea suggested the. eee Phovent that failing in their struggle to pierce an inyisible barrier above them, ober grew horizontally instead of vertically in a claustrophobic attempt to escape FOO from the inadequate space within which they were confined. Farther ‘Inland, however, Hernandia trees of the same stand were erect and ‘attained heights of 60 feet or more. | An interesting feature of these forests, probably to be attributed ‘to accidents of distribution of propagules, was the tendency of the com- ponent species to form alternating nearly pure dganae averaging perhaps O.1 acre in areas In walking parallel to the shore, one might pass from darl patches of Ochrosia forest into sunny open stands of Pandanus tectorius, and from these into thickets of Guettarda speciosa or Allophylus ‘timorensis. Here and there were observed groups of five and six small Souvlamea ‘trees. Elsewhere, as on Takleb island, groups of tall Intsia or Pisonia occurred, or the heavy branches of Cordia subcordata leaned to the ground and © ‘struck root, giving rise to thickets of that species. These alternations of single=species dominance occurred with no regularity, nor did there appear to be any marked tendency for one speciés to replace another, It “is true that seedlings and’ saplings of Ochrosia oppositifolia were con-= spicuously vigorous when growing in heavy shade, but a high proportion of the large fruits of this species failed to germinate among the stones and ~ cobbles of the forest floor, and there would seem to have been no agency , other than heavy storm waves by which these awlward propagules might be carried from place to place, Guettarda speciosa, and the coconut, however, were ubiquitous, and old trees of Tournefortia argentea occurred in - scattered fashion throughout the forest. " Stands of large Hernandia, Barringtonia, Ochrosia, Pisonia, or Intsia were rather exceptional. More commonly the native forest con= sisted of thickets, sometimes containing several sprouting cut stumps; = 30 ~ : daeamaracul behind the ocean beach barrier of peal. frutescens. In * eg Sats. Be a Yih pics situations ae naost cormon igs 2 were Pandanus, Guettarda, and Tournefortia, the latter ‘tyo vith s sous eae crooked boles, sending ‘out many Ascari, spreading branches. (Says Gs eo ae Scaevola frutescens were also abundant, and such lianas as Tponoea tuba and a scendent fesley oredr biflora bound some of these thickets into nearly impenetrable ee Bi inigso bil 8 ah aeons 3 SANDY INTERIORS Primary’ spécies: Pisonia grandis: f Allophylus timorensis _Cordia subcordata .?: tates bi Juge Secondary cee oe Piptaie Bobs Juole fini se '° Bremnat diibusifolda 07-202 Guettarda speciosa hs Pandanus: testorius.: 9 7 4 Lape aM Ground layer: Polypodium scolopendria Height of Canopy?’ 20 to 80:f8ets; 2) i tate te! Density: 15 to 300 trees. per acre over 1 inch dsbehg: 9 0s 12 In 1952 3 massive Pisonia trees stood in the middle of ‘Takleb island, ‘‘gurounded by ‘a forést of ‘sriall kapok and breadfruit ‘treésé .The largest of the Pisdnias was about 80 feet’ tall’:and 28.) feet. in circumference. Other large, isolated Pisonia ‘trees were.observed in the interiors of Enidrik and’ Jilang islandss * Since it appeared possible that these trees might be remnants of the native forests which must have once covered the interiors‘ of these islands, the writer questioned Felix Js, the’magistrate of Amo atoll in 1952, whose family holds land rights on Takleb, as to sPhis dnistony, of ‘Taleb. He very pea oer the Ql owine cee Until 1876 Takleb island was an "island of. bide. fse5" a eee - 31 - bird reservation upon which trespass was forbidden (caf. Tobin, 1952). Periodic visits were made by authorized collectors to gather birds ahd eges. In 1876 the prohibition against visiting Takleb was removed, and pegs 1886 Tobikley a "brother" of Felix, having obtained rights to the use of the land, commenced to cleat the wild forest idea plant iy breadfruit and coconut trees. Felix stated that the original forest of the interior of Takleb did not consist exclusively of Pisonia trees, Rather, the Pismias occurred in two Groups, which alternated with pure stands of Allophylus, Intsia, and Cordia. Species of birds not present at Arno in 1952 and not reported by Marshall (1951), including the "ak" (frigate bird), "nana," and "kalo," roosted in the trees, together with many terns ("rabit," "jakar", etee)o. He) Peekee wad eracenoue much larger than any remaining on Arno atoll iri 1952, The kapok was intro- duced in 1915, seed having been sent from Ponape. Felix!'s account was confirmed by Tobu, the paramount chief of Amo in 1952, who added that Nami island had also been an island of birds until 1876, when the iroij, Lekaman, removed the emo (tabu) from both Takleb and Nami. ‘The writer encomtered numerous small Pisonia trees in the interior of Namwvi, growing on phosphatic sand and rock, All appeared to be of sprout origins In view of the very great powers of vegetative reproduction possessed by Pisonia grandis, I consider it quite likely that the stand of sprouts and iio observed on Namwi in 1952 consisted of the same trees which grew there before 1876 and * The term "brother" in the Marshall Islands is apparently applied to all male relatives of the same generation as the speaker. Thus, Tobikle may have been Felix!s cousin, . 32 that these trees then formed a high forest over much of the phosphatic area. of that island, i) | ' _In spite of its decimated character in 1952, the vemant of ae Pisonia forest was most striking. The pre=1876 groves of Takleb and Namwi must have been. impressive. Qn the basis of my Bape of the Takleb trees the foilewine reconstruction may be hazarded.s | The forest was gomposed almost exclusively of old trees of Pisonia. grandis between. i and. 9 feet in diameter at breast height and 75 to 85 feet tall, Assuming that ae canopy was about 90 percent closed, the density | of stocking was about, 2 eee per acres The forest floor was essentially bare of tree. seedlings or herbaceous plants, but leafy root suckers of the Pisonias must nave been conspicuouse Thousands of birds perched in whe branches of Ds trees ana the air resounded with their raucous cries. Their droppings combined with the ., decaying ewes of the Pima eer to form a black, mucky humus resembling the greasy mor type of the humid temperate regions. “over and through this humus crept , the Serpentine roots of the A hawiuinivaie are occasionally penetrating to depths of more — one ching Phosphatie Bees leaching from, ‘the humus Stained the pinkish Limesands beneath a purple=brown and with the subsequent precipitation of calcium Oiotaee. phosphatic rock _ was eee eee. phosphatic hardpan was in some piers able to perch a water table, so that small pools persisted on the gromd etal heavy rains., In places where the rock had not consolidated completely, how= . , ever, rainwater percolated through the sandy soil to the permanent ground-water table more than four feet below the surface of the soil, Typhoons left their mark on this forest, although they did little to alter its composition, The tops of the great trees in some cases = 336 were broken off, at times 10 to 20 feet above the ground, and some trees were. completely uprooted. These trees, however, were not all killed. From broken trunks and creeping roots aishoue quick=growing -suckers arose, qiuiclly filling the gaps in the canopy. Fallen logs developed roots along much of their length, and rapidly crore sprouts along their upper surfaces later developed into vrees ahah - in. 1952. seemed to be growing in rows, as if planted. ees fallen . branches struck root and developed into trees where ms light an | tensity was sufficient, , In 1952 Pisonia grandis was considered a major pest and had been eradicated as an important species from all islands of Arno atoll ex- cept Takleb, Takleb ej, Enidrik, Jilang, Natirly be Langar. In segs . dition, lone Pisonia trees stood at ie and on Arno island, . Habitats which appeared suitable for the species, however, witeadaley . din the secondary. fores® of me island as well as on Jeon and other - islands reforming between Malel and Matolen, felonies the typhoons of 1905 and. 1918. The sticky fruits of the Pisonia adhere to the feathers of birds roosting and nesting in its branches, and are be- lieved thus to be, carried from island to island, Evidently this . mechanism of dispersal is not very effective in the short PUN. A careful search for trees. of udcoubted seedling origin ewes only ‘a few, although sprouts and suckers were abundant. Once 2 alma however, the tree seemed to be almost immortal. The virtual in- destructibility of the older, trees -~ fire is the only effective means of clearing Pisonia forest «-. combined with their great powers of vegetative reproduction seems to be sufficient to accowmt for the ~ 3h former predominance of Pisonia on parts of Takleb and Namwi. It is possible that these trees were members of clones resulting from the chance long-distance dispersal and establishment of single seeds. The groves of Allophylus, Cordia,:and Intsia, which according to Felix shared the interior of Takleb island with the Pisonias, - were reduced in 1952 to a few trees growing on stony soil near the ocean side of the island. Of these Intsia is the largest tree, at- taining a trunk diameter of 19 inches, and heights of over 60 feet. Cordia subcordata is a much more compact tree than Intsia bijugas- on Arno atoll it reached heights of 5 feet. Its crowns were mostly wide-spreading and its thick branches originated low on the main bole, which attained a maximum diameter of 28 inches. Allophylus timorensis was : common in the interiors of Arno and Langar islands, It was never taller than 30 feet, and its trunks were mostly slender whips. LAGOON SHORES AND DUNES Primary species: Scaevola frutescens Pemphis acidula Secondary species: Suriana maritima Sophora tomentosa... Guettarda speciosa Pandanus tectorius. Tournefortia argentea . Cordia subcordata. Barringtonia asiatica -Terminalia samoensis. Hernandia sonora (planted?) Callophyllum inophyllum (planted) Hibiscus tiliaceus (planted) Herbaceous pioneers: Lepturus repens dala ‘Vigna marina Wedelia biflora Triumfetta procumbens Fimbristylis atollensis = 35 = Between Lukwoj and Jabu, on the long southern island of Arno atoll, and on several of the islands devastated by the typhoons of 1905 and -1918 between Langaz and Matolen eibanre forest and scrub occurred along lagoon shores, The primary species in such situations were Scaevola frutescens and Pemphis acidula. The latter comionly formed pure stands where the beach was composed of sandstone or con- -Solidated fragments of:reef rock, At high tide the Pemphis trees often overhung the lagoony so that when traveling along the beach. one frequently was. forced to wade around the. trees or detour inland, Pemphis also. occurred on unconsolidated sands and dunes along lagoon shores, together with Scaevola and, near Lukwoj, Suriana maritima, Cordia subcordata, Terminalia samoensis, Pandanus - tectorius, Guettarda speciosa, and Tournefortia argentea were other trees frequently encomtered on sandy-soils near the lagoone In. + villages Hernandia sonora and Calophyllum inophyllum were common, Rainwater runing down the trunks of the latter was often collected in large iron drums, SALINE PLATS Primary species: Pemphis sean. Secondary species: Scaevola sie don Tournefortia argentea . Bruguiera conjugata Small interior saline flats occur on several islands of Arno atoll, Comnonly these had the form of elliptical, flat-bottomed depressions about 200 feet long and 50 feet wide, Island sandstone was exposed in the bottoms of some of these depressions; drifting sand had accumulated in others, At high tide: saline ground water might rise above the level of the bottom of the flat, or where a channel opened to the ocean or = is lagoon, salt water washed in with the tides. Commonly at low tide small pools were found in depressions in the flats and the water in these was salty to the taste, | The slightly higher rim of such flats was. composed commonly of consolidated rock, and on this. rock as well as on islands of sand or rock within the flat itself occurred nearly pure stands of Pemphis acidula.s The Pemphis is a large shrub or small tree 15 to 25 feet in height. Its exceedingly hard, heavy wood, which sinks like a stone in salt water, | was prized-as a source of durable construction timbers and coconut husking sticks; formerly it was the preferred wood for spears, Where -sand had filled the bottoms of the flats, Scaevola frutescens and Tournefortia argentea together with the Pemphis formed open stands, .The largest of the saline flats of Amo atoll was found between _Bikarej and Badrbaren islands, Therey on low islands between tidal channels and pools, the Pemphis formed essentially pure standse. At high tide salt water washed over its exposed roots and those covered ‘by sand were undoubtedly bathed in the saline groumd water. Fiddler -. erabs had excavated innumerable holes in the muddier portions of the flat, and permanent. pools a few inches deep were inhabited by gobies, - which.darted into holes or under’ sheets of algae when disturbed. At Bikarej, Namwi, and Enidrik islands Pemphis bordered mangrove % swamps. It differed from the mangroves, however, in that whereas the . Latter were restricted to swamps in which peat formed wder water, _ Pemphis was most abundant on sites which at low tide appeared to be well drained. and even dry. Small depressions in beach conglomerate _ formed along lagoon shores constituted an especially common habitat. * Sonneratia, Bruguiera, Lumnitzera (Combretaceae). Ngee MANGROVE SWAMPS Primary species: Bruguiera conjugata : Secondary species: Sonneratia caseolaris Lumitzera littorea Pemphis acidula Epiphytes: Asplenium nidus Nephrolepis spe Density: About 90 trees per acre over 6 inches DBH, but variable, Height of canopy: 35 to 55 feet. The principal areas of occurrence of mangrove swamp were Tinak, Langar, Bikarej, and Namwi Islands, and the districts of Kinajong and Matolen, on Ine Island, Elséwhere stands consisting of eee a few trees may grow in local depressions. The swamps were of two kinds. The' more common ‘type was the interior swamp or "bat," cut off from any connection with the lagoon or ocean by dunes on the lagoon sides or oy boulder ramparts onthe ocean shores. Certain of ‘these interior muddy depressions apparently resulted from the formation of successive boulder ridges of coarse rock on the ocean sides of islands, The less cormon ("Jjinbatbat") tyne of mangrove swamp occurred at Bikarej and Namwi Islands, bordering the enclosed North Horn lagoon. On both of these islands shallow embayments opened on’ the lagoon, The swamp at Namwi, however, had been cut off from the lagoon by sand Laan drifted across the ovtlet. Drifting sand had also covered the center of the Namrii embayment forming a salt flat’ on which grew scattered shrubs of Pemphis and Scaevola, The @ennection of the Bikarej swamp eee ome -with the lagoon was still open, and water ran in and out of the embay- ment with the tides. During the day at low tide this water became ex~ tremely hot, and evaporation probably tended to increase its salinity appreciably. Brine shrimp: were abumdant. In both places ‘the mangrove trees formed merely a bordering ring around the interior flat. Scattered, mostly prostrate trees of Sonneratia caseolaris, which on Arno atoll was restricted to the Bikare; and Narwi swamps, bordered the central flats of sand or muds their erect, conical pneumatophores extended twenty feet or more beyond the main boles of the trees toward the center of the embayment. ‘Behind the Sonneratia trees and tending to fill the gaps between them occurred the more erect Bruguiera conjugata, perched upon which were aerial baskets of Nephrolepis and Asplenium. Seedlings of Bruguiera also became established in the mud beyond the Sonneratias, but they were apparently short-lived in such situations, Forming a zone around the Bruguiera were thickets of Pemphis acidula, the sprawling main trunks of which sent up erect branches 15 to 20 feet in height. and over 6 inches in diameter. Behind the Pemphis occurred trees of the scrub forest (especially Intsia and Allophylus), impenetrable thickets of Clerodendrum inerme, or coconut plantations .« Mangrove swamps of the interiors of islands were strikingly dif- '. ference in appearance. Instead of a mere ring of trees bordering a barren embayment or flat, a complete cover of trees occupied the inland swamps. Most of these stands contained only a single species of tree, Bruguiera conjugatas Only at Matolen did Lumitzera littorea, prized ‘for fishing poles and garlands, grow with the Bruguiera. Thickets of the attractive white-flowered Clerodendrum, also used for garlands, ut On sometimes bordered these interior swamps. Mangrove swamps of the interiors of islands varied considerably in density of stocking, distribution of diameter classes, and form of the larger trees, At Tinak and Kinajong the largest trées were mostly tall and straight, and the distribution’ of mature trees, youmg poles, and saplings, was relatively hbratintirougnduriite swamps. The largest trees: at’ -Kinajong were 50 feet or more ‘in height and about two feet in diameter at breast height. The trees. at Tinak were some- what smaller, but the proportion of tall, straight trees which had pruned themselves of their lower brariches was higher. — '-The case was different at Matolen and Langar.’ In those places the stands consisted of scattered. groups of old.trees‘of poor form alternating with groups of younger, straighter poles. A dense groiwth of Bruguiera seedlings about 18 inches tall constituted the ground layer, and the bird'senest fern grew perched ‘on the larger trees. Seedling mortality of the Bruguiera was probably high in the heavy shade, for the density of poles 10 feet or more in height was much reduced, The. poor form of the older trees at Langar and Matolen, which were about 35 to 5 feet tall, exceedingly crooked, and commonly with heavy branches originating less than 10 en ground, indicates that they must have grown under conditions very different from those obtaining in 1952. Their offspring, growing densely wider partial shade, had developed into tall, straight, slender poles which had pruned themselves of their lower branches, This suggests that the mangrove forests of Langar and Matolen might have been of very recent ‘ origins The older, poorly formed "wolf" trees must have grown in =- 0 = ‘openings as scattered individuals and small groups. Those parts of the swamps occupied in 1952 by young poles were probably also open prior to the establishment of the latter, for the stocking was dense and uniform and the trees were of nearly identical size, diameter, and form, It is not impossible, of ‘course, that such local openings could have resulted from the uprooting of patches of a former oe during typhoons, but no evidence of typhoon damage in the form of fallen logs partially or wholly buried in the peat could be found; the 1918 typhoon, in which many breadfruit trees were uprooted on Langar, is said to have caused no : appreciable damage to the mangrove forests of that island, "Highegrading," thateis, the practice of harvesting the more desirable trees whilst leaving behind those of poor form, may have contributed to the prepon- derance of wolf trees among the older age classes. In 1952, however, cutting methods consisted of removing only a few of the better formed poles. In the terminology of the forester, this constituted merely a "light. thinning."- Clear cutting was not practiced in 1952 on Arno atoll, and perhaps never was. Thus, previously existing open places of con- siderable size probably cannot be attributed to typhoons or former puttin practices. It seems at least possible, therefore, that the mangrove forests of Langar and Matolen originated in relatively open, swampy depressions with the establishment of the present wolf trees... When these trees reached maturity, their fruits were scattered thickly | throughout the swamp, giving rise in openings to the present dense stands of poles, ‘In other words, the mangrove forests of Langar and Matolen swamps may be no older than the wolf trees still growing in them, Dr. Harold St. John had suggested to the author that certain of the a be interior mangrove swamps might have originated through the deliberate introduction on ‘the part of the Nershallese of fruits of the Bruguiera into open swampse One clear case of artificial introduction and . establishment of Bruguiera was actually encowtered, suggesting that at least some of the existing interior Bruguiera stands of Arno atoll - may have originated in this-ways.: Qn Tutu Island a fresh water swamp was encountered (chloride | - concentration 15 ppm, less than 0.1% of that of the local seawater), in which several young trees of Bruguiera were growing, The know history of this swamp illustrates. the possible origin of other inland mangrove forests. :Yaraj.-(Cyrtosperma) was formerly grown successfully in the Tutu swamp, but the rooting of pigs. made its culture impossible, ; and Bruguiera, previously absent, was deliberately introduced some time between. 1933 and 19:2, In or ayshemcigaene at least two age classes presents The older, larger trees, growing in openings among taller coconuts and pandanus, were of distinctly bushy habit, and . seedlings of Bruguiera were scattered throughout the swamDe ‘The ‘< »Bruguiera plants were. perhaps not so vigorous as those. growing in more saline environments, but they were surviving and reproducing and may, well tend eventually to-displace the other species of trees now present. If this. in fact occurs, a stand will be formed con~ sisting of scattered groups of old Bruguiera trees of poor form, » originating from hands full of fruits casually tossed into the swamp from-points along its margin, alternating with dense growths of young saplings and poles. The stands of Langar and Matolen had precisely this appearance in 1952, so that similar origins are not impossible. It may further be conjectured that as the old, poorly formed trees of oy Bel Langar and Matolen die, they will be replaced by their taller, straighter progeny, giving rise to stands such as were seen in’ 1952.at Tinak. and Kinajong,. ‘The relative antiquity of the stand at Kinajong is further suggested: by the fact that. the place name is evidently:.derived from "jong," the Marshallese name for Bruguiera. That’ is, the mangrove swamp is probably older than the name of the district, if the latter was in fact named for ite It is of course possible that certain Bruguiera swamps of the interiors of islands became established through natural agencies alone. High storm waves washing over. the land may have carried Bruguiera_ fruits into previously existing interior swamps, or stands of Bruguiera formerly growing along lagoon or ocean shores may have been cut off inland by the formation of new dumes or boulder ramparts. The writer, however, walked the entire southern lagoon shore of Arno atoll from the northern end of Arno island on the west to the northern end of Langar on the east ae pbateiae a single plant of Bruguiera growing in such ae In ia among the northern islands of the atoll he noted Bruguiera growing along lagoon.shores only at Bikarej, where scattered trees Seated silaae eas ‘the: interior embay= ment along the northern Lagoon shore: ‘i ttiab ielande Nor was Bruguiera observed growing along ocean cribs, although a special search was made for the tree in such Snelaet new It is a common impression bam mangrove’ eagauciins are invariably saline or brackish. Such was certainly. not via case in the stand developing at Tutu, where the chloride caliden tucitaic of the swamp water was only twice that.of the local rainwater, -Analyses of the a! oc eteee Pera . 3 : . ua derors water. in‘ the Matolen Bruguiera-Lumitzera swamp demonstrated a vsehloride concenttationof:only 640 ppm, about 3 percent of that of the local seawater. The:water of the Tinak swampy however, was . distinctly salty to the taste; and at high tide the water of the ‘Bikarej swamp.mixed freely with that. of the North Horn lagoon. r “Secondary. species: Cocos nucifera: Obviously, then, Bruguiera is capable of growing in swamps having... an extremely wide range of salinity.; Consequently, it should never 8 In passing, it should be noted that large numbers of fish- .- eating seabirds roosted and nested in the: Bruguiera trees of the. ‘'*Langar and other: Arno. ‘swampss: .The:,high: phosphate content of the, : ‘Arno mangrove! péats’' (Stoney 1951L):-may:-be conveniently-explained by their"presencey? ent OF Jn: ore IR eran er Bi iorpesttinlaengs, Sey wien £ a f : u 4 we Sin are a Mum eunsh bak Vn? GRR GeP ae Gisiee Se ro “s Sy TC eT PROC PS) “Oh Re ee UP mapusie deh, Habre : UME EVO) Meier: Eee arn cas CUS ene | ies SOP Bis gee ‘Primary species:' Pandanus tectorius:. ' Hibiscus tiliaceus » . Intsia bijuga . woe! Morinda citrifolia Allophylus timorensis:- Ground layers Eleocharis sp» »itiaivine te 7 eg ait cel Bhan eh: Dryopteris goggilodus OBA IHE ' Polypodium scolopendria’ .. oo 6.9% Epiphytes: Polypodium scoloperidria:. - Nephrolepis spe Asplenium nidus «©... : Fresh-water swamps or“bogs were ‘observed on Ulien, Tutu, and Arno islands, Of these the Ulien swamps were the largest and most o~ lh - interesting. A series of swamps alternated with higher ridges com- - posed of coarse fragments of reef rock, Perhaps the most remarkable feature was a small pond, about 100 feet long and 20 feet wide, which ‘contained knee-deep standing water when visited. This water was fresh to the taste, and apparently not saturated with calcium or magnesium salts, for Stone (1950) stated that soap could be used. Much more shallow standing water occurred in other swamps on Ulien; on still others peat was exposed above the ground-water level, The Tutu swamp contained shallow standing water, but certain small areas in the northwest part of Arno island were merely wet underfoot, ‘Included here also are the abandoned yaraj pits of Arno island, for their vegetation was very Similar, The predominant tree in all these swamps was the wild variety of . Pandanus tectorius,. "Erdwan.". The trees were mostly about 35 feet tall, but density'was variable, depending to some extent).at least, on the depth to the ground water, Thus the small bog surrounding: the Ulien pond supported an open.stand of Eleocharis, Polypodium and young. . Pandanus formed a zone around the sedge, followed by a belt of larger Pandanus and finally, on’ the higher): drier land, by coconut forest with an understory of scattered Pandanus,. In the swamp west of this stand occurred both dense dark stands of Pandanus, containing perhaps: .250 trees per. acre and much more open stands of probably half the density ‘of the former, The more open stands occurred on a fibrous peat which smelled strongly of hydrogen sulfide, whereas the substrates of the dense stands were darks mucks » gopcnnts had been peonved in sig of these Swamps » When windthrown , heise root Sie voms: tore up patches of peat, exposing the standing water beneath.: Intsia, Morinda,.and Allophylus occurred on slight rises, and Hibiscus tiliaceus, elsewhere common «in abandoned yaraj pits, grew along the margins of the swamps, Eleocharis was abundant in open places in the Ulien and Tutu swamps, but was not seen on Ammo island. As noted by Stone, Dryopteris goggilodus grew in open places in the abandoned yaraj pits of Arno, but apparently did not occur elsewhere on the atoll, This suggests that it: may have been of accidental aboriginal introduction, spores possibly having ad-= hered to the corms of Cyrtosperma. On the dry-exposed peat west of the Ulien pond Lepturus repens formed bunches, and Fimbristylis atollensis was common on slightly higher situations, The history of the vegetation of the Ulien.swamps: and the sur= rounding higher land might well be worth intensive study by the methods of pollen analysis.” ‘Such a study would necessarily include an account of the origin and age of the swamps and the wmnique pond, It is sug- gested here that these interior depressions are of natural structural origin, the result of the formation of successive boulder ridges on the ocean side of the island, possibly as the reef grew outward. At least one such interior boulder ridge contained a thicket of Scaevola frutescens similar in all respects to the brush of that species which so commonly borders ocean shores, On the other hand, my informants confirmed the legend related to Stme that an Ulien retting pit was excavated by a star. * A radio-carbon analysis of the layer of peat in which breadfruit pollen first appears might give a minimum date of the occupation of Arno atoll by the Marshallese. » HO « It would seem improbable, however, that the Ulien swamps were formed by a shower of meteorites. eee ee ee ee ce Primary species : : _ Ground layer Allophylus timorensis Tacca leontopetaloides Artocarpus altilis Polypodium scolopendria .Prema obtusifolia Nephrolepis sp. Guettarda speciosa Asplenium nidus Pandanus tectorius Crinum asiaticum Cocos nucifera Dryopteris goggilodus Ochrosia oppositifolia | | (abandoned yaraj pits) Secondary species Randia cochinchinesis Scaevola frutescens Ipomoea tuba Hibiscus tiliaceus Morinda citrifolia Ixora casei Pipturus argenteus In 1952 a considerable part of the interior of Arno island was occupied by a growth of small trees, of which Allophylus timorensis was the most abundmt. Growing with the Allophylus and often locally predominant were Guettarda speciosa, Premna obtusifolia, and Pandanus tectorius.e An aggressive seedless variety of the breadfruit, "Bukaral", maintained itself in these forests by root suckering; a seeded variety, "Vijwan," was also present but less common. The main, closed canopy of the forests was 25 to 35 feet above the ground, but breadfruits and coconuts hO to 50 feet tall occurred scattered throughout, Randia cochinchinesis, resembling a small coffee tree, was common in the shrubby layer, and Tacca, Crinum, and seedlings of Allophylus were con- spicuous on the grounds Although the relative abundance of the component species of trees 9 RN, was spatially variable, their sizes in general were note’ The bread- fruits were mostly 6 to 10 inches in diameter ‘and lacked buttressed roots, The Allophylus trees were slehder. whips, .2 to inches in diameter, and the Guettarda was scarcely larger, The small size of the trees suggested that these forests were not old, This supposition was confirmed by the presence in the forest of abandoned hovse sites; around which Ixora bloomed: profusely, and yaraj pits, occupied by rank growths of the wile coneonus ("Erdwan"), Hibiscus paeenss 2a ia Dryopteris goggilodus, These secondary forests clearly. had. grown up in land formerly utilized for agriculture. An attempt was mede to determine the history of land abiandonnent on Arno. According to several informants, Arno island was once the headquarters of a powerful prince, Lojette, who is said to have con- giered ecven atolis of the Radak chain and to have originated such mst Cuore asa military schdol arid‘a corps of physicians. After Lojettets departure from Arno dissension broke out among his followers; ~ wars, famine, and disease are supposed to have decimated the population. sie Much of the history of Lojette is undoubtedly pure myth, .On the other handy Lojette is listed on the register of the iroij lablab of Arno kept by Lajibili, the former magistrate of the atoll, and Lojettets descendants are among the nobility of*Maloélap and'Majuro, Thé ruling tribe of Jebrik and Kaiboke at Majuro, for example, traces its descent to Lojette and his second wife, Liwarelik, a member of the Rarno Clan. The geneology of at iéast one descendant of Lojette, Karjin of Ailok, is known accurately and serves to place the date of birth of Lojette between 1720 and 1780," ~ 8 = * Lubne X Lojette -} Limajjen -- Neikom <3 Lekam =<) Lajen => Litoene ~+ Karjin, about 50 years old in 1952, ‘Length of generation: 20 to 30 years. Informant: Raymond de Brune Jon, an old man of Arno island, kindly showed the author the site Ora Taee elon he said aes ecu ed by Lojette and permanently atenceded atten Wis departure from Arno, Excavations revealed that the depth of accumiilabion of coral pebbles in “the ancient courtyard varied fron V2 to 19 inches. Pebbles in ‘courtyards of know age were found to eee at whe rate of about one inch every ten years. Thus the Houseeete had been occupied roughly 120 to 190 ears at the time of pa a eae tf Torette was 0 to 70 years of age when the house was Seatdened, then Lt was first occupied sometime between 1570 and 1730. ‘ ee Speed coustnde were fod inland from the one said to have gen occupied by Lojette. Most of eee eeticnecd to the commoners (kajur), sho cmetraved the nearby yaraj pits and harvested the breadfruit, | The interior of Arno island was ones inhabited until about 1860 or 1870, Lajibili, who is now Boe 60 years old, stated that Arno island consisted of 3 inhabited districts during the time of his grand= eather —“~ ocean pe ‘lagoon fies and the interior. During the time of his Baten Gale the obecn and lagoon shores were inhabited, and in 1952, pogeee: for a single Prterior house, dyucilines were restricted to the path near ine | ce apa Yee aes grown in the now-abandoned bwil (ioe during the time of the YBa ee (191b-194), but rooting by pigs ea resulted in the abandonment of almost all yaraj pits before the ar= n hee of he Americans ; T saw ihe one pit on the island which still contained oyrtospermas The jirno people aeaated the abandonment of hdtigesites in the Ipeerar of fone ciene Me a yee ci Mecrence in the human population. fbote 200 people eed on 1 Arno in 1952, but’ many of these were rela- ‘tively Boers soni ona es from Mille, Wotje, and Majuro. A former ‘larger ~ 9 « population is certainly not impossible. Much more food must have been . produced before the yaraj pits were abandoned. n the other hand, a simple migration of biological families from the interior of the island to the lagoon shore may have taken place with no change in the total popula- tion or the island. The writer observed abandoned housesites similar to the Arno ones in the interior of Ebon island, Ebon eect He was informed by Dwight Heine, a resident of Ebon but superintendent of schools at . Majuro in 1952, that at Ebon "ri jerbal" formerly occupied the less desirable interior houses; lagoon shore sites of the watos having been _ reserved for the alabs. At Ebon and Arno all the members of a are in 1952 lived in a group of houses OCR SURE SF ae a commonly shared cook house near the lagoon side of the wabos Five As the population of Aro island was formerly considerebly larger than it was: in 1952, the interior . = mee area ce secondary forest and much of the land under .c oconuts - -= Heel es “resembled that os Majuro islandll in 1952, where interior houses set among large breadfruit trees were still occupied. The yaraj nese were feria bananas and limes were abun- dant; and the underbrush was rept more or less in check, If, on the other ea the popula tion of Aro has not changed ap= . preciably and the biological families constituting a bwij were formerly scattered in a line rumning fron Gesen to lagoon, wild native trees may have made up an appreciable part of the vegetation, Perhaps enty a small area around each hovse was kept relatively clear of brush, It should be emphasized, however, that in the Marshall islands it is iS Hei to make a sharp distinction between selipil paved and waste land. “Managenent consists chiefly in cutting back competing unwanted wild vegetatione -Intensively managed coconut and breadfruit groves are parklikee On the other hand, breadfruit, yaraj, and coconuts are harvested in nnplewl tice areas which might well be papeed as secondary forest. The degree irieeaisdt varies from wato to wato. It-is not wmusual to find a strip of land almost ecnploely devoid of underbrush bordered on both sides Ge aedely impenetrable thickets of small native ee overtopped by coconut or breadfruit treese } At Buonk the time of abandonment of the interior dwellings on Arno, Pea nter es coWtenene tae Gapes ta the MovsneT wetadds, Mepatual iy land under native scrub and unneeded breadfruit forest was cleared and “planted to coconuts. Although the new coconut plantations were generally “““~suecessful near ocean and lagoon shores, the conversion met with diffi- “culty in the interiors of the widest islands of several atolls in the southern Marshalls, After a few productive years, the trees effectively *erenaeed heamiris and many died, In 1952 a few Arno landholders continued tn aetaaln their failing groves, whereas others had abandoned the at- ree to aanance moore in the interior of the island. How much of the Catesene aeean dary forest followed directly in the wake of the failing eqosmibe oo ok now. Probably some breadfruit land passed to secondary Sepect wi thaus first having been planted to coconuts, the older breadfruit trees sen ba salvaged Gor eas. ) ~ Tn 1952 somé areas of poor coconuts were being invaded directly by Pestana. teens, whe Be ee ees More commonly the earliest eagedtene word Wedetia ert Evens, the latter forming thickets of whippy stems six to she feet tall. A bepresentative zonation was ob- served in the ato wun Karel near the ocean side of the secondary foreste ee The sequence was (1) poor coconuts with Fimbristylis and a black alga prominent on the seetnk (2) Wedelias (3) 6«foot Premas (1) Pipturus argenteus with an undergrowth of scattered Thuarea; (5) Randia cochinchinensis and Morinda citrifolia over which Ipomoea tuba climbed; (6) 12=foot Guettardas and (7) closed forest of Pandanus, Allophylus, Succi. and Guettardae te “ In Bas otosiee forest ae es young trees of which were abundant on the forest flats appeared to be increasing at ‘the eee of Premna and Gusttarda, Seedlings and saplings of the Premna were never observed growing under closed secondary forest, and small Guettarda 1 were common _ only in apeitinee and at the edge at the forest. Ochrosia gppost ie a a rare species eases near ocean shores, oer. in abundance at one place in the aay tes of Arno island, It appeared to be ie only species on the atoll capable of groving : in nits ou ¢ dense shade, The area under Boke is not likely rapidly to increase in Size, faeces for the Ochrosia fruits are large ane heavy and are “not usually disseninated far from the parent tree, The same may ioe said for the fruits of Her- nandia sonora. A single large Hemnandia | was found in a a patch of secondary forest near the ocean sida of Arno Ee but seedlings were observed only within 50 feet of, ane veunle of the pare trees Secondary forest on Arno occurred on Arno loamy sand, According to Stone, these soils were ere "under a native mixed broadleaf forest that was replaced in part by the indigenous agriculture and more or less completely by 'copra eulene Jt Thus their development cannot be related to the existing vegetation," The present author, however, encountered an immature Arno soil on the rim of an excavated yaraj pit in the breadfruit ea forest near Ine village, suggesting that the series can develop wider breadfruit trees, It appears fairly certain that a large part of the area of Arno loamy sand on Arno island, including the present secondary forest, eas under a form of indigenous agriculture for at least 130 to 300 years before the advent of "copra cultures" The history of ieee mag ‘be much longer. Since the Arno loamy sands occur chiefly in the Goterd dpe of the wider — -- the regions best suited for breadfruit and gona) culture beeause of Pe Low pe ge of the ground watery it is not un- reasonable to postulate that the | areas under soils of the Arno series in ngo2 were fttOns the first to we occupied by the Marshallese voyagers upon their arrival, ey the atolls With the establishment of the Marshallese agricul tures gee Soils were disturbed bi ue excavation of yaraj pits. The ee of Rees during typhoons ‘and lesser storms has been another factor tending to obscure the otic soils It should further be emphasized that there is no field evidence that the interior of eo island was ever Sarieuea by a native mixed nesacuees Torest. Present stands of wild native trees infrequently visited Dye _ humans are commonly characterized by the presence of seabirds and phosphatic sands or rock. Phosphatic areas indicative of former flocks boo eaars ane and nesting seabirds were entirely lacking on Aamo islands Indeed, Sa analysis of a sample of gaue aay sand from the area of poor coconuts on Arno showed a surprisingly low content of available _ phosphorus in the black surficial horizon. He reported ae 10 pounds per acre, about 10 percent oe that, Les mtne in the productive | groves. WR as ESE gat VEGETATIONAL CHANGE Processes ao teeeeenesié change in the vegetation of Arno atoll were not evident in 1952, It is true that brush of small native trees and shrubs tended to grow up in the coconut and preadfruit groves, but the cutting and burning activities of the Jand-holders kept pace with theese threats Soe in a few places, The encroachment of native secondary forest on the area of oer coconuts of Arno island has been discussed above, bat the invasion did not appear to be strikingly rapid, Special study was made of areas affected by the typhoons of 1905 and 1918, In the pees co een ened oe during the typhoon of 1918 (cf. Wells, 1951; figs 8), Scaevola, Tournefortia, Pandanus ; Terminalia, Vigna, Canavalia, Triumfetta, Fimbristylis, Lepturus, Euphorbia, Polypodium, and Cassytha hate merce under the youmg coconut resi One Scaevola plant growing in a washed~out depression had attained a height of about 16 fect, The Wiehe wae Gopecta rie arene ane the herbaceous plants, Its apparent competitive ability in this situation is perhaps not surprising in view of the relatively low nitrogen eonhent of the immature soil (Stone, 19513; profile No. 27). Walking from Langar eee to Ine village I noted young plants of the following species growing in sand deposited since the typhoon of 1905 devastated much of the eastern part of Arno atoll: Cocos, Scaevola, Calophyllum, Pandanus, Guettarda, Tournefortia, Lepturus, Vigna, Triumfetta, _ Fimbristylis, Wedelias In most places, the sand which had drifted in between parallel ridges of beachrock representing the shore lines of former islands was devoid of vegetation. Yet it was possible to walk the entire distance from Langar to the southern end of Aljatuen Matolen on dry land at mid and high tide, Primary succession undoubtedly takes place, but it is apparently a slow process in such situations. .Pemphis acidula, however, was especially abundant along the lagoon shores of this devastated sa ants of all sizes up..to low trees 15 to 20 feet tall were observed growing on the solid beach rock. Shallow pockets in the rock in which small quantities of sand had accumulated were the seedbeds of this hardy plant. Scaevola frutescens and sprouting coconuts formed thickets along some of the high boulder - «ramparts behind the shrubbery of Pemphis. Ubiquitous was a black blue~ - green alga which covered completely the rock fragments of boulder ram- parts in exposed situations. Further confirmation of the hypothesis that vegetational change was slow at Arno in 1952 is afforded.by the fact that it was possible to use aerial photographs taken in 19) and 195 in mapping the vegetation of the atoll. The photographs were repeatedly checked in the field. and no discrepancies were noted. other. than. slight EU gees land forms, The vegetation of Arno atoll had remained essentially stable for at least Ts years o PHOSPHORUS RELATIONS The atoll habitat, monotonously miform to superficial observers, is actually striking in its variety. Permanent residents, for example, des ToL Sere of the differences in the abundance of fish and sharks in different situations. On Arno atoll the fishing was especially good off Bikarej island. Although there was an abundance of fish on the ocean reefs near the Takleb ee ee there was dangerous because of the great concentration of sharks. The lagoon shores of the windward islands of Arno atoll were often wmpleasantly hot because the coconut tee plantings to windward cut off the ectatie trades, The Lagoon sibee of Aro island and Ine, on the other hand, were for the most pat ladon ly cool, for the trade winds impinged on the villages along their lagoon shores after an uninterrupted sweep across the broad lagoons Mason (1952) discussed the striking Sra a SEBS poe to be found on the aitferent islands of Arno atoll. ig peared that almost every island had a ‘Local "won der" or legend associated se £ The atoll was in effect | a microcosms _ Marked differences in the habitats available to Slants + were a coud. Soils varied in texture from peats and mucks through snap loams to mereetais conmaeee chiefly of ouisnes and cobbles. Some soils, Like those of highs wander? dunes y SNe eee excessively drained, but else-~ where eeahenater swamps were encountered, The salinity of the ground water varied over a 2,000~fold range, and this variation was reflected _in the vegetation, Breadfruit and yaraj plantings were Limited to areas ofy relatively low salinity. | | | The variation in concentration of readily extractable ERSEprer Ee in the soils of Arno atoll was especially noteworthy, Stone gepeetes only 10 pounds per acre in the area of poor coconuts on Amo island,but 860 pounds per acre at one locality on Namwi island. Concentrations in the mangrove peats of Langar island reached 1600 pounds per acree Local concentrations of phosphates and chopRAanairele es at Amo were often correlated with present concentrations of roosting and nesting seabirds, The mangrove swamps of Langar and Tinak harbored large colonies of fish-eating terns, the droppings of which are believed to be rich sources of phosphates. Takleb and Namwi islands, formerly Marshallese bird reservations, Hee local deposits of. phosphatic sand and rock, - 56 # Although imeonsélidated phosphatic sands were Ota tatty neg in areas of native forest dominated by Ochrosia, Guettarda, Intsia, Barringtonia, or Hernandia, corisolidated phosphatic limestone was ap- parently restricted eiuaiescher ses, Sues forest in which Pisonia grandis was or clearly formerly had been the primary species. On ~ Takleb island, for example, phosphatic rock was foumd only on the part of ie eaten sadencere grew 3 large Pisonia trees, probably relicts of the phpincs pouiye eoneake” In order to study this phosphatic sit the "author eeeeecd a deep pit near the largest of the Pisonia trees, The eer ie wae: 0 a2 inches, but extending to a depth on L inches in’ “fissures in the phosphatic rock: Black greasy hunus with Pisonia roots. Crumb structure well developed when dry. Mucky when wet. 2 Whe inches but commonly outeroppine at® surface and in places continwing to depths Sf an ashes: Brown phosphatic limestone, The rock wes highly pee meen eaetelst p and pockets ef wiconsolidated phosphatic gand vere peeeene. : 13 e202 haat oat coral and foram sand with vertical brow phosphatic streaks ona enone of brown phosphatic rock. Grading into’ 22 52 inches: pink ee) aie foram ed, eradually increasing in Mibanstheds of texture with depth, and containing near the bottom of pit well preserved eno on branched corals, it This soil eae developed on a parent material of coral-and~ foram sand similar to that found in the 2252 inch horizon. Phosphatic ‘salts from bird ce aoeen es werden er from the humus layer into the Lime sands beneath, powbeniee with these to form insoluble calcium ‘phosphate. “Movement downward of the phosphatic salts was effectively - 57 - limited to the upper 22 inches of soil by the filter of lime sandse Gradually a matrix of calcium phosphate formed around the grains of limesand, cementing them into solid chunks of roa ‘The mucky phosphatic humus washed down into the interstices of this rock, in places preventing the further downward movement of materials. ivcuae the largest of the Pisonia trees stood 3 pools of water an then or two deep, which served as wallows for the numerous pigs of Takleb island. Qne of these wallows was situated only 11 feet from the soil pit | described aboves Standing water remained in this wallow at the end of a 36-hour rainless period, whereas no water was encountered in the pit at a depth of 52 inches, Evidently this phosphatic hardpan was capable of perching a water table, | A fragment of rock removed during the excavation of the soil pit was suggestive of the length of time necessary for the formation of the hardpans) 1 One side contained a semi-cylindrical cavity approximately 2 inches in diameter. The rock was hardest and darkest around the cavity, which Se odereiy had once been occupied by the root of a tree, It is suggested that the rock must have formed around the root after the latter had attained a diameter of 2 inches, for had the root penetrated the rock after the latter had formed, it surely would have shattered the brittle rock during its increase in diameters Further, the rock must have formed around the root while the latter was still alive or only recently dead. Coarse, unconsolidated sand would quickly have filled in the cavity left by the aecayine root had the latter rotted before the rock was formed. This suggests that the phosphatic rock formed very rapidly, perhaps within less than 100 years, although the length of life of the root of an unknown species of tree under the conditions described is of course not know, oe BO" Felix Je, the magistrate of Arno atoll, whose: family holds land rights on Takleb, ‘stated that the phosphatic rock was first noted in 1905. The island was first: settled in 1886, and certain crops not observed on the atoll in 1952 ~- the pineapple, tomato, an watermelon -- were ‘grow on the fertile Takleb soils wntil the development or exposure of the rock made their culture difficult or impossible. It is possible that . the removal of the wild trees tended to have a dessicating effect on the “gOilse During periods of drought the soil may have been exposed to the dire direct rays .of cuavitten ies? sun, and. the resultant "baking" may have ac- célerated the. formation of phosphatic ack! Qn the other hand, removal <-of the treés.may -have been followed by heavy ercsion during the torrential ‘downpours so characteristic of Arno atoll, The mucky layer of phosphatic _-‘humusy in’which the crops were grow, may simply have washed away, ex~ ‘posing the previously formed phosphatic limestone beneath, “For the most part, productive coconut trees: occurred on soils of intermediate concentrations of phosphates: 80 to 120 pounds per acre according: to Stonets analysis. Although coconuts were successful on ' these ‘soils,: the special crops: += tomatoes, watermelons, pineapples -- in general were nots’ ‘In 1952 Kapok trees had been successfully introduced only on: Taklebs They were stated to have failed on Ine. //. Qn soils of very low concéntrations of: readily extractable phos- phorus even the coconut did.poorly, _In.the interior of Arno island (Fig. 13) where Stone's. analysis showed-only 10 pounds of readily extractable -phosphorus per.acre’in the surface 6 inches of the soil, was an area of “poor coconuts ("laora") forming a semicircular belt. around an-area of "secondary forest; Coconuts observed in clearings of the latter exhibited wo 59 = the same symptoms as those in the area of poor coconuts adjacent to the secondary forest. It is therefore probable that the condition responsible for the poor growth of coconuts was present throughout the secondary forests Combined, the two areas totaled 136,6 acres. Poor coconuts of the "laora" type were also observed by the author in the interior of the widest island of Ebon atoll, in the southern Marshalls. The soil, of which samples were sent to Dr. Stone for analysis, was apparently of the Ammo series. On both sides of this area of poor coconuts but no nearer the lagoon were groves of healthy, productive trees growing on phosphatic soils. These local pvhosphatic deposits, neither of which was larger than 10 acres in extent, had been mined by the Japanese wmtil United States bombers destrovéd the loading facilities; about 50,000 tons were removed, The author excavated a small soil pit in an area undisturbed by mining operations, A highly organic, fine=textured layer 8 inches thick covered 8 inches of gray sand, which in turn lay over solid, hard phosphatic limestone, through which it was impossible to dig with a hand shovel. The fotiebion was apparently thick, however, for the Japanese excavations had pierced the water table and created a fresh=water pond perhaps 1000 square feet in areae To summarize, "Laora" is a disease of coconuts occurring in the interiors of the wider islands of atolls on soils of low concentrations of readily extractable phosphorus. (n local phosphate deposits in the interiors of one such wide island the coconut trees were healthy and productive, whereas the palms on an adjacent, non-phosphatic area were not thrifty. The sites were apparently identical in all respects with the exception of the difference in phosphate concentration of the soil. - 60 = The presence or absence of local phosphate deposits in the interiors of wide islands is provably not fortuitous. At Arno the tops of large Pisonia trees were invariably crowded with flocks of roosting and nesting seaoirds, In fact, the birds appeared to "prefer" the branches of the Pisonia to those of any other tree, The droppings of these myriads of birds would eventvally result in the formation of phosphatic rock, as ob-= served on Takleb island, Like many of the native trees of the southern Marshalls, Pisonia grandis tends to occur in groves -~ that isy in pure | stands less than 10 acres in size» Present local deposits of phosphatic rock may well coincide with former Pisonia groves. The Marshallese voyagers; upon arriving at a new atoll, probably cleared first the forests of the wider islands, In the interiors of these islands the ground water was sufficiently fresh to permit the growth of their basic crops, breadfruit and yaraj. At Arno island yaraj pits and abandoned housesites were concentrated in the eee it is probable that this interior land has had a longer history of con- tinuous human use than any other area of the atoll. From the time it was first cleared wntil 1952 wn at least 220 years and probably much longer =- large flocks of fish-eating seabirds were absent, for the birds tended to avoid villages, where they were persecutede Consequently, any large-scale deposition of phosphates by seabirds effectively ceased in the interior of Arno island with the arrival and establishment of the Marshallese voyagerse Tt is perhaps not tinea ies that after centuries of cropping and leaching losses some of these soils ie not able » supply the coconut tree with sufficient phosphorus to meet its needs, to OM According 3 the data of Walker (1906), ‘a ripe coconut minus the ee and shell Sot G A ov ote aids ene erset (Po0y) On the average about 1850 nuts per ar tae coconut land were converted into copra on Amo atoll in 1951. This is about 3.41 pomds of phosphorus (p) removed per acre in coprac Th age en apane 0.8 pounds of phosphorus was removed from coconut plantations in the shells of the husked nuts, ngleede a total of 1.2 pounds per acre, About 3.8 tons of phosphorus were lost to the atoll in 1951 in the export of copra, The shells , of course, were not wee but were used as fuel locally, Using the data of Bachofen Ghusaaneisy 1900) the eee of phosphorus lost to the atoli in 1951 in the exsort of copra drops to 1,37 tons; the loss per acre becomes 1.3 pounds. Of course, in order to arrive ata more accurate estimate it ahat a be necessary to ameter Arno coconuts or Bag let The lis fee merely ioe nea to indicate theorder of magnitude of the crorping ngeee Indeed, with the abandonnat of the unproductive groves and the establishment of secondary forest, the early stages of a shifting 2» asriculture 6 were ‘present in 1 1952, Only the establishment of isonia groves and rookeries within the sueendeee pores were necessary to Gautie : the cycles i Raa d ve appear that the cycle of movement of shOepnoene rem Seats iene! and back is of considerable importance in See? Soret ea ote nae oe oF SRIaBited Guo Sw ane eee eneeneeehat tel 1 pee tie a atl tae geochemical study of the i aeaigcs ie on in~ habited abit ath ae Wore Re ee Soest eve es general interest, for it dont cabin a ane of the chemist, the soils scientist, the geologist, the Veli deea ae the Sheree the botanist~agriculturist, and the anthropologist in the investigation of a vital phase of the economy of atoliss, “62 = HISTORICAL RESUME 1, Before the arrival of the iershallese voyagers, the vegetation of Arno: atoll was composed ine of ere, ta and shrubs, Along boulder ramparts and extending perhaps 100 feet inland on wind~ ward islands was. a scrub vegetation. composed principally of Scaevola frutescens but also. including Tournefortia argentea, Pandanus tectorius, Terminalia sailoensis, and Guettarda speciosa. This scrub merged inland on stony soils with a forest of trees 20 to 60 feet. tall. Prominent among these were Barringtonia asiatica, Hernandia sonora, Ochrosia cppositifolia, Intcia bijuga, Pandanus tectorius, Guettarda speciosa, Pisonia grandis, and Cordia subcordata, Ground. cover was sparse, being featured chiefly by tree seedlings and suckers, clumps of Asplenium nidus among the rocks, and patches of Peperomia spe On sandier soils nearer the lagoon, groves of tall Pisonia trees served as rookeries for numerous fisheeating seabirds. Hardpans of phosphatic limestones . developed under the Pisonias, Allophylus timorensis, Guettarda speciosa, imisia, bijuga, Pipturus argenteus, Cordia subcordata, and Prema obtusifolia were also present on sandy interiors, although they probably did not form mixed stands with the Pisonias. Along the dunes of lagoon shores a scrub of Scaevola with occasional Suriana maritima and Sophora tomentosa formed a narrow belt at the edge of the forest, Along eroding lagoon shorelines in which beachrock was exposed, Pemphis acidula formed the bulk of the vegetations, Saline flats were covered with nearly pure stands of Pemphis. Saline “swamps may have been occupied by a vegetation of mangrove trees, ad though it is possible that these were introduced later by the Marshallese. - 63 ~ Fresh-water swamps were covered by forests of Pandanus tectorius. Along the lagoon shores of island damaged by typhoons herbaceous plants such as Triumfetta procumbens, Lepturus repens, Wedelia biflora, Vigna marina, Fleurya ruderalis, and Fimbristylis atollensis maintained precarious footholde tn the face of the advancing woody vegetation, 2. The arrival of the Marshallese settlers introduced a period of major change in ee ee The forests of the sandy interiors of islands were partially cleared, gerne due cultivated, and breadfruit and coconut trees planted, Perhaps 15 percent of the land area of thé atoll was given over to agricultures -Breadfruit groves flourished in the in- teriors of the wide islands an coconut treés “lined -their lagoon shores. 3. Housesites in the interior of Arno Teanepoes abandoned in the middle of the 19th century after having been occupied for at least 120 and perhaps 300 or more years. An Hawaiian mission was established at Ine in 1873, marking ime first recorded date in Arno?s history, In- 1876 ot Lekaman removed the restrictions on colonization of the only two sizeable islands of Arno atoll ~ Takleb and Nami == on which the | native vegetation eee! maltered or nearly so; much of the native forest was removed to make way for breadfruits, coconuts, and other crops, ‘Germany established a protectorate over the Marshall Tslands in = 1885, German administrators and traders encouraged the production of coprae ‘Pigs were introduced. ~. | he ae the early part of the 20th century the acreage under coconuts increased rapidly at the expense of the native scrub and forest. Native herbs, however, prospered in the new habitat. Pigs devastated the yaraj pits and uprooted Tacca plants in the coconut groves, The breadfruit groves of the interior of Arno island with their yaraj pits were abandoned, b = @h > and a native forest of Allophylus, Guettarda, Prema, and Pandanus developed in which certain varieties of breadfruit persisted -as scattered trees, reproducing by root suckers and seeds,. A typhoon did serious damage to the islands of the eastern part of the atoll in 1905.6 5. Soon after the beginning of the first World War the Japanese replaced the Germans as administrators of the Marshall Islands, The Marshallese were encouraged to purchase Japanese exports, especially rice. A destructive typhoon hit Arno atoll in 1918, washing out parts of islands and tproatane trees, Japanese agriculturists advised the Marshallese to increase their coconut plantings, Plantations were wsuccessful, however, in the interior of Arno island on the lands surrounding the site of the ancient villages Copra production in the Marshall Islands probably reached a maximum under the Japanese, 6. In 19) Kwajalein, Majuro, and Eniwetok were occupied by armed forces of the United States. Although the copra trade was disrupted, — Arno was not the scene of major military activity. Aerial photographs taken in 19) and 195 Povealed that about 69 percent of the land area of Arno atoll was under coconut trees, This area had not changed ap- preciably by 1952. Native forest and scrub were restricted chiefly to stony land and the ocean sides of windward islands. By 1951 copra production had increased to an output of 622 tons, but secondary forest continued to encroach on the area of poor coconuts of Arno island, Dry land was exposed at mid or high tide the entire length of the eastern part of Arno atoll from Langar island to Aljatuen Matolen, but plant colonization on the newly formed land was slow. Pemphis acidula, however, ~ 65 « had succeeded in establishing itself in large numbers on the exposed beach rock of the lagoon shore of this region and Scaevola bushes grew along the high boulder ramparts» a a Se eet os = — POLST rian ~ 66 o Anderson, De 1951. Anonymous, 1900. Child, Re 1950. Cooke, PeCo 19326 LITERATURE CITED The Plants of Arno Atoll, Marshall Islands. Atoll Res» Bull. No. 7: 1+, i-vii. A Coconut Analysis (by Dr. F. Blohaten’ Queensland Agr. Jour. 6: 297. Cocos. in the Wealth of India, Vol.II, Raw Materials, pp» 262-288, Cowcil of Scien- tific and Industrial Research, Delhi. Investigations on Coconuts and Coconut Products. Dept. Agr. Straits Settlements and Fed. Malay States Gen. Ser. No. 8, Kuala Lumpur e Copeland, £.Be 1931 The Coconut. Third edition, revised. Londons Cox, DeCe 1951 Fosberg, FeRs 1919 Marshall, JeTasJre Wason, Le 19525 Spoehr, Aa 19:9. The Hydyology of Arno Atoll, Marshall Islands. Atoll Res. Bulle No. 8: 1-29. Atoll vegetation and salinity, Pacific Sci. 3: 89492 1951. Vertebrate Ecology of Arno Atoll, Marshall Island, Atoll Res. Bull. Noo 3: 138, Anthropology=Geography Study of Arno Atoll, Marshall Islands. Atoll Res. Bull. No. ll: 1-21, Majuro, a Village in the Marshall Islandse Fieldiana: Anthropology 39: 1266, Stone, Esle, Jr. 1951. The Soils of Arno Atoll, Marshall Islands. —_ Ghee sae 86 Tobin, J.Ee 1952. Land Tenure in the Marshall Islands, Atoll Res. Bull, Noe 11: 1-36, Walker, HeSe 1906. The Coconut and its Relation to the Production of Coconut Oil. Philippine Jour, Scis 1: 56.82 Wells, J.Wo 1951, The Coral Reefs of Arno Atoll, Marshall Wester, P»Ge 1920. The Coconut Palm: Its Culture and Uses. Philippine Bur. Agre Bulls 53. «3 (6G) wi TNOLV ONYV JO NOILWLS9S/A dVW Adv "3uy é@ 3YNDIS Bn VEGETATION OF ARNO ATOLL | 05 1.0 STATUTE MILES COCONUT GROVE BREADFRUIT GROVE SGRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR COGONUTS:"MELLAL” POOR GOCONUTS: "LAORA™ RRR SECONDARY FOREST | € 3YNdI4 S3JTIW SLALVIS O1 GO 10 O ete the } — ern xT COCONUT GROVE ; &\ >? BREADFRUIT GROVE SCRUB FOREST SALINE FLAT MANGROVE SWAMP oO FRESH-WATER SWAMP POOR COGONUTS:"MELLAL" POOR GOGONUTS:"LAORA" NN EERE SECONDARY FOREST VEGETATION OF ARNO ATOLL lI te eee . ~ at ti alee iting wi — Te , 5 - oa a bt nein na Mie i ~ f } ay, ; 5 ‘ ’ bv AYNDI4S 1S340J AYVONO93S »VYOV1,:SLANODOD YOOd » 1V113W,,:'SLANODOD YOOd dJWYMS Y3SLVM-HS3Y¥S JWVMS SAOYONVA LV14 SNIT1VS 1S34¥03 ENYOS 4% VEGETATION OF ARNO ATOLL Il] Ol 0.5 1.0 STATUTE MILES GOCONUT GROVE _—— [EDIT] erenoenur onove SCRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR COGONUTS:"MELLAL” POOR COGONUTS:"LAORA" ERERSRSEREEE SECONDARY FOREST FIGURE 4 COGONUT GROVE BREADFRUIT GROVE SCRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR COCONUTS:"MELLAL" POOR COCONUTS:"LAORA" SECONDARY FOREST “" VEGETATION OF ARNO ATOLL lV COCONUT GROVE BREADFRUIT GROVE SCRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR COCONUTS:"MELLAL” POOR COCONUTS:"LAORA™ ss | oO Ol Os 10 STATUTE MILES SECONDARY FOREST FIGURES COCONUT GROVE S76 2 BREADFRUIT GROVE | CG SCRUB FOREST @ SALINE FLAT _ MANGROVE SWAMP FRESH-WATER SWAMP POOR GOGONUTS: "MELLAL" POOR GOGONUTS:"LAORA" SECONDARY FOREST i VEGETATION OF ARNO ATOLL V COGONUT GROVE BREADFRUIT GROVE SGRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR GOCONUTS:"MELLAL" 0 Ol 05 10 STATUTE MILES POOR GOCONUTS:"LAORA" SECONDARY FOREST FIGURE 6 2 aYNdIS 1S3¥03 AYVONOD3S uwVY¥OVT,, ‘SLNNODOD YOOd uw IVT1I3SN,:SLANODOD YOOd dWVMS Y3LVM-HS3u4 JWVMS SAQOYSNVAN S3TIW ALNLVLS O1 sO ane ial ‘ Z es VEGETATION OF ARNO ATOLL VI GOCONUT GROVE BREADFRUIT GROVE SCRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP RRR SECONDARY FOREST = ——— J y 10 STATUTE MILES POOR COGONUTS:"MELLAL” POOR GOGONUTS:"LAORA”™ VEGETATION OF ARNO ATOLL Vil COGONUT GROVE BREADFRUIT GROVE SCRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR GCOGONUTS:"MELLAL" POOR GOGONUTS:"LAORA" SECONDARY FOREST VEGETATION OF ARNO ATOLL VII =| COCONUT GROVE BREADFRUIT GROVE SCRUB FOREST SALINE FLAT 1.0 STATUTE MILES MANGROVE SWAMP FIGURE 8 FRESH-WATER SWAMP POOR GOCONUTS:"MELLAL" POOR GOCONUTS:"LAORA™ | SEGONDARY FOREST REE COGONUT GROVE BREADFRUIT GROVE SCRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR COGONUTS:"MELLAL" POOR GOGONUTS: "LAORA" SECONDARY FOREST ‘s | \ en ee eee ——oCDClU™té<“‘i‘OSOC;WWU€ULEGEUVATION OF ARNOUATOLL a SCRUB FOREST SALINE FLAT MANGROVE SWAMP POOR GOGONUTS:"MELLAL" POOR GOGONUTS: "LAORA" LRRRRRRRRERE SECONDARY FOREST FIGURE 9 UT GROVE FRUIT GROVE FOREST FLAT OVE SWAMP -WATER SWAMP COCONUTS:"MELLAL" COCONUTS:"LAORA" DARY FOREST 3 VEGETATION OF ARNO ATOLL IX GOGONUT GROVE BREADFRUIT GROVE SGRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR GOCONUTS:"MELLAL" POOR GOCONUTS:"LAORA" | UTE MILES 0 Ol 05 HOSSTPAWIUMTE IE RERRERRERH SECONDARY FOREST FIGURE 10 AAR f COCONUT GROVE BREADFRUIT GROVE SGRUB FOREST SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR COGONUTS: “MELLAL" POOR GOGONUTS: "LAORA" SECONDARY FOREST — ree * aw VEGETATION OF ARNO ATOLL X GOGONUT GROVE BREADFRUIT GROVE SGRUB FOREST Ss) FIGURE II a es ee] oO Ol 05s 1.0 STATUTE MILES SALINE FLAT MANGROVE SWAMP FRESH-WATER SWAMP POOR GOGONUTS: "MELLAL" POOR GOGONUTS: "LAORA® SECONDARY FOREST (LX3L NI NOILUNW1dX3) ‘S3aduL SAILVN NIVLYS9D JO JONVLYOdW! GNV NOILNGINLSIG ‘zi aNNOIS Ze G3LNVId VIVONCNOD VEIINONUS VANGIOV Sit#dW3d SISNSOWVS VIIVNINYSL VYVANY V3INVINOS SNS39S3INYF VIOARVOS +s wo;rmer —<—$—<—<—_— winn ’ ene oH waTEm Lo LAGOON ° ° PANDANUS TECTORIUS 1400 FEET GUETTARDA SPECIOSA ALLOPHYLUS TIMORENSIS PISONIA GRANDIS PIPTURUS ARGENTEUS INTSIA BIJUGA HERNANDIA SONORA PLANTED? Ni BARRINGTONIA ASIATICA OGHROSIA OPPOSITIFOLIA GORDIA SUBCORDATA MESSERSCHMIDIA ARGENTEA SGAEVOLA FRUTESCENS SOULAMEA AMARA TERMINALIA SAMOENSIS PEMPHIS ACIDULA BRUGUIERA ConvuGATA PLANTED i GNV17SI ONYY JO LYvVd 3GIM ‘€1 3auNdI4 a13/4 11v9 7 3SNOH $.3LL3F01 1334 003; O0u: 00S 6) 90¢ 1S3Y¥O4 8NyoS (1S34O3 AYVONOO9|3S NIHLIM) 2 Slid (VYVA JO VIN 4S3403 AYVONOD3S 1333353333393) 1S3¥04 Linusavays AA wW4¥OV7, “SLNNO909 YOod SIMNNOSO09 Goos fae LA. Productive Coconut grove. Ine. Fig. 15. Poor Coconuts ("Laora"). Arno 8/23/52.- Islend. Nearly half the palms are dead. 7/15/52. Fig. 16. Breadfruit grove. Ine. Ground layer is dense in open groves. 8/23/52. ; ere | 7 : | | : iia iy : aus - Se, : Met “ Ms a ete Z Did . a & * ris + t f ~ > b a ~ a | P = 3 x \ b ‘ie Er Fig. F bla 17. 19. Yaraj pit, Ebon island, Fig. 18. Guettarda on stony land Ebon atoll. 9/20/52. near rampart. Langar island. 8/15/52 Ochrosia stand on stony land, Bikarej island. The Fig. 20. Barringtonia stand, stony land, Langar island. trees are about 40 feet 8/15/52. wall. “WA5/526 PESOS TROT ITe MEE a ae a rte Oe pee eae ae — aw Fig. 21. Large Pisonia tree, Takleb Fig. 22. Pisonia trees, Takleb island. island. Tree was 28.4 feet in cir- The trees have originated from sprouts cumference and 80 feet tall. Note along a single fallen log, the remains numerous suckers. 8/5/52. of the roots of which are visible at Wetter 08/5/52. Fig. 23. Mangrove swamp. Compare forms of old trees at right with straight young poles on left. Langar island. 8/15/52. ie eA. Fresh-water swamp. The Fig. 25. Fresh-water pond and swamp. stand consists chiefly of Pandanus The water was about 24 inches deep tectorius. Ulien island. 7/16/52. when photographed. Ulien island. 7/16/52. Fig. 26. Secondary forest. Rank growth Fig. 27. Modern house. Note gravel of Pandanus tectorius in abandoned yaraj courtyard around house. Arno island. | pit. Arno island. 7/15/52 8/20/52. Fig. 28. Gravel in modern courtyard. Arno island. 8/20/52. Fig. 29. Gravel in courtyard of abandoned housesite. Near yaraj pits, Arno) island. 17/15/52. SO Gard AZPNTS No. 17 May 15, 1953 ATOLL RESEARCH BULLETIN 17. Handbook for Atoll Research (Second Preliminary Edition) edited by F. R. FosBERG and MARIE-HELENE SACHET TS ) ane Ay MAY 28 1963 }) Issued by LIBRARY wi a THE PACIFIC SCIENCE BOARD National Academy of Sciences—National Research Council Washington, D. C., U.S.A. ATOLL RESEARCH BULLETIN 17. Handbook for Atoll Research (Second Preliminary Edition) edited by F. R. Fosberg and Marie-Héléne Sachet Issued by THE PACIFIC SCIEYCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1953 ACKNOWLEDGEMENT It is a pleasure to commend the far-sighted policy of the Office of Naval Research, with its emphasis on basic research, as a result of which a grant has made possible the ccntinuation of the Coral Atoll Program of the Pacific Science Board. | It is of interest to note, historically, that much of the fundamental information on etolls of the Pacific was gathered by the U. S. Navy's South Pacific Exploring Expedition, over one hundred years ago, under the command of Captain Charles Wilkes. The continuing nature of such scientific interest by the Navy is shown by the support for the Pacific Science Board's research programs, CIMA, SIM, and ICCP, during the past six years. The Coral Atoll Program is a part of SIM. The preparation and issvance of this Bulletin is assisted by funds from Contract No. N7-onr-291, Task Order IV. The sole responsibility for all statements made by authors of papers in the Atoli Research Bulletin rests with them, and do not necessarily represent the views of the Pacific Science Board or the editors of the Builetin. ee en ee ee Fditorial Staff F. R. Fosberg, editor , M. H. Sachet, assistant editor Correspondence concerning the Atoli Research Bulletin should be addressed to the above c/o Pacific Science Board National Research Council 2i01 Constitution Avenue, N. W. Washington, 25, D. G., D.'S. A. PREFACE TO SECOND PRELIMINARY EDITION The "Preliminary trial edition" of this Handbook, issved in May, 1951, was unfinished, as some of the manuscripts had not come in by the deadline. The intention was to distribute only the few conies needed at that time and to issue the finished book when all the material was reacy. Since there was much delay in this, it was thought best to allow the authors a chance to revise their sections in light of subsequent experience and criticism, and to re-ecit the whole book. ‘this has been done, with the result that several chapters have been replaced entirely, others reorgenized, and several new sections added. In this edition there has been little editing of the content of manuscripts, but mostly correc- tion of punctuation and ambiguity of expression. Differences in style and differences of opinion on certain techniques and scientific points are those of the authors. Such differences of opinion merely reflect the fact that not everything is positively settled ebout coral atolis. In such cases, as weil as in anything else, experience of users pointing one way or the other, or which covld in any manner contribute to an improvement of the Handbook, should be brought to the attention of the editors. Any such notes will be kept on file until such time as a final edition of the Hand—— ‘book is written. © f The editors and the Pacific Science Board again wish to express their thanks to all who have contributed to this edition. = SS be INTRODUCTION {TO PRELIMINARY TRIAL EDITION _/ by F. R. Fosberg One of the projects recommended by the Honolulu Symposium” on Coral Atoll Research was the preparation.of a Handbook on Atoll Research, for use by future | Atoll Project expeditions, as well as by other parties and individuals who might be ina position to make observations Be interest to the Project. Obviously, it was too late to produce such a ierianaate in fini hed chaos in time for this summer's expedition to the Gilbert Islands. It was felt, though, that as much information.as possible: should be placed in the panda of those taking part in this expedition, and that this might be produced in the form of a preliminary trial edition of the handbook. Use of this during one season would doubtless bring out the spats where it most needed amplification or revision. The present feopbaphes series of essays is the result of an attempt to carry this OU A tentative outline was prepared and the sections were given to various workers whose specialties or experience qualified them to write preliminary discussions on short notice, and they were requested to write what they could in a very short time. It is to the great credit of those asked to cooperate that their contributions came in so promptly and so adequately orate the subjects treated, The contributions have been edited only for minor details of punctuation, | spelling, and grammar. Where my own experience has suggested that anything might be added with benefit to the user, since there was not time to correspond with the authors, additions have been made in square brackets thus: [....--Ed.} It must be emphasized that both the topics covered and the essays here presented are tentative and preliminary. The Handbook, in its ultimate form, will depend, for its quality and adequacy, on suggestions, corrections, and criticisms evoked from readers and users by this preliminary version. All such Suggestions will be welcomed and should be submitted in writing to the Pacific Science Board, Washington, D. C., or Honolulu. Although it was decided at Honolulu that Mr. E. H. Bryan, Jr., should prepare the Handbook, it was felt that it would be unfair to ask him to meet the deadline necessary for this preliminary edition, and that since Miss M. H. Sachet was in the employ of the Pacific Science Board, working on the biblio- graphy for the Atoll Project, she could logically be assigned the editing of this draft. Thus, much of the necessary detail will be taken care of in the task of editing the final Handbook. I wish to extend the sincere thanks of the editor and of the Pacific Seience Board to all who have so willingly given their time and written the sections that follow. is Organized by the Pacific Science Board of the National Research Council, with the cooperation of various Honolulu scientists, and held at the University of Hawaii on February 5 and 6, 1951. 2 re TABLE OF CONTENTS PREFACE INTRODUCTION TO PRELIMINARY TRIAL EDITION TABLE OF CONTENTS Chapter 1 -~ Geography INTRODUCTION TO THE GEOGRAPHY OF ATOLLS, by Bh. He Bryan, ers Chapter 2 -— Meteorology | NOTES ON THE COLLECTION OF METEOROLOGICAL DATA ON TROPICAL ISLANDS, by Luna B. Leopold. Chapter 3 -- Geology GEOLOGIC STUDIES OF ATOLLS, by J. W. Wells COLLECTING GEOLOGICAL DATA, by Martin Russell BEACHROCK, by K. O. Emery SUBMARINE GEOLOGY, by Ke. . O. Emery 19 20 USE OF HAND LEVEL AND. BRUNTON COMPASS FOR DETERMINING AND MAPPING MINOR TOPOGRAPHY, by C. K. Wentworth DESCRIBING SIZE GRAIN OF BEACH AND OTHER Rae eae MATERTALS , by C. K. Wentworth Chapter 4 -=- ede AIMS AND TECHNIQUES FOR GROUND-WATER INVESTIGATION, by D. C. -Cox © TECHNIQUES FOR SALINITY DETERMINATION, by D. C. Cox TECHNIQUES OF TIDAL ANALYSIS, by D. C. Cox Chapter 5 -— Soil Science SUGGESTIONS FOR STUDYING ATOLL SOILS, by Earl L. Stone, Jr, Chapter 6 -- Botany | DIRECTIONS FOR STUDYING VEGETATION AND FLORA, by F. R. Fosberg 1. Vegetation i. Plant “Collecting on Coral Atolls Tite Directcns for! Colvecting Information on Varieties of Economis Plants IV. List of Equipement for Collecting Plants on a Coral Atoll -iii- Ab 48 93 29 SUGGESTIONS FOR COLLECTING FUNGI, by Donald P. Rogers COLLECTION OF SOIL SAMPLES FOR THE RECOVERY OF AQUATIC PHYCOMYCETES, by F. K. Sparrow INSTRUCTIONS FOR COLLECTING ALGAE, by Maxwell S. Doty SUGGESTIONS FOR COLLECTING MODERN AND FOSSIL CALCAREOUS ALGAR, by J. Harlan Johnson Chapter 7 -—- Zoology SUGGESTIONS AS TO COLLECTING LAND VERTHBRATES ON CORAL ATOLLS, by J. T. Marshall, dr. an SUGGESTIONS FOR COLLECTING TERRESTRIAL INVERTEBRATES ON PACIFIC ISLANDS, by Robert L. Usinger THE BERLESE METHOD OF COLLECTING SMALL INSECTS AND OTHER ANIMALS, by Joseph B. E. Morrison COLLECTING MOLLUSKS ON AND AROUND ATOLLS, by Joseph P. E. Morrison METHODS OF COLLECTING MARINE INVERTEBRATES ON CORAL ATOLLS, by Robert W. Hiatt DIRECTIONS FOR COLLECTING, PRESERVING, AND SHIPPING _ FISHES, by Leonard P. Schultz COLLECTING SEROLOGICAL SAMPLES, by Alan Boyden Chapter 8 —- Marine Ecology INSTRUCTIONS FOR MARINE ECOLOGICAL WORK ON CORAL ATOLLS, by Robert W. Hiatt Chapter 9 -—— Anthropology ANTHROPOLOGY AND CORAL ATOLL FIELD RESEARCH, by Alexander Spoehr SUGGESTIONS FOR INVESTIGATING THE CULTURE OF ATOLL PECPLE, by Leonard Mason , A QUANTITATIVE STUDY OF CERTAIN ASPECTS OF THE MAN—LAND RELATIONSHIP IN MARSHALLESE ECONOMY AT ARNO ISLAND, by Leonard Mason and Harry Uyehara INVESTIGATION OF MATERIAL CULTURE, by K. P. Emory -iv- Page © 65 69 73 7h 78 90° 96 100 Page Chapter 10 -= Miscellaneous HINTS ON TROPICAL PHOTOGRAPHY, by C. J. Lathrop 123 HINTS ON LIVING ON A BOAT, by J. HE. Randall, Jr. 125 HINTS ON LIVING UNDER RESTRICTED CAMP CONDITIONS, by C. K. Wentworth 128 -V= Chapter 1 == Geography eh as enue TO THE GEOGRAPHY OF ATOLLS by Bo, peor Bryan, Jy, The study ef atolls, died we edt 4 by the Pacific Science Board of the National Research Council, has two broad objectives. The first is” to investigate total environmental interrelationships on atolls, The second is to apply the knowledge gained from such studies to improve the wellbeing of persons living upon atolls. At its best, an atoll presents a harsh environment in which to live. Numerous atolls are inhabited, and several, particularly in the Gilbert Islands Colony, are overpopulated, which adds to the difficulty. These atoll studies are being made in cooperation with a project of the South po pies Commission to improye human conditions on atolls, This handbook presents suggestions for various types of investigation on atolls. By way of introduction, I have been asked to summarize some general aspects of the subject: What is an atoll? What is the range of physical form? Where are atolls found? ‘hat are some of the theories of their formation? What general observations are needed concerning atolls? WHAT IS AN ATOLL? A "eoral reéf" is an accumulation of limestone principally derived from skeletons of marine animals and plants,especially corals, foraminifera, mollusks, echinoderms, and red and green algae. These may or may not be in their original location or position of growth. The structure -extends. from the sea bottom near or quite to the surface. Changes in sea level or elevation of the sea bottom may cause it to protrude above the present level of the ocean. Reefs occur chiefly as fringing reefs, barrier reefs, and atolls, In addition, there may be reef salto askin which do not fall into any one of these categories. Fringing reefs are platforms of reef which extend out a relatively short distance from the shores of both high and low islands, even continents, Such reefs are not at uniform depths below sea level and may be cut by channels, especially opposite the mouths of streams, for reef organisms do not develop in fresh or silty water, Barrier reefs lie at a greater distance from shore, with relatively deep water between. The Great Barrier Reef, for example, lies up to 90 miles off the coast of Australia, with soundings: of 20 fathoms in the inter- vening channel. Barrier reefs may be cut by channels, and there may be islets upon them, but the proximity to other land, which usually is high, distinguishes them from atolls. The barrier reef’ around the Truk group illustrates this point. Although the reef resembles an atoll rim, Truk is not an atoll because of the high islands within its lagoon, ade An atoll is a physiographic structure, composed of reef, broken reef rock, and sand, rising abruptly from the ocean, and typical zy en closing a shallow lagoon, which does not contain "high" islands, The popular conception is that an atoli rim is circular. This seldom is the case. More frequently it is angular or elliptical. There is great range of size, shape, and form, Some atolls are completely submerged, with no dry land. At the other extreme we find "raised atolls", the reef rim of which is raised to a considerable height above sea level, enclosing a dry lagoon basin. . Every intermediate form can be found. Some atolls have a great number of distinct islets (Kwajalein has. about, 95) 5 and numerous entrances or channels, On some atolls the islets may be connected by ridges of sand, with few lagoon entrances. The next form in this sequence (which I call the "doughnut ") has a continuous ring of land surrounding the lagoon, ‘which may vary from very salt (Sydney Island) to quite fresh (Swains Island), depending chiefly upon the rainfall and the permeability of the material of which the atoll‘is composed. The size of the enclosed lagoon becomes smaller and smaller, in our series, until we reach the "pancake" with no lagoon. On the "pancake" island the marginal rim and enclosed basin may | be distinct (as in Baker or Jarvis Islands) or the island may be quite flat 1 (Howland), or so covered with trees that the basin is not apparent (Vostok or Flint). The highest part of such islands generally is just behind the windward beach, and rarely exceeds about 20 to 30 feet, usually much less. Any greater’ hel ent would require an uplift of the fend, as is the case on Fais, Nauru, and Ocean Islands. The question is, when does an atoll cease to be one? The point is | - academic. One has to make his own def finition, depending upon the circumstances, | It was decided at an atoll symposium, held in Washington, D. C. and Honolulu, in 1951, that, for the purpose of our studies, we would include non-elevated limestone structures, with dry land, with or without a lagoon, which were not associated: with high land. We would exclude elevated atolls (reaching a height of more than 20: feet or so), sunken atolls (without dry and), barrier reefs surrounding high islands (such as Truk), and reef structures on fringing shelves or platforms adjacent to land masses (such as the Great Barrier Reef, the islands in the Bay of Batavia, TaN and the like») DISTRIBUTIO:; OF ATOLLS ) The’ writer has undertaken to compile a list of knéwn atolls for the “Pacific Science Board, ‘To date these number about 400. They are located chiefly in the Pacific and Indian’ oceans, most of them between the Tropics. The ones: furthest north are Kure and Midway, at the northwestern end of the Hawaiian chain (28° 25' North). Those furthest south (excepting . Pelsart _ island and Middleton and Elizabeth Reefs) are Oeno Sh Ducie (21,° BOt.S.) The greatest ‘concentration is’ in the Central Pacific Equatorial, Northern ‘Cook, Tokelau’ and Phoenix (together 35); Ellice (9), “Gilbert (16), : Marshall (33), Caroline (40), and isolated in the fereess Pacific (3), a "total of 136. The Tuamotu archipelago contains 75, with.4 more to the east and 7 to the west of it, There are 5 in the Hawaiiah chain;~8 in Fiji (not counting 4 host of shail circular reefs without’ and); 27 between Fiji and Australia, chiefly in the Coral Sea; 6 north of the Solomon Islands; 6 in the Lowisiades; 20 near the Admiralty Islands; 9 northwest of Australia; 15 in Indonesia; 41 in the central Indian Ocean (Laccadive 13, Maldive 22, bie f a3e and Chagos 6); and 26 in the western Indian Ocean. Only one has been recognized in the Atlantic ‘(Rocas Reef, off Brazil), There are 25 potential atolls in the Caribbean Sea, but whether or not they are comparable with those in the oe and Indian Ocean is in dispute. If we follow the custom of the taxonomist, the type locality of atolls would be the Maldive Islands , -where the native name atollon is used to designate the reef rings found there. These differ in structural detail from "typical" atolls in the Pacific. agi ACCEPTED IDEAS REGARDING REEFS AND ee Harold T.- Stearns (1946). has listed ut Bis factors ‘which now are eaneraey accepted by ae gh ae They . are: : "1. Reef building eérals thrive iy at moderate depths, usually less than 200 feet pee sea level in nari, fatty. clear ocean water. : 2. Nullipores f/ coralline mais F bind corals together and make a reef possible. In some places nullipore reefs exist nearly free of corals. 3. Sea level was lowered about 300 feet during the Pleistocene by the abstraction of water. for the ice caps. If all the ice on earth were melted today, sea water would rise about 150 feet, - 4. Changes in the configuration’ of the ocean basins and consequent changes of sea level have occured since Miocene time to an undetermined extent. 5. The Sialic (continental) islands bordering pain sot [ine cluding Melanesia and western Micronesia / are underlain by folded rocks. 6, No folded rocks are exposed in the Simatic (Gecante), ‘Yslands Z Polynesia, except pernaps Tonga and New Zealand_/ except for slight warping which occurs locally, usually in calderas. 7. Emerged reefs of Tertiary age are exposed in many islands. 8. Coral atolls are not generally formed by growth upward from crater and caldera rims. 9. Coral atolls rest on basements of eee or cener non= coralliferous rock, 10. Barrier and fringing reefs require shelves built with their own talus, antecedent platforms or shelves OF some other origin, or gently shelving shores of any origin. ll. The growth of. corals is relatively rapid under favorable con~ ditions, variously reported .from 20 to 90 feet in 1,000 years. 12. Very rapid submergence may drown a reef, . Se} | Yas ay : qt f Byy ede *.y¥ 75 ti ay ‘ 13. - Reefs at sea: level are killed by emergence, Se Wea canal pe aaa most: rapidly at their ‘outer edges in the breaker zone where the food and oxygen supply is greatest. 15. The lagoons of atolls are not caused by submarine solution." LE vidence for-some. solution. of limestone in the inteftidal zone recently has been pointed out, _ Dr. Stearns proceeds to examine the various theories which have been proposed to explain the formation of reefs and atoll s, and to develop some modifications of his own. He points out that the shelves and plat- forms upon which reefs have grown have been built’ by ‘volcanism, sedimentation, or diastrophism, or out by marine or subaerial erosion. Examples of all can be found, he says, and "the origin of many shelves can be determined by study of the geology of the adjacent land, but ‘much ‘drilling and e-opeie* work is needed to determine the origin of submerged platforns." THEORIES FOR THE FOR:ATION ¢ OF ATOLLS , ‘It is a the object of this summary to discuss the relative ie of the various theories for the formation of atolls which have been pro- posed. A brief listing of the principal theories is given panorinid to suggest .the kinds of. ebeenet tons which need to be made. In 1928, William Me Davis! monumental work on coral reefs was published. This summarized the theories for reefs and atolls to that. date. Many detailed observations on the subject have been made since then, and the resulting ideas should be outlined to bring the subject up to date. One of the more recent summaries, which presents a relatively unprejudiced comparison, is "The problem of coral reefs," by Harry S. Ladd and J. I. Tracey, Jre, (1949). Another recent summary is given by Ph. H. Kuenen in his "Marine geology," (1950). During the past 20 years, several other valuable contributions have been published (see bibliography, below.) . Contributions are being made au an accelerated ‘rate. In order of pao the intial theories have bias 1..-.In.1821, A. von Chamisso propésed a theory for atolls based on the natural growth of corals around the edge of submerged banks, observing that massive corals grow best around the seaward edge of reefs, while those within are killed or stunted, 2. Charles Darwints famous "subsidence theory" was published in 1842, It suggested that fringing reefs growing around a sinking volcanic island would build up around their outer edge to form a barrier reef, and that an atoll would result when the enclosed island had sunk entirely and its summit, probably buried by coral debris, was submerged beneath the lagoon. This theory was supported and expanded by James D. Dana and favored by Davis. bac, a 3, Carl Semper (1863), itn in Palau, suggested that reefs were formed by outgrowth on a rising foundation, and that the lagoon of an atoll was formed by solution of the enclosed living reef, e Discovery of Simoneed arate Ye in the sea prompted sir John Murray (1880) to go back to von Chamisso's original idea that atolls formed by the upgrowth of corals around the edge of still-standing, — aggraded submarine banks. Similar ideas were held by J. J. Rein, H. B. Guppy, and F, Wood Jones. W. J. Le Wharton suggested that the foundations could have been completely truncated by i ie aes or «incompletely truncated. be, dn 1910 (and later}, R.A. Daly developed his "glacial control theory," which supposed that during the glacial period, when sea level was lowered by the removal of sea water to form glaciers, platforms were formed, around the edges of which reefs were built up in the warmer seas as sealevel rose, following the melting of the glaciers. Up growth kept pace with . rise of sealevel, forming the present barrier reefs and atolls, 6. In.1928, W. M. Davis amassed a ramendous amount of evidence in ‘support of a combined subsidence and glacial. control theory, in which inhibition of coral growth by lowered temperatures in the marginal belts _ of the coral seas during glacial times played a great part. He favored a modified subsidence theory which did not involve regional subsidence of the sea bottom but individual subsidence of islands separately. 7. J. F. F. Umbgrove (1939 and 1947), unable to find evidence to support the glacial control theory in his studies of the reefs of Indo- nesia, described evidence for crustal movements during the Pleistocene. He stated that elevated reefs in that area were formed in late Tertiary or early Pleistocene, then uplifted and: eroded. He noted that other reefs could be explained by Darwin's subsidence a 8. J. E. Hoffmeister and H S. Ladd (1944) secs evidence for the antecedent—platform theory. - This was based chiefly-on studies in Tonga and Fiji, where there is land movement due. to folding. It suggests that volcanic cinder cones are eroded'to ai depth.of about 150 feet below sea level, and then the resulting platforms are built up by non-coralliferous deposits ane reef—building corals. can: commence: their growth, chiefly -around the periphery. This: theory obviates both subsidence and glacial control. Later studies in Micronesia, particularly as a result of deep drilling on Bikini atoll, have caused these See to modify or broaden their ideas, ; 9. In 1945 and 1946, Harold piv eae suggested that great eustatic. shifts of sea level’occured during Pliocene and Pleistocene time, as a result. of changes in the configuration of ocean basins. On sialic _. (continental) islands, he: said, reefs formed on rising foundations, caused by folds. On simatic (oceanic) islands of the central Pacific block, they developed on subsiding islands. The appropriate hypothesis was.used in each case. . a © 10, In 1947, Ph. H. Kuenen, elaborating Davis! ideas, combined the two chief theories into a "glacially controlled subsidence theory," suggesting that preglacial reefs were cut down to glacial sea level by chemical means, and that reefs grew around the edge of’such platforms. Summary: It- will be noted that the important-.theories fall into two main groups: those that involve a relative change of level between land and sea, and those which do not. Of the former group, some favor a rising foundation, others a sinking foundation, and others a sinking and later rising sea. The chief argument in the other camp seems to be how the foundation was formed and how it became a suitable habitat for the growth of reef organisms. There is good evidence that the sea shifted its level during the glacial period, and doubtless in previous eras. It. is also known that oceanic islands can settle as well as be built up, and that other islands, which are remnants of a more ancient land mass, can move up and down with the folding of. their foundations. The trend seems to be for geologists to cease their feuding and to agree that one theory can account for the formation of reefs and atolls . in one region, while those in another region may have another explanation, even a combination of theories. Professor Kuenen states that he is in complete agreement with the statement of Hoffmeister and Ladd (1935) that "Probably no single reef theory will explain all reefs. Certainly, recognition of the complexity of the problem is essential to its abtution! It does not belong within the realm.of any one subject, but requires the attention of scientists of antl fields, eacn gee: his share," DETAILED FEATURES OF ATOLLS Careful eudads of boltee such as those recently made at Bikini, are much needed. They show that "an atoll is a composite reef made up of | a number of distinctive reef types - windward reefs, leeward reefs, and lagoon reefs, Most of these are zoned relative to prevailing winds and currents. Organic growth, erosion, and deposition of sediment influence each of these zones, but in each the balance of forces operating at the present time may be roughly appraised. Some reef zones appear to be making headway against the sea, others appear to be essentially in a state. of equilibrium, while still others are being eroded. Organic growth is. the source of all the materials that make up an atoll. Some of the ~ skeletons of the reef—building organisms remain in position of growth after death, but a much larger proportion are broken up by physical and organic agencies to form sediments that are deposited on the seaward slopes or in the lagoons, So long as organisms live on an atoll, it continues to grow, but it grows mainly by the accumulation of classic sediments." (Ladd, Tracey, Wells, and Emery, 1950.) The above abstract summarizes the kind of study needed for more atolls. It scrutinizes an atoll zone by zone: the Lithothamnion ridge, the coral-algal zone, the reef flat, the beach; submarine cliffs, eroded edge, boulder ramparts; various parts of the lagoon, including the coral knolls, It attempts to evaluate the balance of forces operating on an atoll, which have produced these formations, TYPES OF INVESTIGATIONS ‘NEEDED Ladd and Tracey (1949) eae the ree iae promising fields in etait investigation. dL "Island eee ee er eedied aelaias of all types should be mapped geologically.".. This raised the need for adequate base maps of convenient scale and considerable accuracy, In cataloging the known atolls, the writer has noted that but few atolls, outside of Micronesia, are mapped or charted in any detail, or with any accuracy. The use of the aerial _photograph, so widely used in the Pacific during World War II, should be of tremendous value in this connection. 2. "Reef studies... ‘Additional ecological studies such as those “earried out on the Great Barrier and at. Bikini on organic productivity of reefs: are needed, a“ 3., "Lagoon studies. Mapping and coring of lagoon sediments yield data that are of great value in the interpretation of elevated limestones and of the cores and cuttings obtained from drilling. When such, information is combined with data obtained from dredging on the outer reef slopes, it is possible to make sound paleoecological interpretations and to determine the significance of the several types of limestone that occur in reef areas." Study is needed also of the internal constitution of the numerous ecuat knolls aeine from the lagoon floor, 4, Submarine geology. More details are’ needed regarding the structure of seamounts, discovered in considerable number in the Pacific. "Dredging and additional coring of these structures should give more clues to their origin and, perhaps, their age. Specifically, it is essential to learn if they have a hard rim with a central depression now. filled with sediment; to determine if the coating of recent’ sediments is thick, and if it contains pebbles of: hard rock suggesting an origin of wave erosion." "Studies of isolated banks in. the coral seas at 300 - 1, 000 feet below sea level would give needed data on the types ‘of sediments that accumulate and the kinds of organisms that live in reef areas at depths below the limit of reef coral growth." Do reece foundations. More Lo. and geophysical investigations are needed. Other subjects in need of investigation include: How do climatic factors tend to modify atolls? These include wind direction and velocity, force of tropical storms , Pens eatery and rainfall. What is the effect upon reefs and toils, of ocean ae, tsunami and other large waves, water Lempenatine and een tye a What are the geographic names in use by the native inhabitants for islets and localities on atolls? How has man modified atolls? He is known to have: removed sand and guano, dredged reefs and cut channels into lagoon tasins, What has been the effect of these? He has made large withdrawals of ground water, dug pits for cultivation, removed the natural vegetation. Items to note regarding human inhabitants include: the number of persons and their areal distribution; land utilization, including areas . used for residences, crops, exploitation of the lagoon, and the like, The general observations, listed above, together with the pease derectaons presented in the pages which follow, should help to make possible a study of total environmental interrelations on atolls, and result in improved conditions Tor those who make atolls and reef islands their home. BIBLIOGRAPHY: Recent summaries of data on Atolls and their formation, . Eicuali Davis, W. M., The corai reef problem, American Geographical Society, Pub. 9, i928. / Presents a very complete summary of theories prior to date. \, Gardiner, J. S., Goval reefs and atolls. Macmillan, oles / Detailed study_/. Kuenen, Ph.H., Geclozy of coral reefs. Snell ius Expecition, i ies ae results, 2, aie “Indenesian region, Hoffmeister, J. EH, & Ladd, H.S., The antecedent—platform pried Journal of Geology, 52: 388-102, Taig . : -+ The foundation of atolls: a discussion. Journal of Geology, 43: 653-665 3 1945. Stearns, H. T., An integration of coral-reef hypotheses. American Journal of Science, 244: 245-262, April 1946, Umbgrove, J. H. F., Coral reefs of the East Indies, Geological Society of America, Bulletin 58: 727-778, 1947. Kuenen, Ph. H., Two problems of marine geology: atolls and canyons. Verhandlingen der Koninklijke Nederlandsche Akademie van Wetenschappen, Afd. Natuurkinde, Tweede Sectie, 43 (3): 68 pp., 1947. Daly, R. A., Coral reefs — a review. American Journal of Science, 246:193- 207, 1948. . Emery, K. O., Submarine geology of Bikini atoll, Geological Society of America, Bulletin 59: 855-860, 1948, Ladd, H. S. and Tracey, J. I., Jr., The problem of coral reefs. Scientific Monthly, 69 (5), 1949. | ~9- Sargent, M. C. & Austin, T. S., Organic vroductivity of an atoll. American Geophysical Union Transactions, 30: 245-249, 1949. Kuenen, Ph.H., Marine geology. N.Y., John Wiley & Sons, 1950. Chapter 6: LiL~179, gives a detailed account of reef theories. / Ladd, H. S., Tracey, J. I., Jr., Wells, J. W., and Emery, K. 0O., Organic growth and sedimentation on an atoll, Journal of Geology, 58 (4): 410-425, 1950. Chapter 2 -—- Meteorology NOTES ON THE COLLECTION OF METEOROLOGICAL INFORMATION ON TROPICAL ISLANDS by Luna B,. Leopold There is considerable truth to the general belief that meteorological records must have considerable length to be useful for hydrologic and climatic work. On the other hand there are some areas of the world where meteorological information is so meager that a relatively small volume of observational data taken by a careful observer can be very useful for several purposes, The accompanying notes describe very briefly the kinds of information a scientist can easily gather in connection with other work. Emphasis is placed on observations of simple nature requiring the minimum of instrumentation, and on data which may prove useful for the interpretation of biologic and ecologic Boras Rainfall The mean annual rainfall over the open ocean is very poorly known over great areas of the world. Observations taken near the seashore in a location essentially free of orographic influence approximates closely the rainfall of the open ocean, The main difficulty is to choose a location which is free from orographic effects. On an island sufficiently small it is desirable to measure rainfall near both the leeward and windward shore, as well as on . a topographic high. On an atoli, observations at three such points might be made without undue trouble in connection with other work. If measurements cannot be made at more than one location, at least a series of observations should be taken at the base camp and the location should be described with care. To measure rainfall nearly any kind of a container can be quite satis-— factory though a standard gage is eight inches in diameter. It has been shown that even cylinders of one inch diameter generally approximate the catch in the standard gage. A very satisfactory gage may be made out of a large size tin can, approximately four inches in diameter and nine inches tall. If a large can is not available a standard tin can holding about two cups may be used. The can should be set on a post so that the open end stands 2) inches above the ground. The open end should be approximately horizontal. From the open end of the can a cone, sloping up and outward at 45°, should not inter- sect any obstruction such as a tree or house. As the above description implies, the most representative rain reading is obtained in a location as exposed as possible despite our intuitive feeling that a gage should be protected from the wind. The gage should be read about the same time every day, preferably in the morning at approximately 7 o'clock, A stick may be calibrated with pencil marks to show the depth of water in the can, The depth should be expressed in inches on the area exposed on the open end of the can. If a bucket is used which converges to smaller diameter at the bottom than at the top, it is Pe Uy Vea Po clear that the actual inches in the bottom of the can will be too large, If a glass jar is used in which the open diameter is somewhat smaller than the diameter at the base, the inches of depth in the jar will give too small a reading. The calibrated stick can be easily made to give the proper reading. If the gage cannot be read once a day the reading should be taken at uniform intervals of a4 week or a month, in which case a few drops of oil should be put in the can to reduce the evaporation. Wind Direction Wind direction should be observed at a relatively open place and should be recorded to at least 8 points on the compass and preferably 16. Because -there is a diurnal change of wind direction as a result of land and sea breezes, wind direction should be recorded both in the morning and evening. Wind direction measurement is facilitated by the construction of a simple vane which can be made out of a stick and flattened tin can tacked on the end. A nail should be driven in the stick so that it goes through the center of gravity of the vane. Such a simple vane can be constructed in a few minutes and mounted on the top of a post, the height of which should be theoretically 18 feet, but in practice need not be that high. If the vane is 8 feet above the ground it will be quite satisfactory. If such a vane is installed near the base camp and as many as 4 readings a day'can be taken, a much better picture of the land and sea breeze may be obtained. Wind speed may be estimated by the use of. the following scale: Force 10 a1. wil. 2s “Table 1 -- Beaufort Wind Scale for Observations on Land Stations Explanatory title Calm Light air Light breeze Gentle breeze Moderate breeze Fresh breeze , Strong breeze High wind Fresh gale Strong gale “Whole gale Storm Hurricane Specification for use Smoke rises vertically Direction of wind shown by smoke drift, not by wind vanes Wind felt on face; leaves rustle; ordinary vane moved by wind Leaves and small twigs in constant motion; wind extends light flag Raises dust and loose paper; small branches are moved Small trees in leaf begin to sway; wavelets formed on inland waters Large branches in motion; whistling heard in telegraph wires; umbrellas used with | difficulty Whole trees in motion; incon— venience felt when walking against wind Breaks twigs off trees, generally impedes progress Slight structural damage occurs (chimney pots and slates removed) Seldom experienced inland; trees uprooted; considerable structural damage occurs ':Very rarely experienced; accompanied by widespread damage Miles per hour 13 19-2h, 25-31 32-38 39-16 47-54, 55-63 niin bette 75 «1L3= Temperature A thermometer for measuring air temperature should be hung so as to be well ventilated and constantly in the shade. Readings should be made at the same hour each day, preferably in the morning if only one observation is made each day. It is observed that when one becomes acclimated to the tropics he becomes increasingly sensitive to small changes in the temperatures. Therefore, at the time of the temperature reading it would be of interest to note, even subjectively, the observed impression of the relative warmth of the day, using such terms as excessively hot, hot, warm, cool or very cool. The maximum temperature of the day will generally occur approximately at 2 P. hi. and the minimum temperature at about 5:30 A, M. If two readings a day are possible they should be made at these times. Measurements of dew are very desirable because data on dew are meager. Estimates of the amount of dew may be made by walking through a patch of grass or by shaking the leaves of a bush, Amount of dew should be recorded as heavy, medium, light or none. Estimates of dew should be made at the same time the temperature readings are made in the morning, Water Temperature At least weekly records should be made of the temperature of the ocean in a relatively exposed spot, not in a lagoon. Because ocean temperature changes are relatively small the ocean témperature should be read to the closest half degree F. or quarter degree centigrade. Observations should include a description of the place where ocean temperatures are made, Wind Direction Aloft The direction of winds aloft may be estimated by observing the direction of movement of different levels of clouds. Clouds may be classified in three classes; fair weather cumulus or trade wind cumulus are considered low clouds. Their height is generally about 2,000 feet and their tops may be 4,000 to 8,000 feet. Their direction shouid be estimated separately from clouds of other heights. Middle clouds in the tropics generally occur at 14,000 to 20,000 feet and may be identified by their relatively thin appearance. The most common type in the tropics are alto-cumulus which may be identified by — their resemblances to tradewind cumulus except they are arranged in some linear or repetitive pattern. Alto-cumulus are often described as small lambs-wool clouds. The high clouds are cirrus clouds which generally occur above elevations of 30,000 feet. They are generally identified by their wisp—like pattern, and the common mares—tail clouds are of the cirrus type. A record of the direction of movement of these three levels of clouds is more important than © the speed of the clouds, Diurnal changes in cloudiness should be noted, particularly the repetitive patterns such as afternoon thunderstorms and sea~breeze clouds caused by the meeting of trade wind and sea breeze, i Chapter 3 -= Geology GEOLOGIC STUDIES OF ATOLLS by J. W. Wells The significant features of atolls are the reefs, lagoon, and islands. The reef is the primary structure; the lagoon is enclosed by it; and the islands are transient features of secondary importance. The following outline of geologic features of atolls gives an idea of the data and materials to be collected and studied. iii eheets: A. Seaward Reefs. 1. Windward. 2. Leeward. Relations to prevailing winds and currents. Fasses. Algal ridges. Width of flats. Distribution and zonation of coral genera and species, and other hermatypic organisms over the reef. Outer slopes: surge channels and steepness of slopes, Correlation of reef types with classification of Ladd, Hoffmeister, and Tracey. Distribution, origin, and condition of niggerheads and rubble. Traverses and profiles to implement data. B. Lagoon reefs. Width of flats and development with relation to size and strength of lagoon waves. Lagoon margins and slopes. Source of sediment on flat and slope. Zonation of corals. C. Coral knolls. Pattern of distribution. Any relation to reefs and passes? II. Lagoon. Depth. Nature of bottom sediments. Coral zonation on slopes and over bottom, III. Islands. Distribution. Beaches. Boulder ridges. /depressions - Fa. / Beach con- glomerates and sandstones. Island sandstone. Dune ridges. Phosphate deposits. IV. Processes. A, Degradational. Effects of normal waves, swells, and local currents. Degradational beaches. Effects of typhoons: frequency; destruction of islands. Evidence of exposed beach conglomerates, sandstones, and island sandstones, bearing on former islands, Development of rubble tracts on lagoon and seaward reefs. Development of new islands subsequent to typhoon destruction. * ~1A san 7 Ges B. Aggradational. Movement of sediment over reefs into lagoon. Tidal deltas. Building of beaches. Source and composition of wind-—transported particles and development of dune ridges. C. Uplift and eustatic changes. Evidence of these? Data on all these points are important in the general problem of the origin and natural history of atoils, as well as the more specific problem of the recent history and relative development of a particular atoll. Well documented collections of corals and other hermatypic organisms dre” significant in the problem of the geographic distribution of the Indo- Pacific hermatypic fauna. ~libw COLLECTING GEOLOGICAL DATA by Martin Russell Many features of importance to the geologist will be seen by observers whose main interest may not be geology. Because geologists need more information on atoll composition, structure, and development, the non-geologist could make valuable contributions to the geological sciences if he would record observations about easily recognized features as he pursues his own particular scientific interest. Examples of the significant geological features of atolls have been described in one form or another, But, like words in a dictionary, though they may be known, they have little significance until they are fitted intelligently into a coherent story, by a writer in the case of words, by an intelligent observer and scientist in the case of geological features. If you would record what you see, you could, with little effort, do much to help the geologist. A knowledge of certain basic features of an atoll is necessary in almost any field of atoll research. The chapter "Introduction to the Geography of Atolls" should make clear the names of these features, which become obvious to anyone after a few days on an atoll. Once understood, they should be freely used for orientation in making notes. I. Reefs A. Seaward reefs ois Windward ae Leeward B. Lagoon reefs C. Coral knolls II. Lagoon III. Islands Ae Seaward beach B. Lagoon beach C. Inland surface A notebook and pencil should be carried at all times and if a large-scale base map is available it may be used to record the location of the features described. The following are the features to be seen that deserve a few lines of description in your notebook. One should not hesitate to report descriptions if similar features are observed in two or more locations. Some of the names of the features and the characteristics which should be noted are underlined. REEFS. The growth, extent, and form of the reefs depend to a large degree on the winds and water currents passing over them. The observer should start out on a reef an hour or more before low tide, noting width, which ranges from a few tens of yards to a mile or more. Commonly there is a distinct zonation. On windward reefs there may be a reddish 17 tinged algal (or "Lithothamnion") ridge a foot or more higher than the reef flat, upon which the waves break; note the steepness of the outer slope and the surge channels running perpendicular to the reef front through which the sea rushes forward then back with each wave. Step carefully here for the reef may be very cavernous or honeycombed and a false step or slip would result in a nasty fall or an unexpected bath. Immediately shoreward of the reef front may be a pavement—like flat covered with mossy algae, close inspection of which may disclose multitudes of tiny sand-grain sized, cream-colored Foraminifera, The calcareous shells, or tests, of these animals are swept back over the reef to form a considerable portion of the beach sands and loose floor of the reef and lagoon, Note their concentration in different sandy deposits through— out the atoll. Between the reef front and the shore may be many different reef environments which should be hoted: isolated coral growths separated by waist-deep, sandy bottomed water; shallow sandy areas; rough coral flats; and isolated channels or areas of deeper water, Note the different types of coral growth which commonly are restricted in distinct areas ae branching colonies, solid smooth heads, bluish palmate colonies, etc. On the leeward reefs, the picture may be much different, zonation may not be distinct, the reefs are commonly much narrower, and individual coral and algal forms may be more delicate and varied. Note the distribution and condition of reef blocks (negro heads) torn from the reef edge by tropic storms and cast upon the reef flat. _ Anywhere on the reefs may occur sand or rock groins, platform-like structures which stand clearly above the level of the general reef flat. Any measurements of their height (above the general reef flat, in feet) and notes on their distribution, and orientation would be a valuable aid in determining the extent of geologic uplift and/or change in sea level. Most atolls have distinct breaks oF passes in the. continuity of the reef ring. Distinguish whether these are clearly breaks with no reef growth or mereiy parts of the reef over which the water is deep enough for navigation. On parts of the reef, commonly near islands, some currents form sand bars, and spits curving away from islands, Be sure to note, if possible, any difference in their size, shape, or location from those shown on any map you may have. The reefs on the lagoon side may exhibit different forms around the atoll depending on the size and strength of waves generated within the lagoon. It is generally found that lagoon reefs on the windward side of the atoll (but leeward of their reef or island) are less developed than lagoon reefs on the opposite side where the waves generated by winds sweeping across the lagoon are sufficient to encourage extensive coral growth. Here again any zonation of reef type, observable features of the reef margins, slopes and character of sediments should be noted, ‘Should you be fortunate enough to fly over the atoll and its lagoon watch for coral knolls, which are masses of coral that grow to within a few feet or tens of feet of the surface of the water. Note their size, shape, relative numbers, and pattern of distribution if any. Note in what relation to passes, islands, or abrupt change in direction of the reef ring these knolls occur, ISLANDS. Beachrock, a common feature of the intertidal zone, generally dips (angle between a horizontal plane and the layered surface of the rock) toward the open water, a reasonably definite indication of the slope of the beach sands from which the beachrock was formed. It follows that should you record beachrock dipping toward the island instead of the open water, that there has been a shifting of the islands (by storms, currents, wind, etc.). In some places on the reef ring, the existence and even outlines of former islands can be inferred from lines of beachrock, now submerged, out on the open reef and between present islands. A feature generally found only on the seaward beaches of leeward islands is the boulder rampart.* Immediately above the intertidal zone there may extend a mass of water-worn and rounded boulders, A cross section of this mass perpendicular to the shore may show two, three, or four "steps" or ridges. These are formed during the infrequent but particularly severe storms. | Other island features of note include a) mangrove swamps containing brackish water; b) guano, a light gray, powdery material filling crevices or encrusting rocks; from action of bird droppings or calcareous rocks} c) layers of cemented phosphatic sandstone, commonly 4 to 10 inches thick,formed from action of bird droppings on calcareous rocks, but occurring only a few inches beneath the humus mat of Pisonia grandis forest; d) any unusual depressions; e) wind-formed sand dunes or irregular ridges; and f) the depth and thickness of any hard, cemented layer encouraged in excavations on islands. Because calcium carbonate (limestone) is the only material deposited by the reef—building organisms, the occurrence of any other rock material (hard, soft, or loose) is of especial interest (pumice, for example). In describing the sands, rocks, gravel, and so forth, it is well to remember that four types of organisms are responsible for about 95% of the reef and island material on atolls. These are 1) hard or massive coralline algae (reef flats, reef front); 2) coral (colonies on reef flats, coral knolls); 3) segmented algae (the lagoon); and 4) Foraminifera (reef flat, lagoon) . * This is termed a boulder ridge or a beach ridge by some workers. There are several other uses of the term "rampart", especially the application to a ridge lying on the reef flat, with water behind it, by those who have studied the Great Barrier Reef islands, Australia. -- Ed. on “ . wate . may ’ , 4 oN ‘ eee mer ue “19s BEACHROCK by K. O. Emery Beachrock occurs around many islands of the Atlantic, ener: and Indian Oceans and even on the shores of some of the large lakes of the western United States. At Bikins Atoll, in the Marshall Islands, it covers perhaps 15 percent of the beach area, exhibiting no marked preference for seaward or lagoonward sides, It is practically restricted to the inter- tidal zone. Except for cementing calcite, it is the same in composition as the loose beach material tHat adjoins or overlies it: corals, Foraminifera, echinoid spines, shells, and calcareous algae. Usually, the grain size is also similar. to, ‘that of the beach material, and, just as in the beach, the sand grains in the beachrock form alternating coarse and fine-grained laminae from 1/16 to 1 inch thick. These laminae, as well as the surface of the beachrock, have, in places, the same slope toward the water as does the surface ofa one qeese beach sand; where the beaches are of cobbles, the beachrock is apt to exhibit a pai composition. In some areas the beachrock is collected and used as flagstones, building stones, and grave markers. It may form discontinuously, but. certainly some of it in Kwajalein Atoll is of very recent origin, because of the inclusion.of a Coca Cola bottle, shell cases, and other wartime debris. Although there are numerous published references to the presence of beachrock, little is known of the mechanics of its formation. One theory requires humic acids carried by rainwater to dissolve calcium carbonate from the interior of islands and to ‘deposit it again when the water seeps out through the beach, Another relates it to unknown biochemical processes. A few physical—chemical measurements made in a ‘loose sand beach at Bikini ~ ‘showed that the interstitial sea water a few feet below the surface was acid enough to dissolve calcium carbonate, whereas that near the surface was alkaline enough to cause precipitation. Upward capillary movement of inter- stitial water is known to exist from studies of other beaches. It is supposed that the water-bearing calcium bicarbonate in solution rises close to the surface, where the higher pH, higher daytime temperature, and high evaporation cause loss of carbon dioxide £00,_/ and precipitation of calcium carbonate as cement. The accuracy of this theory is not yet known; it is hoped that more data can be secured in other investigations. Measurements of CO,, alkalinity, pH, temperature, chlorinity, and calcium ion should be made on interstitial beach water at several depths over a 24-hour period to test this suggestion. Methods for some of these analyses are given by ~Doax Cox in connection with studies of ground water (see Chapter 4), More- over, an attempt should be made to observe, if possible, an area where beachrock is being actively formed. Perhaps. this would be indicated by ‘weak aneipient cementation, — Once deposited, the beachrock soon begins to be destroyed, Abrasion wears and polishes it to a shiny surface, often with flutings, and it also develops deep cylindrical pot holes. Wave action undermines it. Solution by sea water, charged with CO, from respiration of animals and plants at night dissolves some of the peachirock, forming shallow flat-bottomed solution basins that eventually join and deepen to develop a jagged surface. | ~20= SUBMARINE GEOLOGY by K. O. Emery Little is known of certain phases of the topography of coral atolls and still less is known of the distribution and types of their sediments. It would be highly desirable to obtain soundings and samples from atolls of the Gilbert, Caroline, and other island groups for comparison with the northern Marshalls, which have become relatively well known as a result of recent studies. Among the more important problems that: can ve eer with makeshift equipment are- ce ap Lsowene: le Steepness of olution shbpeds In the —e Marshalls’ ‘the® orn of the leewax -d reefs is very steep, locally vertical, to depths of 100 to- 200 feet.: Most of the windward reef, however, is bordered ‘by a’ slope of © about 15° to a depth of 20 to 30 feet, beyond which a terrace may Locally extend several hundred yards to depths of 40 to 50 feet. The greatest width is within broad lagoonward indentations of the reef. The terrace probably marks the position ofan old reef level, the geographical. distribution of which is unknown. “Soundings can be made during times of’ low waves from aboard a skiff, using ‘a light line marked: in feet and weighted at the end with a 5=poundl Piece of scrap iron or a rock. For the steep leeward reef, a distance out from the reef can best be determined by paying out another marked ~ | line that is attached to the reef, taking a sounding at each distance | interval of 2 to’5: feet, cee the. pesvard emeecey USS s a bay reef can best be estimated.” © ee tui en ' 2. Lagoon terrace.’ In the ndivehieusl uaieshdials the seaward terrace has its counterpart in a lagoon terrace that is several hundred yards wide, particularly within seaward ‘projections : ‘of the: ‘reef. Its depth ‘is 50 to 60 j feet. Sounding profiles across it can be obtained in the same manner as for a the seaward BEDEHES, but under easier wave: hss. E 36 Coral: masses. Bout teneat about the: aelesois of the northern Marshalls are roughly’conical coral knolls that range from less than 100 feet to more than a mile’in diameter. They rise from any ‘depth of the lagoon floor, and-their tops’may be at’'any depth, including sea level, At Canton and Johnston Islarids, howéver, ‘Snistéad of conical coral knolls, there are elongate ridges that: divide the: ‘lagoons * ‘into compartments. It would be of ; interest to know the geographical distribution of the two forms of coral ‘a0 growth of the lagoons, their composition, and details of their shapes. - esi) Such investigations’ may be conducted by simple sounding operations and by © observation with a.face mask or glass Ae ig box. 4. Lagoonal sediments, In the risen Marshalls the-sediments of the lagoon are arranged in a series of concentric belts.’ In water deeper than 180 feet the sediments consist chiefly of small Foraminifera. In shallower depths,. enough sunlight reaches the*bottom to | ‘permit growth: of a calcareous green alga, Halimeda. This plant’ grows faster:than the Small Foraminifera, so'at depths less than 180 féet it is domifiant, completely - masking the Foraminifera. Within a mile or two of the shore or reef and at depths of less than about 100 feet the Halimeda itself is masked by detrital calcium carbonate mostly of fine sand size, carried by currents from shore or from the reef, Along the shore the sandy beaches consist largely of Foraminifera again, but the types are different from those of the deep water, They have their origin in the sand mantle of the reef flat. Most of them are discoid or flattened forms and are 1 to 3 mm in diameter. Coral is concentrated in knolls and thickets at any depth, and is of various types. The fine sand is the most rapidly deposited of the sediments. The small maximum depth of small lagoons is probaoly a result of the rapid filling that takes place when the belt of fine sand reaches to or near the middle of the lagoon. Profiles of bottom samples across the lagoons of atolls in other areas of the Pacific would provide welcome new information. Such samples can be obtained with simple equipment. For example, a useful dredge can be made by attaching the bail of a stout bucket to a line with a 5 to 10 pound weight tied a foot or so above the bucket so that the lip of the bucket can cut into the sediment when dragged for a short distance across the bottom. A portion of the sample should be dried and placed in a cloth or paper bag labeled with the depth and position, The dredge can also be used to obtain samples from the upper part of the outer slope, thus providing more information on the differences between lagoonal and seaward sediments. The great scarcity of information on submarine geology of atolls makes every new fact of great value in interpreting the composition and history of coral reefs. bo kata, sii Yims Ott whodet we im. oGond gicer't sey tie’ hth Sth fy Boots BHe rer t Teed 83-36 wel bady (odd nt alae ahehes wtb at AR Gd She bi toh tena we Bs oMdgeb wie td etek), bow etlioed ak. ae. ia te. bod Ri é tab wp Laat Hong, ‘oot. ak agi, UREN Nidgilerg i hy Sapopal, thaws ta; pe loadir he) ott ee vi Litt tlie recy Sel: ad Per sation . tak Hay motor > aed Haat SuoSdew abbas pede ir adlont ‘grid, bo BHO pIBNe 0%. er pe: ite wits dity bowtedde we es a sagt pecete: te Lao etl? gotiondoa yd ae oe: Syd we oyadea Gk en dont rs bobs cay ® not bawesth weilw. danse thes Dy Sdith” dasa ' se POND ae bigods © re id wilt: ; MOPS LBRO Paria. 5 qedo. gadd- be. dred 4 ont SEOUL. meow easy athe ie ; 7 : eliole 20 waoloan anivgmiue mo. ao itante tt bo rye terty fh DAS AGLT hoon et aa ldewandal wt “Lev ieory 5 ya’. a | Toles aie oe Lalit eta anol USE OF HAND LEVEL AND BRUNTON COMPASS FOR DETERMINING AND MAPPING MINOR TOPOGRAPHY by C.K. Wentworth In the absence of surveying instruments and without precise methods, useful mapping of small details can be achieved and made more useful by attention to certain principles and practices. If a hand level (with sighting mirror) is available it should be checked against the ocean horizon from the beach for gross. errors. The observer's height to the eyes should be measured, and this value used. If horizontal distances per shot are not over 50 feet, successive hand levelling will give fairly accurate differences if care is used to mark or identify successive points. Even without a hand level,fair estimates can be made in places very near sq level if the ocean horizon can be seen and projected. When climbing a fairly steep bluff, differences of level can be measured within 2 or 3 feet in a hundred with a hand level, and for more gentle slopes and lower features results within 15 or 10 per cent by hand level will have greater validity than an off-hand guess, and should be so described in notes. Level lines of more than 50 feet per shot can be run by hand level but will be open to question unless checked and unless the observer has unusual skill and describes the precautions taken. Two short pieces of glass tube with 25 or 50 feet ef rubber tube wili make a more reliable levelling apparatus for projecting long shots (see that air bubbles are eliminated). A sighting compass such as the Brunton, commonly used by geologists and which includes a clinometer, is useful:in many ways and can provide a part of the control needed in making a fairly reliable sketch map. Nearly the same results can be obtained with any compass having sighting points or a square box and with a two- or three-inch needle that swings freely. Distances can be paced on smooth ground with fair consistency if the observer makes a sustained effort to calibrate his pace. For somewhat more accurate yet rough-and-ready quadrats, or traverses, a cord or rope with 10-foot knots or paint marks is convenient and often faster than a metal tape, even if the latter be available. Calibrate the markings occasionally if the cord gets wet. In using the compass, whether using the mirror at eye level, or viewing the needle from above, keep the box level and the needle free. Check the readings with yourself or other observers; form your own judgment as to whether your readings are reliable to 1 degree, or are in doubt up to 4 or 5 degrees. The latter may need some remedy in instrument or procedure. If the Brunton compass is used for dip or slope measurements, it should be so held that the axis of the pivot is normal to the direction of steepest slope and the sighting line should be either directly up or down that slope, or just at right-angles to it. Regardless of what instruments or devices are used, the accuracy and utility of a map is based on use of lengths and directions of lines and of angles and triangles arranged to make the map a firm, small-scale replica of the features on the ground. The larger the map, the more it needs reference to a Single long line, to one or more strong triangles, or to a well-defined grid of some sort. Any of these is preferable to a non-planned locating of minor features successively to each other. Any such sketch maps or sections should be made clear as notes by giving scale, compass directions and a legend or code of symbols; even the observer himseif may forget. 296 DESCRIBING SIZE GRADES OF BEACH AND OTHER SEDIMENTARY MATERIALS by C. K. Wentworth Lacking special equipment or the services of a specialist, the range of coarseness of sedimentary accumulations can be described for the record by systematic use of simple devices. The following scale of grade terms conforms to most general usage: Name of material - of grains boulders cobbles gravel fine gravel pebbles “very coarse sand | grains ecearse sand -_ medium sand fine sand very fine sand AVES ULE clay For extensive work sieves are necessary; it will, however, be very useful if casual observations are made specific and consistent with the above terms. For larger sizes, one can select and lay out samples showing approximately the above limits for more ready visualization in making detailed. notes.- Inter- mediate and smaller sizes can be laid on a grid background of quadrille paper for rough estimation megascopically or under a magnifier. Any essay of this sort is greatly superior to snap guesses which tend to become stereotyped without valid basis. Proportions are best stated by weights or volumes reduced to percentages (even when only estimated). _ Chapter , -~ Hydrology AIMS AND TECHNIQUES FOR GROUND-WATER INVESTIGATION Ra ch by Doak C. Cox Critical Factors: The aim of & gi round-water eat imnien on the islands of an eae is 68) indicate the quantity and quality of ground water available, and the variation in both in space and time, as a basis for determining the effectiveness of the ground water as an, ecological control. Generally, the fresh ground water will be found to rest on sea water. On Arno the. permeability of the reef. platform was so high that the fresh water body probably behaved according to the Ghyben- Herzberg principle that the ratio of the depth of fresh water below sea level to . the head, or height of the water-table above sea level, is the difference between - the dine Saale gravities of the fresh and salt water. Ascertaining the generality - of this principle should be one of the purposes of ground-water investigation. : The mean head is the most convenient index, though an imperfect one, to the. quantity of ground-water. Chlorinity, hardness and density are convenient’ indices to the quality. It is the job of the hydrologist to relate the magnitudes and variability of these indices to the magnitude and. variation of. the causative factors. On Arno the relations were worked out with considerable intensity. but. almost entirely for.one island only. A more extensive treatment seems desirable now, but will be attained with difficulty so far as the head is concerned because. the head can be usefully approximated only by rather intensive investigation. Salinity is easily investigated extensively, however, and on a dry island will have more interesting variability than on Amo.. It is, the salinity, moreover, that is most important as an ecological factor. More emphasis on the salinity than was made on Arno thus seems justifiable. he Measuring Points: Ground-water may be measured and sampled at any land surface depression in the land surface that venetrates the water table; in natural ponds, or in arti- ficial retting pits, taro pits, or wells. On Armo artificial pits were compara- tively plentiful but on drier islands where the ground-water will be of poorer | quality they will probably be scarcer. Where the land surface is less than 3 or 4 feet above sea level, as it is in much of the ,interior.part of atoll islets, it is simple and quick “to Gig pits to the water-table.. Such pits. should be small, a foot or so in diameter, to avoid capacity effects, and .it. would be best to. excavate such small pits on the side of the large excavations also, rather than using the large pits themselves for water-level readings. Where the ground surface is higher it will be more expeditious to drive or drill small diameter wells. Cheap drive pipe and weil points will, penetrate sand but will break at . joints when driven in coral limestone or beach rock. Drive pipe with special joints might be satisfactory, but simple wells can be driven to hard-rock by | excavating with a soil auger inside a 13" pipe, using hand drilling witha rock: bit and a small sledge inside the pipe for breaking up boulders. The coral con-'- glomerate and harder rock will not require casing so that when the pipe.is driven: through the sand layer the hole can be extended as needed with the hand drill. - and soil auger alone. water levels in such holes may be measured by lowering a stick or tape a measured distance into the hole and noting the level to which it is wetted. Samples can be withdrawn by suction ee a rubber tube or by | lowering an elongate cup. es Pe ee Measurements: The head is difficult to obtain with sufficient accuracy because of low water-table gradients and because of variation in sea level and well water level. Useable estimates of mean head require averaging of tidal variation and accurate leveling by telescopic level, transit, or alidade. Brunton or hand leveling is not good enough. Leveling could be probably done satisfactorily with a long (100 ft.?) rubber tube with glass tube ends, filled with water (and no bubbles) and a pocket tape or rule or two. Determination of tidal variation in a well is easiest done by a recording gage, but can be done very simply and well by observing frequently over at least a 24-hour period. Determination of tidal fluctuation in ocean is easiest and best done by recording gage of Coast & y | Geodetic Survey portable type. The use of the gage is described in a Manual of | Tide Observations, U. S. Coast & Geodetic Survey Spec. Pub. 196, 1941. This gage fits a 4-in. standpipe which must be at least 6 feet longer than expectable tidal fluctuation. Plastic pipe, very light-weight, now available, would prob- ably meet this need admirably. Without a recording gage frequent observations over at least a 24-hour period will suffice, but the water levels observed must have wind-waves damped out. Observations can be made in a partly submerged, vertical 2" plastic tube plugged at the bottom with a rubber stopper containing orifices from 1/8" to 1/2" depending on ratio of wave height to tide range. Or observations can be made by mercury manometer of the suction in a gallon jug partly filled with water and connected with the ocean by rubber tube filled with water. - Suction recorded in height of mercury can be reduced to height of water level in jug above temporary sea level. ‘The rubber tube furnishes the damping. (Wentworth: Wash. Acad. Sci. Jour., vol. 26, p. 347, 1936.) Mean head as measured above a singie day's ocean tide level will be subject to a considerable error owing to long period tides, only Lae: compensated by adjustment of Speed ane level. Salinity: measurement is treated in a separate section. Analysis of Head: In space, the mean head will increase with the width of the island, with the distance from the shores, with the permeability, and with the rainfall. The estimation of the mean rainfall is treated in a separate section. It will be nearly constant over an atoll. Measurements of head should be combined with observations of the size of the islands, with the positions of the measurements on the islands, and with whatever observations are possible on the distribution of rocks of varying permeabilities in the island, to permit analysis of the effects of these independent variables on the head. It should be noted that the head measured at any time or even over the longest period of observation permitted on an expedition will not be the mean head. Estimation of the mean” head will, depend on knowledge of the variability of the head and correction for it. The variability in head with time is caused by variation of recharge of the ground water body caused by variation in rainfall, variation in the rate of with. drawal (of minor importance on atoll islands), variation in barometric pressure (results negligible on atoll islands), and variation in the rate of discharge caused by variation in sea level. Rainfall measurement is treated in the section on gathering weather data. Rainfall correlation with head has not yet been done for the Arno data, so no tested rules of procedure can be laid down. ss Gyaas The correlation made must include recognition that the rainfall of a given interval of time has an effect on the head not only during that or the first succeeding interval, but also for a number of intervals, the effect presumably dying out in according to an exponential decay law. Before analyzing the rainfall effects on heads, it will generally be easiest to correct the observed heads for tidal effects through a tidal analysis or at least to obtain a mean for at least a day. Methods of analyzing the tidal fluctuations are given in a separate section. The parameters obtained from the tidal analysis will again be found to vary with the size of the island, the distance from the shores, and the rermeability. It should be pointed out here that the damping of tidal fluctuations is a function of their period. Fiuctuations with periods of two weeks or more pass across ground-water bodies in smali islands almost undamped, and as they commonly have amplitudes of at least several tenths of a foot, heads not corrected for them may differ very materially from true mean heads. Without fairly intensive tidai analysis the best head measurements are subject to errors iarge in proportion to the heads themselves. Analysis of Salinity: The salts in the ground-water are derived (1) from solution of salt crystals in the air or at the ground surface by the rain, (2) from the solution of minerals in the ground by the ground-water, and (3) from the mixing of hie fresh ground-water with the sea-water in which it is in contact at the botto edge of the fresh water body. The salts in the air and on the ground surface are derived from the evaporation of the sea-water spray. They should, therefore, show the same balance of composition as the sea-water. The chloride content of the water collected in natural or artificial rain-catchment structures may be used as an index of the total salts so dissolved. Density may be similarly used but with less precision, particularly in the low salinity range. The salts derived from mixture of salt water with the fresh water in the rocks should again show the same balance. The difference between the chloride content of the ground-water and that of the usual rain-catch water will provide a good index to the degree of this mixture. The degree of mixture will depend on.the thickness of the ground-water body and on the amount of fluctuation of the head, and therefore, dike the head and its fluctuation, on the amount of rainfall and its variability, and the size of the island, the distance of the point of observation from the coasts, and the permeability. Thus, the salinity alone wiil indicate the same set of conditions as the head, and the salinity is much easier to measure. On arno the rainfall was so great that except on very narrow islands or within a few hundred feet of the shore the chloride contents were so small that differ- ences from the chloride content in rain water were not significant. On drier atolls, larger islands should show interesting variations in chloride content to their centers. The technique for determining the chloride content of water is described in a separate section. ey The degree of solution of the rocks is also of very great interest from the standpoint of its importance in physiographic processes. The rocks of the atolls are almost entireiy limestone or perhaps limestone and dolomite. The hardness of the water, its content of calcium and magnesium ions, is an index to the importance of this solution, but the hardness attributable to sea water admixture must be subtracted. It seems a logical assumption that the hardness attributable to sea water admixture should be proportional to the chloride content. The ratio of hardness to chloride content should, therefore, be determined for the sea water and for the ground water. The hardness attributable to rock solution is therefore where the H's are hardness concentrations 7 Clis: are chloride concentr ations and the ee | S indicates the concentration resulting from rock solution G indicates total concenvrat ion in the ground-water O indicates Sinasiteeon in ocean water Ho/Clig will be about 1/4 or 1/3 A simple and accurate method of hardness determination is given in the section on salinity determination, which also includes a discussion of calcium hardness determination (analysis of calcium ion concentration alone) which should be interesting in studying he relative amounts of dolomite and limestone available for solution. Ecologic Controls: The study of atoll ground water is interesting not only on its own account | but because it provides certain ecologic controls on vegetation and man. Taro culture is almost certainly controiled ‘by the distribution of fresh ground water. _ Breadfruit is apparently so controlled on Arno. Influence of salinity on other economic plants may be found. Mangroves and some other plants require ground-— water in a very brackish range. The extension of recognition of salinity con- trois, and the investigation of their effect on the pattern of cultural adjust- ment of man to the various islands and on individual islands, constitute the chief reasons for support of ground-water work on Pacific Science Board projects... oye TECHNIQUES FOR SALINITY DETERMINATION by Doak C. Cox Chlorides: As discussed in the section on ground-water investigation, the saline constituent of chief interest is the chloride ion. The determination of chloride ion concentration depends on titration with silver nitrate solution - using potassium chromate as an indicator. Chloride concentrations encountered on an atoll may range from less than 10 parts per million to nearly 20,000 parts per million. In analyzing this very large range it will be helpful to have two concentrations of silver nitrate. The following reagents and equip~ ment will permit analysis of at least 100 samples in high and 100 samples in low Cl” range: i mg. Cl”) (4.791 g.per liter soln.) 1 pint AgN03 soln. (1 ml. in dark bottle. 1 pint AgNO, soln, (1 ml. = 10 mg. C1™) (47.91 g.per liter soln.) in dark bottle. Boz KoCrQ), soln.; part in 1 oz. dropper bottle. 2 pipettes, 5 ml. cap., graduated to 1/10 ml. 2 casseroles, porcelain, 100 ml, cap. 2 glass stirring rods. The reagents can be made up in any chemicai laboratory, and the equipment is available at any chemical supply house. Keep AgNO, soln. in dark to inhibit deterioration. The procedure is as follows: ‘1. Determine range of Cl™ content. by hydrometer (see discussion later) or by taste, or by testing first as if high range. . Range Taste Use AgNO, soln. 0-200 None 1 ml. = 1 mg. Ci 200~2 ,000 Flat or brackish TAN site a8 1G) aaah Ord ey 2000-20, 000 salty Gilute sample (see step 2) and use AgNOz 1 ml. = 10 mg, Ci~ 2. Measure 25 ml. of sample in graduated cylinder and transfer to casserole. To avoid excessive use of AgNO, soln. dilute samples containing more than 2,000 ppm. C1i™ as follows: Meastre 2.5 ml. sample with pipette (do not use this pipette for AgNO, solns.). Transfer to graduate. Dilute to 25 ml. with distilled or rain water. Use this diluted sample for test. If rain water is used it must be tested for Cl” and a correction made as indicated in step 6. 3. Add 5 drops KoCrO),, stir. Soln. will turn bright yellow. DG 4. Fill pipette above zero mark with appropriate AgNO, soln. Drain to zero mark, controlling flow with finger at top of pipette. add AgNO slowly from pipette to sample in casserole, with continual ne chess” yellow soln, will become turbid, very turbid if high chloride. Each drop of AgNO3 will make a brick red flash in the soln. in the casserole, which will disappear on stirring, As the end point is approached the red flashes will become larger and more | persistent. The end point is reached when the whole soln. in the casserole | acquires a faint permanent red tinge. It is sometimes helpful to pour only about, 23 to 24 mL. of the sample into the casserole at the start, titrate quickly past the endpoint, then add the remainder of the sample, reversing the endpoint , and titrate to the endpoint with care. 5. Read the amount of AgNO, soln, used from the pipette. 6. Compute the chloride content as follows: With dilute AgNO3 With concentrated AgNO (1 ml. AgNO3 soln, = 1 mg. C17) (1 ml. AgNO, soln. = 10 mg. C1) for 25 ml. sample ppm. Cl~ = ml. AgNO3 soln. x 40 ppm. Cl7 = ml. AgNO3 soln. x 400 limit of accuracy, equivalent of 1 drop AgNo3 soln. | Approx. 1.3 ppm. Approx. 13 ppm. With 1 to 10 dilution Determine Cl™ ppm. in test eante with concentrated AgNO3 as above. Cl- ppm. of original = ‘10 (CL™ ppm. of eee sample) - 9 (C17 ppm. of diluent water), Limit of accuracy approx. 130 ppm. Chloride may be reported as NaCl as follows: Chlorides as ppm. NaCl = 1.649 ppm. C17 Total hardness: As discussed in the section on ground-water, the hardness of the water is a | key to the amount of limestone it has dissolved. The following simple and accurate method for the determination of hardness depends on titration with an organic agent which segquesters calcium and magnesium ions, in an alkaline soln. using an organic indicator. Total hardness.is usually expressed as parts of ' CaCOz per million, both Mg+: and:Ca++ being treated as if they were Ca++ and combined with CO,-~. Total hardness encountered on an atoll may range from less than 10 to nearly 7,000 ppm. CaCO,. As in chloride determination, two concentra tions of the titrating reagent will be convenient. The following reagents will permit analysis of at least 100 samples in high and 100 samples in low range. ney Bs \ i pint hardness titrating solution. (1 ml. =1 mg. Cac0z ) 1 pint hardness titrating solution. (1 ml. = 10 mg. CaC03) 2 oz. hardness buffer reagent. (1 oz. in dropper bottle) 1 oz. hardness indicator. (in dropper bottle) ‘pHydrion paper will be useful in checking pH in buffering. The buffer reagent, indicator, and dilute titrating soiution may be obtained from W. H. & L. D. Betz, Philadelphia 24, Pa. -For instructions for making ail reagents see Betz and Noll, (Am. Water Works Assn. Jour., Vol. 42, p. 49, 1950). Equipment necessary is same as for chloride’ determination. Use pipette used in dilutions for hardness titrating soln. or carry a third pipette. The procedure is as follows: 1. Determine range of hardness. In general, hardness as CaCO3z will be in same range as Cl™ content except for sea water, which will have about 1/3 or 1/4 as much hardness as C17 content. - 2. Measure 25 ml, sample as in chloride determination. Dilute if necessary as in chloride determination. 3. Add 5 drops buffer reagent and stir. Solution should have pH of 8 or % Add 2 or 3 drops of hardness indicator and stir. Soln. will turn red. 4. Titrate with hardness titrating soln. in same manner as in titrating with AgNO. soln. for Cl” determination. As endpoint is approached solution will start changing from red to blue. Endpoint is reached with final discharge of red. 5. Read the amount of titrating soln. used. Add a drop more to the evaporating dish to ascertain that there is no further color change, 6. Compute total hardness as follows: With dilute hardness titrating soin. . With concentrated hardness titrating soln. (1 ml. = 1 mg. CaC03 ) es ml. = 10 mg.. CaC0z ) Hardness, as ppm. CaCO = ml.: Hardness, as ppm. CaCO, = ml. titrating soln. x 40 titrating soln. x 400 Limit of accuracy epprox. 1.3 ppm. “Limit accuracy approx. 13 ppm. With 1 to 10 dilution determine Caco, ppm. in test sample with concentrated titrating soln. as above. ii , Hardness as CaC03 ppm. of original = 10 (CaCc03 ppm. in test sample) -9 (CaC03z ppm. in diluent water) Limit of accuracy approx. 130 ppm. ate Calcium hardness: In the above method there is no separation of Ca‘'+ and Mg++ ions. They are lumped and the results computed as if they were Cat+. With the use of a different indicator and sodium hydroxide for pH control, calcium alone may be determined with the same titrating solutions as used in total hardness deter- mination. The indicator, a solid, is available from Betz in 50 gm. bottles with a measuring dipper. Equipment is the same as for total hardness deter- mination. About 8 oz. of NaOH 1.0 N are required. The procedure is as zortows 1, Determine range, Probably 1/4 to 42 total hardness, 2. Measure 25 ml, sample as above. Dilute if necessary. 3. Add 1 ml. NaQH 1,0 N and stir. Add 1/2 dipper of calcium | indicator and stir. Solution will turn salmon-pink. 4. Titrate with hardness titrating solution as above. Solution will turn purple as gE is approached. pT ‘is final change to 2 GbR BS 5. Read titrating soln. used as above. 6, . Baume Ca hardness as Caco in same manner as computing total hardness as Caco. Compute calcium ion if desired as foilows: Cat+ = .,400 Ca hardness as CacOz3, Compute magnesium ion if desired as follows: Mg hardness as CaCO, = (total hardness as CaCO, ) - (Ca++ as CaC03) Mg++ = .243 (Mg hardness as CaC03) | Calcium and magnesium hardness were not differentiated on Arno because the materials for this method could not be procured in time. Total salinity: In ocean water the ratio of the Eilers ‘content to al total salinity is nearly constant. Gi= = ,55 (Sverdrup, Johnson, Fleming, The Oceans, total salinity p. 166, 1942) The chlorinity is thus a convenient index to total salinity in any water diluted from the sea water if diluent has roughly the same balance of salt as sea water. This will not be the case if there is much Cal03 dissolved from the rocks in the water. In sea water or its dilutions, density also is a very convenient index to total salinity. Densities may be measured by hydrometers. «De. A set of three, as manufactured by G. M. Manufacturing Co. for the Coast and Geodetic Survey, with ranges 0.996 to 1.011, 1.010 to 1.022, and 1.020 to 1.030, covers the range of sea water to distilled water in most temperature ranges. A convenient cup for use with the hydrometers is available from Mercer Glassworks, 725 Broadway, New York 3, N. Y. The temperature of the water must be measured to 1° C., as a temperature correction is critical. Tables for the temperature correction to densities and density to salinity reduction are given in the Manual of Tide Observations, U. S. Coast and Geodetic Survey Spec. Pub. 196, 1941. Density measurements were not made on Arno because the one hydrometer carried broke. S33. TECHNIQUES OF TIDAL ANALYSIS by Doak C. Cox Introduction: As is further discussed in the section on ground-water investigations, the study of the head of a ground-water body in an atoll island must include some sort of analysis of the tidal fluctuations of both the water-table and sea level. Simple averaging of a day's water levels will suffice for some purposes. For P ging a : others, a more intensive analysis is required. The tide fiuctuation in a ground- | water body is itseif a key to conditions in the rocks. A study of this fluctuation depends upon separation and individual analysis of tidal components of different periods, because the periods affect the changes in the tides as they | progress from the ocean through the rocks. The techniques discussed here have not been published yet, but have been applied to a number of problems (Cox and Munk, Hawn. Acad. Sci. Proc. 1951, in preparation). Anaiysis of tidal components: Many tidal components of different periods go to make up the tide. Several have periods near 24 hours or near 12 hours, the diurnal and semidiurnal components respectively. The separation of the individual components in either diurnal or semidiurnal group requires a long tide record, but the separation of diurnal from semidiurnal components can be made fairly simply from a short tide record, as short as a day. These components are analyzed and expressed as if their periods were exactly 24 and 12 hours respectively. The procedure is as follows: 1. From tide records, choose a period of an integral multiple of 2 hours for which records are available for ocean and wells to be compared. 2. List water levels for each station, as measured from any datum, for half hour intervals starting with the time of the period of the analysis (t= oh), If the original records are discontinuous, plot them up so that half hour readings may be interpolated. 3. Make two tables of 24 columns each, for diurnal and semidiurnal components ge gee © each record. Enter readings for hourly periods (for t= 0,0, =.1.08, etc.) in the diurnal table and readings for half- hourly per ieds cee £2 i eae = 0.54, t = 1.0%, etc.) in the semidiurnal table. Set first 24 readings in a row across the columns in each table. If there are more than 24 readings put. the next 24 in a second row beneath the first, etc. Readings should complete the last row. 4. Add algebraically each column and obtain a mean. This will give 24 mean values for each table. 5. Add these 24 values and divide by 24 to obtain a general mean for each table. Subtract this general mean from the individual means to obtain departures. For example the 24 departures for the diurnal tide in Hilo Bay, Hawaii for a day beginning at 1/24/51, 0:00 are: EE a a 34, BO ee eGo, 1 eOk, ht) UL 761) 21151, -.251, e251, +02h9, +2549, +.649, +2599, +2399, +2199, +.049, ~.051, -.001$ +2249, +.6495.+.949, +1.219, #1.249, +1.149. These values are analyzed on the attached ue wo 6. Prepare a form Like the sample for each component for aaah station. Multiply the first twelve of the above departures by 100’ with the signs shown in the” twelve ‘coritesponding columns of row d of the form, and enter the products in row 1. Enter in the twelve columns of row 2 the products of the second twelve of the departures multiplied by 100 with the - signs shown in row e. Example: the 13th departure is +.599. Column O,: row eis (-). +.599 X(-100) = -59.9. Enter -59.9 in: column 0, row 2. Te Add algebraically row 1 and row 2, entering sums in row 3.: 8. Transfer numbers in columns 7 to 11 inclusive of row 3 to columns 1 to 5 inclusive of row 4, but in reverse order (column 7, row 3 goes to feos ok row 4) entering original signs (top) and also reversed signs (bottom). 9. ‘Ada columns 1 to 5 inclusive ne rows 3 and ie “using originali (top) signs and enter sums in row 5,.. Add columns. 1.to 5. inclusive of rows 3..and 4 using new gee: signs in row 4 and enter in row 6. Does Ite wey peames in row 5 by corresponding figures in row b and pith products vertically in column 0, rows 4 to 8 inclusive (Column 1, row 5 times colum 1, row b goes to column 0, row 4). Multiply figures in — 6,. columns to 5", by. corresponding figures in row c, and Sausage products vertically. in column 6, rows 4 to & inclusive. 11. Add’ rows 3’ to 8 in column O and etitienn sum in row 9. Add rows 3 to 8 in column 6 and enter sum in row 9. t 12. Divide figure in column C row 9 by 1200 to obtain "p". Divide figure in colum 6 row 9 by 1200 to obtain "q"". 13. Compute the amplitude of the component, A = V0 a. A will be in feet if the original water levels were in feet. 14. Compute the cotangent of the phase angle of the component, cot ef= p/q. Use trigonometric tables to determine e< in degrees. The cotangent will not discriminate between two values of ©*, 180° apart. Actual discrimination may be made by inspection of the record or By trial when both components: ‘have been analyzed for a station as described below. If t, = A is the time when the height of the component in question reaches a maximum, measured in hours after zero time of the observations, o in degrees = 360t/T, where T is the period of the component in hours. 15. When both diurnal and semidiurnal components have been analyzed for a station, compute the tide and compare with the actual tide for a check. -35- tc r)) cr Re 1 = diurnal amplitude; .«<- 3 = diurnal phase angle in degrees. 2 = Semidiurnal amplitude; o< 2 = semidiurnal phase angle in degrees. t = time after zero time of observations. 4 i height of water at any time above chosen datum. foo | iH] mean height of water above that datum. (Step 5 mean if above original datum) h =h 4 Ay cos.(15t ~2 1) . + Ap cos.(30b -o€2) Because of the approximations involved in using only one diurnal, and one semi- diurnal component, this equation cannot be used to extrapolate the tide fluctuation beyond the period of observation. However, comparison may be made of equations at different stations for the same period. Comparison of ocean and well tides: The tides in a ground-water body whose discharge is controlled by sea level will, at any well, bear certain constant relations to ocean tide. Using the same symbols as above and the additional subscripts © for ocean coefficients and W for coefficients at a given well: A,y/Ajo and Apy/Aoo the damping | ratios of diurnal and semidiurnal components, and o< w- “<10 and S< aw - 29 the phase lags for the two components will all be constants. Any of these may be used as parameters of the conditions between the shore and the well. The damping ratios have a constant relation to the phase lags and phase lag values may be computed from them for comparison. Furthermore, the phase lags for the two components are related through the periods of the components, and a value of a single parameter describing the whole tidal difference may be computed from them. If L = phase lag in degrees. ] lL, = X“iw- “10 180 |. Log f Axio Ti M Aa \ iW , Ly = Sai - 2€20 = _180_ log/_420_\ eee hoy | Where 1/M = log,l0 = 2.303 poe? | Sania ga6 7M logs are to base 10 Let C = the tidal difference in units of hrs. c= L Vr Wad «tage -76i~ Where T = period of component in hours Vi, = (2h = 4.899; \ip = V2 = 3.464; Yo = 0.09847 180 Gs Ly = Lo 4.899 3 6k, Comparison of tidal difference constants for different wells: The exact way in which the constant "C" is controlled by the conditions between the well and the coast is not known. "C" is inversely proportional to the square root of the permeability and directly proportional to the square root of the porosity. It is also proportional to a distance which is apparently some function of the distance from the ocean to the well, If there is an impermeable layer at a depth small compared with the distance from shore to well, it may be inversely proportional to that depth. Because of the present inadequacy of theory, quantitative study of permeability variation indicated by variation in tidal difference constants is impossible or unsafe. It is obvious, however, that if two wells are at equal distances from the shore, the depth to an impermeable layer, if present, is equal and the porosity is roughly equal, and if they still have very different tidal difference constants, the permeability is higher between the shore and the well with the lowest constant. Determining tidal corrections for single head readings: Once the tidal difference constant for a given well is determined, the mean head can be computed from a single head measurement for any time at which tides are still known in the ocean so that a new component analysis can be made for a short period of ocean tides including the time of the single measurement of well head. If Aj)’ andazo' are ocean coefficients from the new analysis, Ay/Ao and gue 6. Wood specimens: © oe | ban taeda A The nature of specimens or samples of wood ‘to be collected. -: depends upon the sort of information desired from them. Microscopic structure can’ be determined from.a very small. piece of: -wood, ‘while. ful: - engineering tests demand large logs which are ordinarily ‘impractical’ to-: eo" collect on an ordinary expedition. Although a. few general directions “may . be given here, if any extensive investigation of timber’ properties is in= tended, previous correspondence should be had with the laboratory whith is- ay to do the testing. Wood-testing, particularly of so-called "engineering properties", is an expensive procedure, and, unless previous arrangements have been made to have the tests made or unless funds are certainly available, much effort is likely to be wasted. o/s ane Size of samples collected will, undoubtedly, be determined by the practical matter of available transportation, If it is not practical to collect samples of more than a few ounces weight, structural properties based on microscopic examination, and macroscopic physical characteristics determined by inspecion are all that can be determined, The fact that should be borne in mind is that characteristics of wood from the trunk of a tree differ significantly from those of branch wood. Trunk wood should always be selected, preferably from clear portions of the trunk, free from knots, bends, diseased or injured wood, or worm-holes. If a very small piece, only, can be collected, then a portion of heart-wood, midway between the center and the sap-wood should be taken, A better sample is a plank-=shaped or pie-shaped section, at least an inch thick and a foot long, running in width from the center to the outside of the tree-trunk. Seasoning of such small pieces does not usually cause difficulties, but they should be exposed to dry air on all sides, or turned frequently. Density is a property of wood from which estimates or approximations of certain other valuable properties may be obtained. If there is serious interest in the economic properties of woods, the minimum test that is of value is a measurement of the density. Dimensions of minimum samples for determi- nation of density are contained in the following quotation, which is from a letter from the Director of the Forest Products ne of the U. S. Forest Service dated March 13, 1953: a, ‘The minimum sample size for specific gravity and shrinkage determinations depends upon the diameter of the log, the presence of knots or other defects, characteristics of grain, and the like. In general it is -. preferable to obtain sample sections from living trees that are freshly cut in order to be sure that they stay green during transit and to minimize stain and decay. Only heartwood of dead timber would be sound, It is desirable to secure sections from a reasonably uniform height in the tree, say 10 to 16 feet above the ground (for better comparison among species and with data in Tech, Bull. 479 where samples ordinarily represent the 8 to 16 foot height). The section should preferably represent the full cross section except in large trees where shipping costs of a full section become excessive. It is difficult to lay down hard and fast rules but, in general, if the diameter is 12 inches or more a section 12 to 15 inches in length should suffice. For trees smaller than 12 inches in diameter we would prefer longer sections, say 18 inches in length down to about 8—inch diameter, and not less than 2 feet in length if less than 8 inches in diameter. The ends of the tree sections should be coated promptly after the section is cut, preferably with not less than 2 coats of aluminum paint or a heavy coating of asphalt paint. If coating cannot be applied the sections should be increased in length to provide not less than a 6—inch cut off at each end upon arrival at the Laboratory . While such provisions may appear somewhat: rigorous, we have seen many instances where sample sections were of little use because of drying, checking, and decay ‘when proper precautions were not taken, and the results of tests on such material are, to say the least, subject to a great deal of question, "! a5ju III. Directions for Collecting Information on Varieties of Economic Plants | A number of the most important economic plants in the atoll groups have proliferated into series of horticultural varieties, Of these very little is known, and what is known is not adequately backed up by specimens. More information should be collected, and an expedition where an anthro- pologist is present provides an ideal opportunity to do this. His co- operation should be sought arid made use of, both-in dealing with the natives: and in cross-checking information. The results should be of equal interest : : . to him and to the botanist. Brief directions | as ‘to what is needed for the principal groups of plants follow: 1. Pandanus. Varieties differ in size and shape of fruiting head, sculpturing of distal end of phalanges, number and arrangements of styles, size and fleshiness of base, juiciness, sweetness, flavor, and irritating qualities of flesh, prickliness of leaves, toughness, pliability, and other weaving qualities of leaves. The first. half dozen of these features will be shown by the herbarium specimens. The features of the flesh must be found by trying them, and by asking the natives! opinions. The natives must: be consulted as to utilitarian qualities of leaves, also on use made of fruits. These may be eaten raw or used in several’ preserved forms. Native names should be secured, as well as any information available on origin of the varieties, how propagated, whether they come true from seed, etc. All © of this information should be carefully associated with specimens. .. ; 2. Coconut, Varieties differ in size, shape, color of nuts, - retieer in a cluster, flavor of water, thickness of meat, edibility of husks, and perhaps other characters. Very little is known of these varieties. Con-. - siderable instruction will be needed from the natives before any understanding will be gained. It is of especial interest to know if these are. really characters of fixed varieties, or if they are merely characters that . segregate independently. Do the characters have names , or the individual . trees, or general types of trees, or actual varieties? | What are the differences in uses of each? Origin?. 3. Cyrtosperma, The tangible a5 Pfereneds| here are in door and ce | liness of petioles and pedunclés. Perhaps there are differences in edible qualities, or cultural characteristics. Botanical specimens with notes will show prickliness and color. These must be correlated, if possible, with other features, Names, Origins. Seasons. Do not confuse with Alocasia. 4. Colocasia, Taro varieties differ in stature, number of leaves fully open, shape of blade, color of petiole, veins, and blade, color and arrangement of tubers, edible qualities, cultural Fi hha aigaege ead Names. Origins. Seasons. er 5. Alocasia. Possibly two varieties or species, one with green petioles, the other with purple. Differ.in stature. Possibly varieties with— in these, differing in edible qualities. Determine if actually eaten, if” any way to prepare to get rid of acrid qualities, under what circumstances eaten, if actually cultivated, names, origins. Do not taste. -5h— 6. Breadfruit. Many varieties, in two main series, differing in being seedless or with seeds, Varieties differ in leaf outline, size, shape, and surface of fruit; edible qualities, whether can be eaten raw, or only cooked, flavor; season of bearing, if fruit is preserved in any way for off-season. Evidence particularly desirable as to whether seed trees are ever fertilized by pollen from seedless, whether two species were originally present, one entire leafed and seedy, other incised leafed and seedless, now mixed by hybridization. Origin of varieties. Names. Place me pana of natives. ‘ Ue) lacea. Careful watch should be oe for varietal differences. _ Extent of use and modes of preparation, as well as whether actually planted -or cultivated should be determined. 8. Other economic plants. Notes should be taken, with specimens, indicating cultivation or use of any species of plants that in any way enter into the native culture, past or present, This should preferably be done in close cooperation with an anthropologist. Information gained locally, either from native informants or from missionaries or other white residents should be carefully verified from independent sources before being accepted as fact. 1. 2e 36 seat ARE ak RY Me ein 200 corrugated aluminum anects A owpeieenasean ts 12x "7 inches. The corrugations should not be over 1 cm, wide, _ and ‘must run transversally rather than longitudinally, 400'felt driers or "blotters", sheets of deadening felt 12 x 17 inches. (If aluminum ventilators are not available, 500 hard finish double-faced corrugated cardboard sheets, 12; x 17 inches , will do in place of both driers and ventilators.) Newsprint, folded into quires 17 x 12 inches, or newspapers *— with the folded end torn off so as to make quires of this size. Enough for 1500 or 2000 folders should be enough for a season ' on.an atoll, depending, of course, on the number of duplicates a eo we collected. An atoll may be expected to have between wpe AS and 150 species, some of which will have to be collected more ‘than once, ; | 100 sheets of stiff cardboard, 12 x 17 eee for tying bundles. Several cones of the heaviest butcher's twine, A roll of waterproof paper. Five pairs of wooden press frames or boards, 12 x 17 inches. Five pairs of 9 foot two-inch webbing trunk straps. Five pairs of 12 x 17 inch sheets of beaver—board or other very heavy cardboard for field presses, Five pairs of oilcloth envelopes just large enough to slip the sheets of beaver-board inside of, with a flap and snaps. Five pairs of four or six-foot, one-inch webbing trunk straps. Two gallons 40% commercial formaldehyde. Two gallons of 95% ethyl alcohol. Three l-inch or 1 1/2-inch paint brushes, Ten lbs. para-dichlorbenzene, 10 yards of gauze. 109 medium sized paper bags. Three Harper "Beatrice" flat-wick kerosene stoves, or equivalent, or six ordinary kerosene barnyard lanterns. Extra wicks. 19. 20. 21. 226 236 2h. 256 26. 27s 28. 29. 30. 31. 32. 336 —56— Coleman lantern, extra mantles and generator. Two sixteen-inch or eighteen-inch Collins machetes, Two pairs small clippers. ; Two boy scout knives, 200 shipping tags. Five bottles insect repellent, Record books, fountain pen, steel pens, permanent ink (Parker 51 pen and ink are good, as the ink dries very rapidly), blotters. soft pencils. Small chest or foot-locker with lock in good order, for notebooks and other valuables. Piece of chick wire 2 x 6 feet. 5 yards of white muslin. Waterproof matchbox or lighter that works. 2 or more 1 or 2 gallon tins with spouts, for kerosene and gasoline, for filling stoves and lanterns, One drum of kerosene, One drum of white (unleaded) gasoline, to be used for cooking as well as for light. eth: tices’ aiebtod de anda ffl Ne Rn hard z\ ah eb faba tt AooL rivibw ‘anstool~ hag os ot waakionsy ay SUGGESTIONS FOR COLLECTING FUNGI by, Donald P, Rogers a __.. Fungi parasitic on leaves or other plant parts are almost entirely unrecorded from atolls. Many of these occur as or in leaf-spots; the fungus fructification may be a superficial pustule or an embedded fruiting chamber (perithecium or pycnidium), often surrounded by a dead or discolored area of the leaf. The distinction between infected leaves and those merely damaged by insects or mechanical injury is not always easy in the field. Often close inspection by a hand-lens is necessary to reveal the fructification; and even then the ostiole of a pycnidium may appear as no more than a brownish, or blackish fleck in a discolored area, Other parasites, the sooty molds , may form wide-spread blackish incrustations. Occasionally a fungus (such as the downy. mildew on Boerhavia) causes marked distortion of the host. All parasites on herbaceous plant-parts are treated, as’ the host alone would be-- that is, pressed, preferably along with sufficient host material to confirm the identification of the latter. Parasitized woody parts are dried without pressing. Slabs cut from a stem will dry more quickly than sections of the entire stem, and since the fungus fructifications are not apt to be deep, are just as. useful as larger pieces. .Where there can be ‘any question of the identify of-the host, material sufficient for’ identification should be ahah a to accompany woody specimens. iy Although one specifi¢ identity of the ieubendata of non—parasitic fungi is of less importance, it should wherever possible be noted. Such fungi occur on soil (but ‘not so commonly as in:temperate regions), on dung, and on dead plant material, including worked. wood and fiber. Standing trees or stumps, fallen logs. (a rich and interesting source of fungi), accumulations of dead vegetation on the ground and especially in trash-pits—all serve ‘as substrata for saprobic fungi ‘and should be .examined. Fungi are apt to. be present although not visible in soils and humus of the first shrub-zone of the upper beach, of forested areas and of: the wet pits planted to .. + Cyrtosperma and banana; samples of soil, and-of dung, thoroughly dried and wrapped so as to minimize contamination, can serve as a source of cultures of such fungi as mucors and water-molds even after months or years, Myxo aree . The fructifications resemble minute puffballs or stalked ‘globes or eggs,:-or may be sessile and effused, They.are extremely fragile; and although: usable specimens may be kept in paper packets, the usual -and better practice is to :wedge or pin portions of their substratum into’ ‘the tray’ of a penny match=box or similar small box. Myxomycetes are found on decaying plant debris and on stumps. and logs, or sometimes on sound trees or | wood where they have aoe to fruit. Ay Phycomycetes. ‘Most forms will be aaGEADIe and must be heen ave c culture methods from dung or. soil samples. One is known as a systemic: parasite of Boerhavia; other such downy mildews may be found. Glaziélla forms hollow red fleshy fructifications, up to 2 inches across, in trash on the ground. If it is found, the exact color of living specimens should be noted. A piece of printed paper, baits BS: a: cigarette. Ce es will serve as. a color-standard...._:: nlioweia * ‘tc Bs {Wires TA any TR anos Ailes vr ae -~58- Ascomycetes. Many leaf-spot fungi belong here. ‘In addition, some saprobes will be found. Cup—fungi occur on soil or dead wood, and may occur also on dead herbage. They are firm enough to be dried readily. Pyrenomycetes, hard-fleshy to woody or carbonaceous growths, sometimes bright-colored but often coal-black, develop especially on dead wood. Some are minute and mammiform; many are embedded in the substratum, with only the ostiole or beak of the perithecium protruding; others form encrusting layers or large nodular to club- or antler-—like oe cwulice Readily dried, Basidiomycetes. For the satisfactory identification of most Hymenomycetes (those Basidiomycetes with exposed fruiting surfaces) a spore-print is of considerable use, although for most it is not of critical importance. To secure a spore-print the fruiting-—body should be placed, fertile (usually lower) surface down, on a piece of white paper, and unless the atmosphere is nearly saturated, covered with something such as an inverted tumbler or empty can that will prevent drying out of the specimen before spores have been deposited. If black paper is available, such as that in which photographic film is packed, a part of the specimens may be placed over white paper and part over black. The color of the spores can be learned (and should be noted from the fresh print) from the deposit on white paper; the black will show whether spores have been shed. A microscope slide will do as well as the paper. The spore—print is dried and filed with the specimen. If the weather is very wet, or if only a single specimen of a fleshy fungus is available, spore-printing may destroy the specimen, and it is then better to do without the print. Twenty~four hours should be time enough to secure a good print, if the material is going to yield one at all. Most larger fungi, including nearly all Basidiomycetes, should be wrapped when gathered in pieces of newspaper, one collection to a packet. As soon as possible the packets should be opened and the specimens spread for drying, on a flat roof, table, or sandy area, if the atmosphere will dry the material, or above an reeds source of heat if that is necessary. Fleshy “specimens need quick drying; woody ones are useful’even if conditions for preparation are poor, E The jelly—fungi (cieecreiaties) may be ear—like and hairy on the sterile surface (Auricularia), jelly—-like and translucent (Tremella), or only a mucous or gelatinous layer'on a dead log (Sebacina). Spore—prints aré desirable, but the basidia that produce the spores'are of greater importance for subsequent study, and if the specimen is held a day for a spore-print and then too slowly dried the basidia may nearly all have dis- charged their spores and disintegrated by the time drying is complete. It is often better, therefore, to dry a part of a collection at once, and to use only a part for a spore-print. If -the collection is small, ora choice must be made, it is better to secure immediate drying. If the heat is not so great as to kill the fungus by cooking, a well-dried specimen can often be revived months later and spores secured. The genus Septobasidium produces lichen-like growths on living, plant parts, usually twigs or smooth bark, A very interesting lavender-gray species has been found on Pandanus in the Marshalls; one should be sought on Citrus twigs. The fungus is usually felt~-like in texture, and under the lens may show three layers in section-— a continuous basal layer in contact with the bark, a layer of pillars and air-spaces, and a continous fruiting | | -59- surface-— all in a thickness of one or two millimeters. It is parasitic on scaleinsects which in turn parasitize the host-plant; and any collection of Septobasidium therefore presents a triple problem in biogeography-— that of the host-plant, the insect, and the fungus. This should be dried like any similar fungus; specimens in preservative are aiso desirable, Thelephoraceae (and some Fungi Imperfecti which resemble them) and Hydnaceae may form films or sheets closely applied to the substratum (resupinate), or may be reflexed or occasionally free or even stalked. The last family is characterized by a tuberculate or toothed fruiting- surface, and the other groups by a more or less even one. Even the most delicate ones, which appear almost mold-like, are worth collecting; these should not be spore-printed unless quick drying is possible, or unless a part of the collection can be dried at once and a part sacrificed for the print. Unless the fructifications are stipitate, all should be collected with a slab or generous fragment of the substratum, Polyporaceae may be shelf—like and leathery, fleshy, or woody, or may be resupinate like some of the groups earlier listed; all have a fruiting surface marked by pores. They are easily dried; the resupinate ones should be printed if possible, even if this necessitates first moistening the fructification or soaking the attached substratum. Polyporaceae may be shelf~like and leathery, fleshy, or woody, or may be resupinate like some of the groups earlier listed; all have a fruiting surface marked by pores. They are easily dried; the resupinate ones should be printed if possible, even if this necessitates first moistening the fructi- fication or soaking the attached substratum, Agaricaceae, the gill-fungi, will be found both on dead wood and on the soil. Some are extremely fleshy and decay very very quickly, and for . these the preparation of specimens will be a race between the drier and the processes of decay. The smaller, less fleshy species will retain their form and texture fairly well; but specimens of the larger species are usually worthless unless accompanied by notes showing color, attachment of gills to cap, and presence or absence of a ring around the stalk and a cup for membrane at the base. Color-notes are desirable for all. It is also desirable to make a median longitudinal section of one or two before drying. Gasteromycetes, the puffballs and stinkhorns, will probably be rarely found, but are of considerable interest. Puffballs are worthless except for food unless the spore-mass within has taken on its characteristic deep color, Stinkhorns, which may have a simple columnar receptacle covered at the tip with a sticky greenish spore-mass, or may have a number of free or interwoven arms, are greatly sought by carrion flies, and may have to be sprinkled with formaldehyde before drying or put at once in preservative. Rusts and smuts are rare on atoll plants. If found they should be treated like leaf~spots. Lichens. As is generally known, these are compound structures composed of a weft of fungus threads in which are embedded algal cells. In moist areas they are quite common on trunks of trees--either crust—like -60— forms or rosettes; rock-incrusting species are not usual on atoils. Lichens are apt to be whitish or grayish when dry; rain or artificial soaking usually shows up the green component and shows them to be lichens rather than simply fungi. Specimens dry readily, and ordinarily are not much damaged by re- maining moist for some time. They should be collected whenever found. Whether many of the atoli piants are mycorrhizal is not known; informa~ tion on this point would be of importante not only to mycologists but also to the ecological understanding of the vegetation. If soil fungi are found it would be well to note whether they are associated with particular plant species. if they are, or perhaps in any case, portions of the absorbing tins of roots, held in preservative, should provide data on the mycorrhizal or non-mycorrhizal nature of the plants. . ate COLLECTION OF SOIL SAMPLES FCR THE RECOVERY OF AQUATIC PHYCOMYCETES by F. K. Sparrow Equipment needed: Tablespoon; wide-mouthed jar of 710% ethyl alcohol; 2 oz. cardboard cylindrical drug cartons, 24" high by 15" in di-. ameter; tare (adhesive or scotch). Collection method: Scrape azay topmost inch of soil, fill carton with soil, using spoon,.tape. carton shut, alae collection data on ess ee a spoon in alcohol. te It is better to collect only dry soils. Coliection data: Locality, depth, any special features such as proximity to beach, tide line, etc., date, coliector. ~62- INSTRUCTIONS FOR COLLECTING ALGAE _by Maxwell 5. Doty Algal forms of three categories should be sought in: particular: 1. Algae that are prospectively important in the structure of reefs, in the filling of lagoons or in contributing to the sand of which the non—volcanic islands are formed; 2. Algae of use to -the natives; 3. Soil algae that contribute to the binding of the sand, humus formation, or nitrogen fixation. Algae of the first category should be sought on the reef, encrusting coelenterates and rocks, or waving free in the water or on lagoon bottoms. Aigae of the second group are to be obtained with the aid of local guidance as to names and uses. The algae of the last category appear as strands, scums and crusts that vary from elusive stains to tar-like crusts of various colors or black. They appear on sand, dead wood or in pools, both on dry land and in the water. Labels indicating, in these respects, the roles of the different algae collected will greatly enhance the value of the collec~ tion e There are two methods of preservation recommended. The simplest is drying the algae in the shade in clumps or spread on paper or leaves. Drying will be hastened by the elevated temperatures above a heater (e. g. Coleman lantern). When dry or nearly so, the algae should.be wrapped with their labels in bundles and packed so as to prevent crushing. The second and better method for field use is "pickling" in ten per cent formalin in water, ten. per cent sea water formalin being used for marine forms. "Alcoholic. specimens" are good. It is feasible to set aside a can. of four per cent formalin to which, in the course of other work, the different kinds can be added, as found, until it is felt thet a representative collection from the area has been obtained. Most phycologists working in the Central Pacific recently have sealed their specimens in tin cans with a can sealer, as in the process of home canning. This method is greatly superior to older field methods, as it avoids breakage and leakage, and identifying marks painted or scratched on the can are perma- nent. Algal herbarium specimens mounted on herbarium sheets or mica are, of course, most welcomed. Preserved algae may be shipped mixed with cther preserved materials such as fish or invertebrates, but take measures to prevent the rock-—line corallines from crushing the more fragile forms. It is desirable but not essential that the algal collections be kept separate. Again: The value of a collection will be inestimably enhanced if labels will include information relative to the roles outlined in the first paragraph above, or information which will show critical consideration of distribution. —63- SUGGESTIONS FOR COLLECTING MODERN AND FOSSIL CALCAREOUS ALGAE ” by J» Harlan Johnson Calcareous algae are those seaweeds that deposit lime within and . around their tissues. They are.of interest and importance because they | assist in the formation of. coral reefs and often contribute appreciably ae the formation of limestones. In the- course of geologic tine,. some members of most of the major - groups of algae. have developed the ability to secrete or deposit calcium carbonate. In connection with studies of tropical reefs, however, only two groups are important, the red coralline algae and the green Codiaceae, (represented by Halimeda) and Dasycladaceae, (represented vi Cymopolia and and Acetabularia), The red coralline algae include both crustose and articulated types. In general the crustose corallines are important on reefs and.on reef lime- stones. They form the Lithothamnion Ridge, (which is usually constructed mainly by a few species of the genus Porolithon),and play an important part in binding together the various organic elements of the reef, They may contribute appreciably to the mass of the reef, Halimeda species, the most important calcareous green algae, grow in the lagoons, on the reef flat, and sometimes on the reef face. They may develop banks or meadows over large areas, especially on lagoon bottoms. Some bottom samples from the- lagoons at Bikini, Saipan, and Palau show that Halimeda remains are the, principal constituent of the sediment. In some places,.Halimeda fragments ‘contribute to the sand and beach rock, Collecting Modern Algae The massive, branching, encrusting, and nodular forms of the crustose corallines need relatively little attention. After the specimens are collected, they should be soaked in fresh water for several. hours. or over- night... Then all extraneous material, especially worms, ‘sponges, and other animals, should be picked out. The specimens should then be washed care-— fully in fresh water and allowed to dry for several days, until they have bleached white and. until any enclosed animal remains have ceased to smell. They. should be wrapped well and packed, carefully in strong containers for _ shipping. The large, highly branching forms , are heavy and very brittle, . so they require careful packing with paper, shredded paper, cotton or straw. If it is desired to preserve the color or the associated organisms or both, the specimens should be "pickled" in a sdlution composed of 90% sea water and 70% formalin and packed in containers. Since crustose algae are both heavy and brittle, only one or two specimens should be put in a con tainer, and -they. should be packed so as to fill it. If specimens are packed too loosely or too many to a container, crushed and broken material will result. For the same reason, crustose algae should not be packed with articulated corallines, Halimeda, or noncalcareous algae. bigs Articulated coralline algae are very fragile. Specimens should be washed or soaked in fresh water to remove all salts, Then they should be placed on blotters or newspapers and allowed to dry. slowly (not in sun) for several days or a week. Specimens should be wrapped carefully in tissue paper and packed in shallow boxes or trays, padded with cotton or soft paper. These boxes should be packed in strong containers for shipping. Halimeda, Dasycladaceae, and other calcareous green algae usually contain more moisture and organic matter than the calcareous red algae. They are also commonly fragile. After the specimens have been soaked and carefully washed in fresh water, they should be spread out on blotters. or newspapers and allowed to dry slowly, possibly for several weeks, During that time the specimens should be turned over every day or two. Care must: be taken to prevent mold or mildew, -which can permanently discolor or even destroy specimens. The dried specimens are fragile and need to be packed carefully. Fossil Algae Fossil algae are best known from fragments scattered through lime- stones, especially limestones containing corals and large Foraminifera. Nodular masses and large crusts may form large lenses or beds of limestone, Sometimes algal material may weather out of chalky or marly beds. In:most cases the algal fragments are firmly imbedded in the rock and samples of the limestone SL bP. collected. Take material as little weathered as possible. To identify ‘She algae, thin sections are necessary. This means that fairly large specimens (3 x 4 inches or tareer) or several small specimens should be taken as samples. Calcareous algae are also known to weather free from marly or tuffaceous deposits; more attention should be paid to obtaining such material, as it may supply the best-preserved specimens. Loose specimens, which may be studied in their entirety and by means of oriented thin sections, are ideal for comparative systematic studies but are difficult to obtain--so they are worth searching for and collecting carefully. They should be wrapped in tissue paper or cotton and put in vials or small boxes. They can be cleaned and freed of matrix in the laboratory. ’ : Desirable Field Data It will greatly help the specialist who studies the specimens if the collector will label the material carefully as to locality and will supply notes on distribution, ecological association, and depth of water. Chapter 7 -— Zoology SUGGESTIONS. AS TO COLLECTING LAND VERTEBRATES "th ON ee ATOLLS Dyiidin’. als Marshall, A bea Each member of the party dhe qe provided with a copy of "A field collector's manual in natural history" publication 3766 of the Smithsonian Institution, Washington, D. C. This concise pocket book: gives complete instructions for collecting and preserving all types of animals. The Field Notebook Preferably loose-leaf, of medium size, the: field notebook is much more important than the specimens. In it are recorded observations on the behavior, feeding methods, habitat, breeding activity of the various animals. Species Accounts: Keep the records for each species in a series of pages devoted to that species alone. Memory is treacherous, and to be of value, these observations should be written down in the field while oa are watching the animal. Catalogue ofS Specimens: The catalogue duplicates the information on the specimen label, and is the place where additional information (for which there may not be enough room on the label itself) such as coloration, stomach contents, parasites found, slides ‘prepared, condition of gonads, amount of fat, habitat, — time of collecting is entered, Number each specimen in serial anders in the order in which they. are collected, using the same consecutive series of numbers — for all the vertebrates. Each number then is unique for the particular specimen, and this number appears both on the specimen labei and-in the catalogue. Label: Labels will be provided by the Smithsonian Institution. They should be of tough paper which will not go to pieces in liquid, and should be written upon clearly with Higgins Eternal Fountain-pen Ink. For any vertebrate specimen, the information which must appear on this labe].is _ exact locality ‘date ' your signature: your catalogue number In addition, for birds, the sex, as determined by opening the body cavity and examining the gonads, must be entered; for mammals, in addition to the sex, the following measurements must appear, and to save space, you can merely put the numerals down in this order, in millimeters: total length, tail length, hind foot, ear. In addition, for all vertebrates, put as much of the further information as stated above under catalog for which there may be room. A, strong label, the "skull tag" besPing your initials, cat.’ er sex, is tied to the mammal skull. 2. gp a The Specimens With the bookkeeping methods well in mind, let us now turn to the actual preservation of the specimens. ~ Reptiles and amphibians: Don't neglect the nocturnal species. They may easily be found by their eye-shine, seen when you hold the flashlight near the level of your eyes. A head flashlight is a great convenience. These animals are easily caught by hand while they are active, or when uncovered in their hiding places. For the shy forms, a noose of leader or grass at the end of a long wie -66- stick is an efficient collecting device. Tie a label on one hind leg, slice open the body wall and big muscle masses in a few places to permit penetration of the preservative, and place in 70% ethyl alcohol. Alcohol can usually be obtained from Navy hospitals. Spike or change the alcohoi before shipping home. Record coloration in the catalogue, for these animais fade. Birds: These can often be obtained from the poiseeas who have various methods of snaring them. It is not absolutely necessary to take a shotgun along to collect birds, as the forms found on atolls are pretty widely distributed and well known from the standpoint of specimens. They are practically unknown as regards behavior, annual cycle, etc. You can make a good skin of a bird shot with a small bore rifle, as long as it is not hit in the head. Powdered arsenic, dry sand, cotton and scissors and forceps are needed in preparation. Liberally sprinkle sand all over the bird while skinning, so that blood and juices may not. soil the feathers. The process of bird skinning in a nut-shell is this: make an incision lengthwise of the belly from about the middle of the breast muscle to the cloaca. Lift the skin away from this opening and turn the skin completely inside-out, thus removing it from the carcass, and cutting it away from any attachments to the carcass in the following order: Cut each leg from under the skin.at the knee; cut the tail through the caudal vertebrae, again under the skin, as you expose this area in turning the skin inside-out; cut the wing through the humerus, pull out the ears, cut the eyelids between the skin and skull; cut out the back of the skull and roof of the mouth. Do not cut off the bill. You now have the skin inside-out, with the bill, skull, wings, feet and tail all attached by the skin. Clean off all meat, brains, fat, and if very fat soak in gasoline or scrape and keep sprink- ling with sand. (Caselineasoaked skins easily fluff out and dry if held up in a brisk wind.) Now turn the skin right-side-out again and from the ventral incision, fill each eye with a ball of cotton. (Sprinkle arsenic on the skin before turning right-side out.) Put a stick wrapped with cotton into the neck, and anchor the sharpened end of this stick in the base of the upper mandible. Roll up a piece of cotton the size of the carcass and put it into the skin, with the neck stick ventral to it - thus the neck stick runs along the throat, and we have left a hollow space at the back of the neck into which the neck feathers can find their proper alignment. Sew up the incision, cross the feet and tie the label around both at the point where they cross. Wrap in- cotton until dry. The object of this preparation is to have a specimen resembling the dead bird, which will be well filled out with cotton.so as to reveal the plumage and coloration, and from which measurements of the bill, wings, tail and feet may be taken. This requires that the bill should be closed in a natural fashion. For large birds, it is necessary to skin out the muscles of the wing. This can be done by turning the wing skin inside out as far out as the wrist. This of course involves stripping the secondary feathers off the ulna, but is a justifiable procedure as long as you are certain to puil these feathers back along the ulna to their original locations when you turn the wing right-side out again. The sex determination appearing on the bird label should be based on an examination of the gonads, which lie at the anteroventral portion of the kidneys, near the adrenals. Two oval white ¢h yellowish bodies with tiny tubules showing through the transparent covering, which has a smooth surface, indicate a male. In the female, there is usually only the one gonad, on the left, and it is granular and irregular in shape, owing to the presence of numerous ova, ~67~ Mammals: Again, it is well to bear in mind, especially if you are restricted as to luggage, that it is not absolutely necessary to have traps. The little Polynesian rats, and the larger house rats can readily be taken as follows: break open a fresh coconut and lay it at the-edge of a pile of coconut husks, or a rock wall, or a rotten log. Sit quietly~. with a stick and club the rats on the back as they come out to eat the coconut meat. The natives can get you many in this way, though they make the mistake of hitting them on the head, thus ruining the skull for scientific study. The skinning method ts Similar to that used for the birds with this important difference: the four measurements must be taken before skinning, and the entire undamaged skull must be removed from the skin, the brains blown out through the foramen magnum, and a skull tag attached to the skull, The tail bones also are entirely slipped out of the tail skin, to be replaced with a long wire carefully wrapped with long-fibered cotton of just the right thickness to get all the way back to the tip of the tail. After dusting the skin with arsenic and turning it right-side out again, it is filled with a cylindrical piece of cotton, pointed into the snout, and wires wrapped with cotton are thrust into the four feet. These four wires and the inner end of the tail wire lie between the skin and the stuffing, on the ventral side of the animal. After sewing up the incision and tieing the label on the hind foot, the rat is pinned down on a board, with the soles of the feet down, © Problems of Interest on Atolls Large series of the rodents and lizards should be preserved. Look for signs of the introduction of house rats, and determine if they are eating the green coconuts. Apparently the Polynesian rat does not eat the nuts on the tree, Among the reptiles, the skinks are of particular interest. for they are apt to show, polymorphism (within the same species) and you are likely to observe different proportions ‘of the various color phases on different’ islets of the same atoll, You may even find in the skink genus Emoia that two closely related forms, differing markedly in color replace each other about the atoll (apparently not occurring together upon any one islet), ‘Exceedingly valuable data will accrue from a brisk survey of every islet of. the atoll, whereby you collect representative samples of such a species and note the environmental conditions under which each form flourishes as well as the conditions upon islets where neither is found. Thus one might be permitted to evaluate the possible role of natural selection in bringing about such strange distributional patterns. In other words, environmental conditions about an atoll may not be as uniform, from the standpoint of land vertebrate habitats, as one might suppose.’ Striking irregularities in distribution may be disclosed by such a canvass:of the entire vegetated land area of the atoll--vastly worth-while even though brief, For birds, the items of interest are a determination of the breeding period for each species. (On Ascension Island, the sooty terns are now known to breed every 9 1/2 months, year in, year out!). Also observe and record location of colonies, time of activity, whether flocking or solitary, behavior in general, The most valuable observations accrue from 2 ga gas or anal observations joe the same individual or group. wre | . Parasites: ©... | | After going such a long distance to get your specimens, it is | rather too bad if you do not take full advantage of all the information that they may yield, before they are preserved. It is worth the trouble, there- fore, to take along a microscope and look for rectal and caecal protozoa, and for intestinal helminths, These are easy to find, for if they occur at all, they are usually swarming. For instance, a drop of liquid which a lizard exudes from his cloaca when handled, may be just a living mass of flagellates. The same applies to the caeca of birds and of rats. Take along a stock bottle of Bouin's fixative, and.a supply of round coverslips, and vials into which they just fit, round papers, the size of the covers. Smear the fluid from the rectum or caecum on one side of the cover slip and immediately (before it begins to dry) float it face down upon the surface | of the Bouin's solution. After a half hour or so, transfer it to 50% | alcohol, then to 70% alc. in the vial. On top of the series of covers all | coming from one vertebrate specimen, put a paper label bearing your catalogue number for that particular vertebrate, and the organ from which the smear was made. It is of great scientific interest to do the. same for the pro-— tezoa living in the hind gut of termites, Keep a vial of the termites in alcohol with a catalogue number corresponding to that on the label for the coverslip preparations, so they may be identified. Roundworms are killed in hot 70% alcohol (this makes them straighten out) to which is later added a little glycerine, if handy. Flatworms should be placed in Bouin's fixative overnight, then placed a few minutes in 50% alcohol, then stored in 70% alcohol, If time and space are very limited it is possible to get good results by just leaving these cover smears and flatworms in the Bouin's, but it is Leek pags to store in alcohol after killing in the Bouin's, . Ectoparasites: Mites, ticks, lice, fleas, hippoboscids, and other external parasites are only collected efficiently in.conjunction with _ collection of their hosts. Mites which infest lizards are automatically preserved with their hosts in alcohol, but, because they may drop off, the lizards should, at least at first, be placed in separate receptacles of alcohol, separating the species and localities. When the alcohol is changed any mites that have dropped off may be picked up with a medicine dropper and placed in a vial with a label. Wrapping bird and mammal skins in a thin layer of cotton while they are drying will serve to gather many of the ectoparasites, which leave the host as it dries. They do not live long,. especially if the skins are dried in the presence of paradichlorbenzene or napthalene, and when they leave the skin they usually lodge in the layer of cotton. Rats, which usually have fleas, may be left for a short while in a jar with paradichlorbenzene crystals or a few drops of gasoline or ether, The fleas will be killed and either remain in the hair of the rat or fall to the bottom of the jar. All of these parasites may be preserved in alcohol, Like the helminths, they should be kept in a separate vial corresponding to each host specimen, as proclaimed by a slip of paper in the vial, which bears your catalogue number of the vertebrate host, as well as the location or organ where the parasite was found. SUGGESTIONS FOR ‘COLLECTING TERRESTRIAL INVERTEBRATES ON PACIFIC. ISLANDS by Rohesh Td cee ese: Introduction... The best general advice on collecting terrestrial invertebrates is to look’ everywhere, collect everything, preserve it care- fully, and label it adequately. This comprehensive statement is especially true for the little-known islands of the Pacific where even the commonest. invertebrates may prove to be of exceptional interest. The following remarks © are intended.to provide detailed information on specific points in collecting. | It may appear that a disproportionate amount of time is. devoted to the in- sects, but the collector will find that this is the largest group of ss cele Fe trial arthropods, both in species and individuals. Collecting. The various methods of collecting may best be disdiseed separately, though in actual practice the procedures are carried on almost Ma, simultaneously. The most obvious method of collecting insects is.with a net. . Flying insects are encountered in greatest numbers when the sun is shining. Sweeping and beating vegetation is possibly the second most important method of collecting. In this connection a sturdy net bag is essential, or. a. can- vas sheet, or an inverted umbrella can be placed beneath vegetation and the . limbs can be beaten with a heavy stick. Beating and sweeping yield large numbers of inconspicuous insects which would be completely overlooked by other methods of collecting. The third type of collecting which should not be neglected is ground collecting under stones and logs, on roots, and in leafmold «+ ; In addition to. the above-mentioned types of collecting, which are per— . haps the most important, there are numerous specialized methods that need to be mentioned. Aquatic collecting ‘ ‘is one of: these. Specimens should be sought for in ponds, taro patches, wells, cisterng,.tree holes, opened coconuts, bases of leaves and fronds of Pandanus , palms and, various epiphytes, and in protected coves: on coral reefs or even in thé-open oceam. Mosquito larvae (or better still, pupae) should be ‘collected with a net or dipper and kept alive until the adults emerge, Light collecting is another specialized, but sometimes very profitable, method. “Insects are, usually attracted to lights _ in greatest numbers on evenings when there is no moon and when the atmosphere | is relatively humid. Electric lights may be used, preferably with a white - background, or a Coleman lantern is satisfactory. In places where a Coleman. . lantern is used, care should be taken to. provide.white gasoline, Wood-boring- insects may be collected by beating dead or dying branches of trees or by gathering such branches and rearing the larvae. out in closed containers. A rich fauna will -be found under loose bark and in rotting wood. Equally pro- ductive are décaying breadfruit and ‘rotted :pandanus fruit as well as other . fruits which have fallen to the ground. An entirely different type of life. will be found in animal carcasses and dung. Ectoparasites of man and animals. | should not ve overlooked. These are found on the animals or in some cases | in the beds or nests. Plant-feeding caterpillars including leaf miners should be reared by collecting the infested parts of the plants and placing them in cardboard or tin boxes. Additional food must be provided from time to time. This type of rearing is also very productive of parasites ‘which might ot herwise olf be overlooked completely. To complete the coverage of the insect fauna, close attention should be paid to small insects which occur on the leaves and flowers of plants. These include scale insects, thrips, aphids and microscopic plant-feeding mites. The latter may only be visible under the magnification provided by a small hand lens. Other terrestrial invertebrates include the land snails and fresh- water snails, earthworms and leeches, nematodes and planarians, Crustacea and other arthropods such as millipedes, centipedes, spiders, pseudoscorpions, etc. All of these will be encountered in the course of the various types of collecting mentioned above for insects,’ and especially in ground collecting, bark collecting, and aquatic collecting. Two highly specialized types of collecting should be mentioned for the sake of completeness, though they should not be undertaken except under the direction of a specialist. The first of these is plankton collecting of Protozoa, rotifers, etc, A plankton net is necessary for this type of col- lecting. The other, and most specialized type of collecting is the.search for endoparasites, and for blood flukes of man and arrange (See section by Marshall). Preparation and Preservation of Material. It is obviously useless to collect material unless it is to be cared for in such a way that it will serve some useful purpose. The preparation and preservation of terrestrial arthropods is especially important in the tropics. The following generali- zations may assist the non-specialist, though it is recognized that special techniques may be devised to fit unusual situations. The commonest method of killing insects is with a cyanide jar. Material killed in this way should be protected in the jar by means of tissue paper, and should be removed from the jar before specimens become damaged by rubbing or by accumulated moisture on the sides of the jar. Bees, wasps, true bugs (except aquatics), ae flies, mosquitoes, lacewings, many Orthoptera, moths and butterflies, dragon- flies and damselflies, crane flies, and other fragile insects too numerous to mention should be killed in this way. Such insects should be curated at the end of the day's collecting. Failure to do this will result in stiffening of the appendages and excessive breakage. The moths and butterflies, dragon- flies, damsel flies,’ and crane flies should be folded individually in paper triangles or envelopes. This prevents rubbing and is the only satisfactory way to preserve such fragile insects. The remaining types of insects killed in cyanide should be spread out on layers of cellucotton in cigar boxes or ~ other safe boxes for shipping. The cellucotton layers should be double between each layer of insects so that specimens will not be damaged when they are renoved layer by layer for mounting. Specialists may wish to pin a few of the most fragile specimens such as small gnats and micro-Lepidoptera. If it is desired to do this, minute pins should be used and a regulation insect box with tight-fitting cover should be used to store the specimens. Small Hemiptera and perhaps some other insects may be glued to small paper points and stored in this same way, but this is only recommended in cases where pinned material can be cared for properly in the field, All of the other groups of insects and most other terrestrial invertebrates should be collected'in 70% ethyl alcohol. Small procaine vials, used by dentists in most varts of the world, may be found convenient for ee ial fay collecting the smaller invertebrates, but some large vials and bottles will also be needed. Land snails, which are such a characteristic feature of many Pacific islands, are better collected in water and transferred to alcohol after a few hours. This results in an expanding of the soft parts and greatly facilitates later study. Labeling, is of the greatest importance and should be done at the time the material is prepared. Because of the large numbers of specimens, it is best to label the material as fully as possible and make sure that these data accompany the specimens. The exact locality should be given in every case, together with the date and name of collector. Ecological data should be added whenever possible. Elaborate cross-reference systems, including index numbers and field notebooks, are to be discouraged because of the danger of loss of one or another of the essential elements in such a system. Storage and shipping. Numerous hazards are likely to beset the col- lector in the field, ‘the most important of which are mold, cockroaches, ants, and rats. To. avoid such hazards, material should be kept in tight boxes and not allowed to remain out on tables overnight. In some cases it is necessary to hang spreading boards or exposed specimens from the ceiling by strings, and cover the strings with sticky material or protect tables by treating the logs with a DDT solution. Usually it will suffice to finish all curating in a single evening and place the material in tight boxes. Mold is an ever present menace and should be guarded against by storing material in a hot locker or in glassine bags provided with silica gel. Specimens should be shipped to a museum or home base at the earliest possible time. Specimens may be shipped in the original boxes in which they have been layered by adding naphthalene (which keeps out museum pests and seems to prevent mold), or by adding silica gel crystals. In either case the boxes must be wrapped carefully, and preferably with water-proof wrapping because of the hazards of rain and salt water in Pacific island transport. Alcohol vials should either be full or a cotton plug should be inserted to prevent damage due to the action of bubbles. Vials should be wrapped separately in paper and then mailed in rigid cartons. Equipment List. Collapsible net, e. g. nena ae net, or other (preferably with nylon bag) Cyanide bottles (several sizes--Alka Seltzer bottles with a cork top are a useful size, and smaller jars of plastic material are valuable because they are unbreakable) Procaine vials and larger bottles 70% ethyl alcohol (this may be available locally at medical supply depots) Paper envelopes or cellophane envelopes, glassine bags Silica gel Naphthalene Cellucotton Cigar boxes, or other wooden boxes, unless air-tight containers are available, in which case silica gel is absolutely essential to prevent mold and rotting a7 Insect box and pins, including minute pins (only if handled with great care) Coleman lantern and white gasoline Hand lens, forceps, scissors Barking tool (crowbar, screwdriver, or heavy knife) Small tins or pill boxes for rearing Cotton Conclusion. Finally, it: is a good guiding principle to concentrate on small forms, because these are so often overlooked. It is also axiomatic in good collecting to take a series; one specimen of a new species is not enough and excess specimens can easily be thrown away. The opportunity to collect ona remote island ney not come again for many years, if ever. A useful reference oe peered iabinctiiy technique is the pamphlet by Oman and Cushman (U. S. Department of Agriculture, Miscellaneous Publication No. 601, Washington, D. Cay 1946. Price 15 cents). This small pamphlet ‘includés ucerets ones of the common orders of insects, as well as specific instructions for collecting and preserving insects. Most of the above-mentioned equipment can be obtained from supply houses any place in the continental United States. One of the best known ‘sources of such equipment -is Ward's Natural Sothern Establishment, 302 Goodman Street, North Rochester, New Ogee ; : 1-73 = THE BERLESE METHOD OF COLLECTING SMALL INSECTS AND OTHER ANIMALS FROM LEAFMOLD, SOIL, MOSS, OR OTHER SIMILAR MATERTALS qa aN by Joseph P.E, Morrison The apparatus consists essentially of a funnel supported in an upright position with a shallow fine mesh. screen-bottomed tray,at the top,to hold the sample of leafmold, and so forth, within the rim of the funnel, and a F homeopathic vial half-full’ ofi preserving fivid attached (outside) ne the neck of the funnel sitaipscaucdt catch the insect and other’ apnea that fall out of the sample. ee: Cel GKe The material such as leafmold is dried by the gentle application of’. heat from above.. If direct sunshine is insufficient, or the humidity too high, or the material is to continve overnight, as is usual, a small light. bulb (25 or 30 watt) under a conical shade is used’ to direct heat onto the: top of the slsyiggre The screen on which the sample is spread. in a layer one or two inches. thick should be about 20 mesh to the inch. The inside of the funnel and neck must be smooth inside, so that the specimens falling onto the steep slope will not catch anywhere on. projections, but enc up in or on the pre-.. . serving fluid in the a. eine eis) os eh) Many of the small insects, mites, and so forth, so collected, may te floating on the 70% alcohol used as a preservative. A drop or two of ether, added when the vial containing specimens is removed from the ap- paratus to be labelled and: stoppered, will readily permit the floating specimens to sink into the fluid, so they will not be lost or destroyed around the cork or stopper:: After labelling the vials completely, they are set aside, all sorting of specimens being done in the home laboratory. m/e COLLECTING MOLLUSKS ON AND AROUND ATOLLS by Joseph P, E. Morrison In order to obtain the greatest scientific benefit from any specimens ~ collected, the place and/or zone or region of the atoll must be carefully recorded. In mollusks, as in most other animals, one finds different species in the different habitats or environments in, on, and around an atoll. Marine shells may sometimes be collected in good condition, abundance, and in great variety from beach drift along or just above the high tide line. In the absence of time and equipment available to collect living marine specimens, and prepare or preserve them, the collection of drift material, particularly of the smaller species, makes a very valuable contribution. In the absence of preservative, the smaller species may be dried thoroughly, then packed in newspaper, and sealed in boxes or cans for later cleaning. Iron stain on mollusks shells, from rusting of tin can containers is to be avoided; when packing material in cans, wrap well with several layers of newspapers, etc., between the shells and cans. Field cleaning of marine mollusks taken alive may be accomplished by killing and cleaning them in jars of fresh water. Specimens from different localities should always be kept separate to maintain the locality records. A change and washing in fresh water once or twice a day, every day until the shells are clean (that is, do not smell too much) is necessary for this method. If left too long without changing, the acids of decay will etch the shells, leaving them poor and chalky in appearance, Certain small marine (and fresh water) species may be collected in numbers in a minimum of time by vigorously washing or shaking them off rocks or plants in a bucket full of water. The water, if muddy in the bucket, may be cleared of mud and floating plant material by repeated filling and pouring off the top 3/4 of the water. The bottom residue containing snails is then poured out of the bucket (through) on to cheese cloth or field screens or a dip net, and dried or preserved. Pulmonate fresh water snails (that float when disturbed) may be recovered in this process by pouring all water out of the bucket through cheese cloth or whatever other fine strainer may be available, The smaller land species of mollusks are usually the only types found upon atolls. These may be collected in numbers by taking quart size or larger samples of leaf mold or other detritus where ever these small snails are seen in abundance in the leaf mold or on or in the soil surface immediately under the leaves, etc. The mass of large samples may be greatly reduced after drying by screening the leaves, larger rock fragments, sticks, etc., out through screen of 4 or 6 mesh to the inch. The fine material containing the smaller land snails may then be dried and sacked or boxed for later separation, at the museum, of these minute land species. Excessive dust may be screened out through mesh as fine as 20 mesh to the inch. Caution: Ordinary window screen (16 mesh to the inch) may pass most of the fine species with the dust. If there is excessive sand or foram sand or rock in the sample, the sample may perhaps be concentrated, after drying, by water separation. In this water separation, the sample is put in a bucket or jar of water, the rock fragments sink, and the minute snails, etc,., are skimmed off the surface, 15m thoroughly dried again, then packed. Drift along fresh water lakes and streams, as well as on ocean beaches, is sorted by nature in this fashion; all the species of snails may not float, so check the residue before it is discarded. Particular attention should be given to the recording of exact data on the habitat zonation probable on all atolls. The zones expected will probably include: the windward reef edge (surf beaten); the lee reef edge, reef flats, or reef pools of shallow water at low tide; channels through the reef or between islands; island shores; the shallow margin of the lagoon; and the deepest waters of the lagoon. On island shores zonation should be recorded in relation to the distance inland from shore as well as the type of dominant plant cover or habitat. In the lagoons of atolls zonation may be present along the inner reef edge, the inner shelf of the atoll ring (up to 5 fathoms?) and the general lagoon bottom. Shoals or "coral heads" up= raised from the lagoon floor may be particularly interesting in fauna, if facilities are available for such collecting. Even one person in a small row boat or canoe can reach the lagoon bottom fauna by dredging. A prac- tical small dredge for work by one person should have a frame with an opening of about 5 x 12 inches, and a bag about 18 inches in length. If a launch and other equipment is available, a dredge of about 10 x 20 inch frame, and a 30 inch long bag will obtain bigger and better samples from the lagoon bottoms. This is about the maximum sized drag-dredge that it is practical to use without power winches, The frame of a dredge may be made of any readily available metals such as bar or strap iron. If all parts are galvanized it will last longer in salt water, and will not rot out the cloth dredge bags by rusting in such a short time. The outward flare of the dredge frame is not always necessary; a simple plate, with inner edge beveled, will work well on most occasions. The dredge bag is made of netting of "chalk line" cotton cord with 1 inch mesh. If special dredge netting is not readily available, burlap sacking can be used; this will retain even the smallest snail shells, The bag will last longer if it is wired onto the frame with copper or monel wire. It is usually advisable to protect the net bag from rocks and coral by fastening an open ended sleeve of canvas to the frame around the bag. The dredge line of 3/4 inch or 1 inch rope (large enough to provide a good grip for hand hauling of the dredge) is securely attached to only the longer of the two triangle halter frames. The short one is tied to the longer (arrows in diagram) only with two turns of marline cord or string. This cord will break, and release the dredge from obstructions, before the line breaks. Usually this is a sufficiently effective device, even on coral bottoms. If the dredge "hands up," backing up on the line and pulling in the reverse direction should clear it. About three times as much line as the depth of water is the proper amount to use for this type of dredging. The speed of dragging should be slow enough so as not to pull the dredge off the bottom or have it only skip along. With experience one may judge even the type of bottom the dredge is moving over, by the simple method of holding one hand on the dredge rope and feeling the vibrations from the dredge. te a scene = fe SS ee eee Pf i ce sl LLLP LTL: ee ee r a en eres re ee ee ern mee Kee = ate pln ry emnbs e e 10 mesh 20 mesh | 3 gem ae AE ia ne OCT aE. ae metal hut ae angles metal frame nesting wood frame nesting Fel Bye giSieCe yi EY Nu ao dredge line marline oY ( string safety at ah - J NSA os Pe } rolease rity ‘ We i 4 omega! ‘ cs, yokes frame mesh bag or burlap ual uy / / if ox ae ae | a Va a Ma gil 1eTl oD) vy UN ee Eun a a a Nee “ SO Sache ae frane bolted or riveted F ‘ahh ty ay s Op MEERA iat Phpilsa' i F eT jes “78 | METHODS OF COLLECTING MARINE INVERTEBRATES ON CORAL ATOLLS by Robert W. Hiatt INTRODUCTION While it would be presumpt uous to inform a feliow marine zoologist about collecting methods along our continental coasts it is certainly justifiable when coral atolls are concerned, for such coilecting is very different and requires special information and experience. I trust that our several years of experience in collecting on coral reefs will serve to introduce the uninitiated to the problems and their methods of solution, so that his preparation for and handling. of work in the field will be greatly facilitated. A check-list of equipment appears in the section on "Instructions for Marine Ecological Work on Coral Atolis." The first portion of this section of the Handbook concerns ways and means of collecting; a second portion is devoted to methods of preservation. ASSEMBLING EQUIPMENT Emphasis should be placed upon the selection of special equipment for the special collecting tasks ahead. Both clothing and equipment should be selected for its utility, not merely because it is the most readily available. Moreover, wherever possible, carry out all the critical stages of preparation personally. Remember that you will be a long way from any source of additional equipment once you reach an atoil -— in fact, so far that you cannot expect supplementary shipments of forgotten material. Obviously, another person never cares quite so much about the material as you would, nor would they know precisely what use was to be made of the tools. SAFETY MEASURES IN THE FIELD Because medical care is frequently unavailabie, perhaps for weeks, it is important to be even more cautious than would normally be necessary. Certain safety rules should be kept in mind. These are: 1. Never collect alone in boats, when using diving gear, on rough slippery shores, or while swimming. 2. Exercise care in handling poisonous or venomous species, e.g., certain echinoderms, coelenterates, annelids and fishes. Avoid handling all sea urchins having slender, long spines, starfishes having sharp spines (Acanthaster), stinging corals (Millepora) and other corals if an allergic reaction is noticed, medusae and siphonophores, annelids _ with long whitish setae (Eurythoe), and scorpaenid fishes. 3. Treat cuts and scratches immediately with an antiseptic and merthiolate. Recurrent wetting hinders healing. 4. Keep a weather eye on sizeable sharks in the neighborhood. These ubiquitous creatures are curious and often come close to you, possibly because their eyesight is poor. Our exper- ience shows them to be more frightened of you than you are of them, once they see you. It is often stated that if this is true, their fright knows no bounds. ~79= 5. Give moray eels a rather wide berth. They will not attac. a wader or a swimmer unless the person inadvertentiy steps alongside their shelter or thrusts his hand into a hole which he has not examined previously. Morays are usually easily seen by a person wearing a face mask or using a glass=bottom box. 6. While collecting on reefs it will be found desirable to wear a shirt, long trousers, wading shoes, and naa cloth or leather gloves. FIELD NOTES Getting back with the animals in good condition is only one part of the job. For purposes of modern taxonomy, for ecology, or even for obtaining further | specimens to carry on. certain work, collecting is meaningless without at least minimal data. The marine zoologist is handicapped in his note-taking because it is impossible to write fieid notes while in the water. Consequently, he must make his collections in a systematic manner so that a single set of notes will suffice for a number of animals. In all cases field notes should be written as soon as the worker reaches the shore. When.collections are made in quadrats with code numbers, and in certain species of corals within the quadrats, note-taking can be kept to a minimum. Well-planned field procedures can do much to reduce copious note-taking for the vast number of species living on coral reefs. For many species of invertebrates colors and color patterns are of taxonomic ‘significance. In such cases reference should be made to Ridgway's, Color Standards and Nomenclature (1912). METHODS OF COLLECTING Unlike continental shores, where collecting can be accomplished on exposed surfaces as the tide recedes, coral reefs offer very little intertidal area so that collecting in general is accomplished only under water. Thus, specialized underwater equipment is needed. A brief account of such equipment follows. . Underwater Equipment Glass=-bottom boxes These may be made or purchased as desired. They are perhaps the least useful of all underwater collecting. devices because (1) the observer's, fieid of view is highly restricted, (2) one hand is required to hold the box, thus leaving only one hand free for collecting maneuvers, and (3) the coilector is restricted to very shallow water. .I. recommend against this piece of equipment. Face mask Face masks of many types are inexpensive and readily avaiiablie. Their chief advantages are (1) the observer‘sfield of view is limited only by the turbidity of the water, (2) both hands are. free for collecting maneuvers, and (3) the collector is not restricted to shallow water. He may be restricted in depth of eollecting by his skin-diving ability, but he is restricted in observations only oils — BO by the turbidity. of the water. In. Clear, tropical waters it, is often possible tO see Shei to a oa of two: ‘hundred Pape ; In e Mireee water the pole one may wear, wading shoes, but) in feeder water. where diving is essential or where considerable distance is to be traversed, swim fins,are useful... With them one can propel himself entirely with the feet, leaving the. hands free ‘to carry equipinent or for collecting. * With .a little practice a person can dive 15. to 20 feet, Experienced skin- divers can collect, at 50 or 60 foot depths. _ This outfit supplemented by a gebtectans Pap and inner tube Egat has. proved highly satisfactory. yee slope one gear Aino’ Both een and compressed air diving equipment are available. Since the advantages and disadvantages are held in common, both types will be con- sidered together. The chief advantage of this type of gear is the extended period of submergence permitted while observing or collecting. With oxygen equipment submersion periods up to two hours are possible at a depth of 25 feet, while compressed air outfits will maintain a person at this depth for © only about twenty minutes. The chief disadvantage of such equipment for atoll work is the necessity of transporting tanks of oxygen or air. Far less oxygen is needed than compressed air, so less shipping space is involved with the. oxygen type. However ,. the increased safety of the compressed air type makes it more desirable if shipping is no problem. eee ee aoe gear If adequate personnel and shipping space is available the pump-type diving gear is quite useful and probably the safest. However, a minimum of two persons above for each person submerged is necessary, thus virtually eliminating such equipment from consideration in atoll work. The submerged observer is also restricted in his movements because of the attached air hose and line. The greater depths accessible and the longer periods of submersion are the chief advantages. m Collecting on Different Types of Substrata The Coral reef Methods of collecting Fishes. on coral reefs are detailed in another section of this Handbook. It’ should be noted at this point that cephalopods are also killed by. rotenone and should be collected along with the fish. i ee on coral reefs” may be collected’ in a variety of ways. Perhaps the most important method is the collecting of living or dead coral heads: or porous coral boulders for their entrapped invertebrate fauna. These coral heads and boulders should be ‘carried: to shore and there broken apart with a geology pick or similar instrument. In interstices of the branching : heads and in burrows in the solid heads will be found a multitude of animal types. Crustaceans, molluscs, polychaete Worms , bryozoans , tunicates, sipunculids, hydroids, echinoderms , sponges, | ‘nemert earls and small fishes are easily and efficiently collected in this manneér. °° “Indeed; if this method is not used, a large proportion . of the spécies” ‘present will be missed. Moreover, such a PeEnOe lends itself admirably t6° collecting in transect quadrats. Studies on biotic interaction in specific coral head biocoenoses are also facilitated io employing this method . AIG -81- Many species of gastropod wigdases, particularly of the families Strombidae, Cypraeidae, Conidae, Muricidae, Fasciolariidae, and Buccinidae are found on rock or coral rubble between coral heads. A host of species will also be found in burrows or crevices in the reef rock. Many boulders, particularly near shore can be over=turned to reveal the motile species which seclude themselves beneath it and the attached species which grow on the lower surface. Species of Tridacnidae will be found occasionally either attached to or lying free on the reef rock. Near the outer reef edge the collecting becomes more difficult because of the rougher nature of the coral and nullipore growth. Here echinoids will be found secluded deeply in holes where only a pry-bar can be used effectively to expose them. Boring species of tridacnid clams can also be extricated in this manner from their attached positions deep in coral heads. Usually the outer reef edge area is rather weakly consolidated so that large sections of it can be fragmented with a pry-bar. Oftentimes this reveals a host of species, particularly crustaceans, ophiuroids, and gastropods which may not be collected elsewhere, Collecting in sand and mud. Most animals in sand or mud are either burrowers or exist just below - the surface making them relatively difficult to see and collect. Perhaps the most successful method of collecting under these conditions is to become acquainted with the habits of the animals. For example, many gastropods live just under the surface in sandy areas and carry on most of their activity of moving about at night. If the sandy area is not disturbed too greatly by wave motion it is frequently possible the next day to observe through a face mask the slightly excavated trail in the sand which marks their most recent movements. By diving down and scraping your hand through the top layer of the sand, marked by a slight elevation at the end of the trail, you can collect the gastropod. These sand trails made by subsurface movements of gastropods are ubiquitous in sandy regions of atolls. Members of the families Terebridae, Conidae, Olividae, Mitridae, Alectrionidae, and Cerithiidae are common in such locations. Among the echinoderms cake urchins and heart urchins are often abundant just beneath the surface of the sand. The animals usually disturb the normal appearance of the sand above them, thus marking their position. Probing and digging in the sand at that location will expose the specimen. Holothurians frequently lie just below the surface of the sand, and they may be found and exposed in the manner described for the echinoids. Many species are true burrowers, making permanent burrows which are easily seen. These animals are usually so deep in the sand that it is not feasi- ble to extricate them without special equipment. Such equipment consists of a shallow box with a wire screen bottom into which the sand or mud may be shoveled and strained. Many species of pelecypods, crustaceans, polychaetes, and echino- derms will be gathered in this and in no other fashion. Use of traps, nets, dredges Certain species of invertebrates, particularly the larger crustaceans and those inhabiting depths not available to the average skin-diver, can be collected only by traps, nets or dredges. Such species as stomatopods, palinurids, scyllarids, portunids, etc., may be taken much more efficiently in suitable traps than in any other manner, It is often impossible to secure suita- ble traps from native fishermen, so considerable thought’ should be placed on constructing wire take-down type traps which weigh but little and occupy only a small space. The trap should be constructed somewhat along the lines of the accompanying diagram. \ : 1 [ *> i. m i. Pout THe yee ie ea h ay | ve | ba ay =93- Fish make excellent bait for such a trap. Baited traps such as this will also attract moray eels, so frequent inspection of the trap may prevent the loss of valuable specimens to the eels. With the exception of smail dip nets very little use can be made of seines for invertebrates. The field collector is advised against taking large nets where space is limited. The naturalist's dredge is a most useful tool if a power boat is available to haul it. Usually such a boat is not available, so make inquiries before you send such equipment into the field. Native sailing canoes are not very suitable for dredging. Plankton nets Much has been written about collecting plankton from both qualitative and quantitative standpoints. The field man is advised to investigate suitable nets for the purpose intended. A letter to the Secretary of the American Society of Limnology and Oceanography wiil make availabie the Se Gace on plankton sampling issued by the Society. METHODS OF PRESERVATION Perhaps the most important point to remember in preserving specimens in the tropics is that bacterial and physiological decomposition is accelerated by the high temperatures, thus more attention must be given to methods of immediate preservation. Preservatives FORMALIN is perhaps the most useful liquid for keeping animals temporarily, but not for preserving them permanently. It has greatest utility in that a much smaller volume of liguid need be carried, thus, it is very important for the atoll worker. If no restriction is placed on the amount of collecting gear which may be taken, more ethyl alcohol and less formalin is desirable. There are two principal objections to formalin. First, while it is a very useful provisional fluid for many animals which are not contractile, and especially for those which contain no lime spicules, skeletons, or shells, it will attack the . calcareous portions of animals containing such and dissolve them or cause them > to lose brilliancy, or both because of the free acids in the fluid. Substantial relief from the free acids may be achieved by neutralizing the formalin with an excess of either sodium carbonate or lithium carbonate. Usually it is suffi- cient if an excess of the salt be kept as a sediment in the stock jar or barrel of formalin. Second, unless specimens are transferred to aliconor after two or three years they begin to disintegrate. Preservation of different types of animals should be accomplished with different concentrations of formalin. With large and relatively impervious animals such as fish, large crustaceans and cephalopods under tropical conditions it is necessary to use up to 30 ver cent formalin. If refrigeration is avail- able they should be preserved under refrigerated conditions. In any case injec- tions of formalin into the body cavities and into thick muscuiar areas should be made. For smaller animals the normal concentration should be approximately 10 per cent formalin, while soft or gelatinous animals should be preserved in 2 to 5 per cent formalin, the siniipitt rule being that the softer the animal, the weaker the formalin. With formalin, as with other preservatives, only one, or at any rate only a very few, objects should be put into the same receptacle at the same time, and there must be a good amount of fluid in proportion to the animal matter present. Either fresh or salt water may be used in making the solutions. The sea water solution often preserves the transparency of gelatinous bodies better than the other. For certain flaccid animals the hardening accomplished by formalin is advantageous. ALCOHOL, ethyl preferred, is without doubt the most indispensable liquid for permanent preservation. Fully denatured alcohol is to be avoided because of the deleterious effect of the denaturants on specimens. In certain instances alconol is required for the initial preservation, so a small amount is essential in the field. In most instances, except where anesthevtizing is necessary, specimens may be dropped directly into 70 per cent alcohol. Certain soft—bodied forms such as jellyfishes must be first placed in 50 per cent alcohol and after 6 hours or longer placed in 70 per cent. The alcohol must be diluted with fresh water, as a white precipitate hare appears and covers the specimen if sea cae is used. Meese: which possess calcareous shells, spicules or other structures of a similar nature, should if possible be placed directly in alcohol, since the acid formalin may seriously damage the specimens even within a short time. it is not necessary to use alcohol in concentrations in excess of 70 per cent, indeed, it may be harmful to do so. The effect of both formalin and aTrono! is to a ponsiderapls extent to replace the water of the tissues and thus prevent bacterial decomposition. The removal of the water not only prevents decomposi- tion but at the same time stiffens the tissues. Alcohol at 70 per cent accomplishes this replacement most efficiently. Regardless of whether formalin or alcohol is used the colors will fade eventually. Anesthetizing reagents - MAGNESIUM SULPHATE, commercial grade, is ‘perhaps the most useful and least _ expensive of all the narcotizing reagents. Animais to be narcotized shouid be carried to the laboratory in sufficient sea water to keep them alive, then placed in a pan with sea water barely covering them. Crystals of magnesium sulphate should be added roughly in concentration of one tablespoon per quart of sea water. As narcotizing progresses more reagent may be added. This method works well for gastropods, echinoderms, sipunculids and echiuroids. CHLORAL HYDRATE, commercial grade, is useful in a very weak solution, from 0.1 to 0.2 of 1 per cent in sea water, for narcotizing forms difficult to pick off of algae or madreporarian beads. CHLORETONE is useful in weak aqueous solutions for narcotizing small forms. ALCOHOLIZED SEA WATER is also a useful anesthetic for quieting small marine species, particularly the worms. Care should be taken not to put many individuals together during the process as a copious secretion of mucous results in an inextricable entanglement of the tig thus rendering the specimens useless. sESuSPortation of specimens There are a abies of points to remember from the standpoint of returning specimens to the laboratory and from there back to the institutions where they will be handled. ae Be Don't crowd your collecting containers in the field. Don't place antagonistic species together in the same cont ainer, é.g., large carnivores with small species, crabs with soft- bodied types, etc. Only animals taken at a particular sbebien should be placed in one container. Carefully segregate mucous secreters from other specimens in the field, as the entanglements caused often result in worth- less specimens. If specimens are to be returned to the laboratory alive, suffocation must be avoided by (a) having only a small number in each container and by (b) having a minimum of water plus a good deak of algae for cover. Aeration may be supplied by an Oxy-bomb outfit obtainable from A. Daigger & Co. 159 West Kinzie Street Chicago 11, Lilinois Avoid leaving containers in the sun. If specimens are returned to your institution in metal cans, they should be opened immediately upon arrivai, and the specimens should be transferred to 70 per cent aleohor in glass storage jars. Notes on preservation of particular animal groups PORIFERA should never be preserved in formalin as the tissue degenerates at once, leaving only the skeleton which may also be in poor share. Drop the specimen into 70 per cent alcohol immediately and renew alcohol when it becomes discolored. Horny sponges. may be dried by first washing them in fresh water for a few hours, then they should be placed in 70 per cent alcohol for a day or so, and then placed in the sun to dry. Sponges treated this way will not have an offensive odor. For best taxonomic results a small piece of the sponge should be cut off and dropped into Bouin's fixative. COELENTERATA provide a variet by ler. situations which must be dealt with individually. Corals, including the madrepores, millepores and other hydrozoan types with stony skeletons should-be broken from their attachments as carefully as possible. If a piece of a large coral head is taken, it should be obtained in a typical part of the head and should contain exposures in as many planes as possible. If the atoll is in a rainy belt all that need be done-is.to place the corals in the sun on a slightly elevated structure where drainage is good and where no vegetation will touch them., Alternate wetting and drying, accompanied by frequent turning of the specimen, will soon clean and bleach it. Where rain- fall is slight and undependable, immerse the corals in containers of fresh or tepid.-sea water until the tissues disintegrate. Wash them well and place them in the sun to dry. Labels, especially temporary ones, during the rotting and bleaching period, should be on heavy Reeee ener eae attached with stainless steel wire. Anemones and hydroids should be covered Lae sea water to which a small amount of magnesium sulphate has been added. When they have expanded completely and are fully narcotized, transfer them to formalin. The soft-bodied siphonophores and scyphozoans may be immersed directly in approximately 5 per cent formalin. This results in highly contracted tentacles, but this does not impair the specimens for taxonomic study. NEMERTEANS present great difficulties because of the highly contractile body and danger of fragmentation. It is best to place them alone in a sizeable container of sea water to which chloral hydrate is added. Allow them to remain there for about 12 hours, then drain off the nareghsee nels solution and replace it with 70° Per cent aicohol. — ne : SIPUNCULIDS may be narcotized with chloral bpoeaee in sea Dee until the tentacles are expanded. Often it is necessary to wait until the proboscis is extended, then seize the body with one hand and with a pair of forceps seize the extremity of the proboscis to keep the tentacle extended, then immerse the animal into 10 per cent formalin or 70 per cent alcohol until it dies. BRYOZOANS are somewhat resistant to anesthetics but some extension of the zooecia may be preserved by narcotizing slowly with 70 ver cent aicohol added to sea water. Preserve in.70 per cent alcohol or in 5 per cent formalin. POLYCHAETE worms are difficult to. preserve properly in the field in the best taxonomic condition. If they are not completely narcotized at fixing, they will contract greatly, twist out of shape and frequently will fragment. Narcotizing may be accomplished by separating the worms into finger bowls with just enough sea water to cover them. . Then add alcohol slowly and let them stand for several hours. Just before they are completely immobilized place them in 5 per cent formalin to which a little glycerine is added. MOLLUSKS presént special problems because some are shelled while others are not. The shelled gastropods are identified by. shell characters only, thus it is unnecessary to preserve the soft parts unless they are needed for special study. The chief problem is to remove the soft parts from the shells without damaging -87- the structure or brilliance of the sheil. I have found it most satisfactory to place the living shelled gastropods into a container of sand so oriented that the rotting fleshy material will drain from the shell into the sand. The bottom of the container should be perforated so that all decaying materiai can drain or. - be leached out of the container. . The container should be placed on the ground so that ants can enter the decaying shells and hasten, the removal of organic substance. Remember to cover the container sufficiently well to prevent the entrance of land hermit crabs which will be attracted to the area. If the hermits are able to enter the containers you will soon find that many beach-worn shells have been traded for your prize specimens which now decorate the abdomens of these ubiquitous robbers. When most of the decaying material is out of the shells, after a month or more in the container, they may be washed in water, labeled and packed away. Tectibranchs and nudibranchs should be anesthetized using chloral hydrate or magnesium sulphate. When they appear immobile preservation may be made with alcohol or formalin. Large specimens should be PS eee with preservative. Pelecypods, except for certain boring species (Teredo, Bankia, Rocellaria, etc.) are identified by shell characters so the soft parts may be discarded. In both shelled gastropods and shelled pelecypods some specimens have brilliant clean shells (sand-dwellers) and the shells of others are more or less encrusted with other organisms, especially calcareous algae. It is virtually impossible to make shiny shells from these latter specimens, and indeed, for taxonomic purposes they should be left in their natural condition. The cleaning of such shells has little to do with scientific endeavor so the methods are omitted here. Cephalopods can be dropped into 10 per cent formalin after the body cavity has been injected with preservative. No particular problems will be encountered with this group. ; Te ay et CRUSTACEANS present individual problems in fixation, but in general all should be preserved in 70 per cent alcohol to which glycerine is added in a ratio of 1 part to 10. The greatest danger in preserving crustaceans is the loss of appendages. Glycerine will help in this cai by maintaining some flexibility in the arthrodial membranes. Shrimps except for snapping shrimps (Crangonidae) can be preserved without narcotizing in alcohol and glycerine_or formalin and glycerine. Snapping shrimps are prone to autotomize [detach - Ed./ the large chela, especially if they are © dropped directly into the fixative or are crowded with other specimens. Unfortunately some taxonomic characters are based on the morphology of the ein cheliped. If shrimps of more than one species are crowded together and many: autotomize their'chelae, a near hopeless confusion results in matching Shrimps - with cast appendages. I find it preferable to permit them to die in fresh water or tepid sea water with but one type in a container. Under field conditions this is difficult to do, but if a sufficient number of small vials are taken along it can be accomplished. Many anomurans and brachyurans will autotomize appendages if stimulated too strongly with the fixative. The impermeability of the integument prevents rapid ~88— penetration and death, thus the prolonged thrashing about often results in mutilated specimens. It is best to put them into fresh water or leave them exposed to air until they die. Land crabs (Cardisoma, Geograpsus, Coenobita, | Birgus, etc.) present special problems because exposure will not kill them and fresh water is not particularly effective. The only suitable method is to drop them into formalin. Fortunately,.they do not autotomize the. appendages readily. Large specimens should be, injected with preservative. Where it is necessary to transport many preserved specimens in a.common container, each specimen should be protected by wrapping it adequately. in cheesecloth which is tied securely with string. Whenever possible, flex the. appendages so that the tied animal will | be compact. I find it best to preserve hermit crabs in their shells so as to | have precise information as to the shell occupic’d. Oftentimes both the crab and its shell will require identification by specialists, Most specimens can be | pulled from the shell rather easily following preservation. | Barnacles must be pried from their attachments ‘and preserved in 10 per cent formalin. It is not necessary to relax them first as taxonomic determinations -are made from the valves. Wet preservation is superior to drying because it insures the safekeeping of the valves until VIN are cleaned for use by the taxonomist. Smaller crustaceans such as ostracods, copepods, amphipods, isopods, stomatopods, etc., should be dropped directly into 70 per cent alcohol or 5 per cent formalin. ECHINODERMS siheerit a variety of problems of ceesee ion! The hard-shelled echinoids may be dropped directly into 10 per.cent formalin. Large specimens should be punctured through the peristomiai membrane to insure internal _ preservation. After a day or two they may be placed in a shady spot to dry. The sun will bleach the colors rapidiy if the specimens are not shaded. Asteroids and. ophiuroids may best be handled by anesthetizing with MgSO), in a pan with Sea. water barely covering’ the specimen. ‘When the asteroids have flattened out remove’ them to another pan of 70 per cent alcohol or 10 per cent formalin. After two or three days in the fixative they may be dried in the shade. Ophiuroids must be handled with extreme care as the arms are readily autotomized. Do not crowd them in containers in the field and place them in MgSO, solution as soon as: possible. When the arms no longer react to touch arrange them all parallel to one arm so that a comet-shaped form results. Remove the narcotizing solution and replace it with 10 per cent formalin or 70 per cent alconol. After two days lay them out in the shade to dry. Holothurians require more care than other echinoderms, because they have soft and highly contractile bodies and all are furnished with tentacles which contract or retire within the body on contact with a reagent. Moreover, certain -Species are prone to expel their viscera shortly after being immersed in the preservative or anesthetic. While this behavior does not impair the use of-the specimen taxonomically, it is unsatisfactory in many other ways. Allow the animals to relax and expand in a pan of sea water to which MgSO, is added. After a few hours of such treatment many species will become narcotized with their tentacles expanded. - They should be injected with 70 per cent alcohol, wrapped in cheesecloth, and tied, and dropped into 70 per cent alcohol. Never leave sea cucumbers in formalin for any length of time because the acids will -89- quickly destroy the tiny calcareous cutaneous spicules so important for correct identification. Thin-walled apodous species must be nandled carefully to avoid fragmentation, The tentacles cannot be inverted within the body so complete relaxation is unnecessary, indeed, it is usually to be avoided since many of these species are several feet in length. It is best to place them in a shallow pan containing 50 per cent alcohol. When killed they should be transferred to 70 per cent. Crinoids have a tendency to fragment regardless of their treatment, but this can be minimized by immersing the animal in 90 per cent alcohol, then violently shake the vessel to hasten death. Usually the field collector has no suitable containers for transporting these animals to the institutional laboratory in wet preservative. J have found it useful to hang the specimen in a shaded place to dry after it has been in preservative a few days. They must be packed separateiy in rigid boxes for transport. LOWER CHORDATES may usually be dropped directly into 10 per cent formalin. However, if the specimens are wanted for microscopic structure the investigator shovld consult a guide book on animal micrology. If time and facilities permit it is better to narcotize cephalochordates and tunicates with chloral hydrate and enteropneusts with alcoholized sea water. 34) yt ome satuagie huonedvee esoare hia be orig oc - soy wo lope’ ox TREE ecoucind ad) te ‘poste oes ae bk wk Rea ort, aE ts ee ‘ sy, Sy VS snap ge w x 2a/ J ; i “ a jaunts Dye or } Gerin; Yoxit’ i’ pot =p ; bias» £ nt SIIIITE IAIOSRO" i iA it ti party ny “el a geelorori miea = isitae ody J J thee id aetabrotse: al 4 4 ‘ 4 ‘= CF Ont y P 7 se ; * : ch ie een : wre “ x Pde al s 4 r , omQQ = poi DIRECTIONS FOR COLLECTING, PRESERVING, AND SHIPPING FISHES by Leonard P. Schultz COLLECTING FISHES The usual methods of collecting fishes are baited hook and line, trolling, spearing, dredging, trawling, attracting them to the surface with a light at: night, and using various nets and beach seines. Poisoning is the most important method for shallow water fishes of ocean reefs. Powdered cubé or derris root with a ae SC pao of rotenone ea or rotenone in concentrated form is” best. Coilecting by Means of Fish Poisons on Coral Atolls” Thirty-five minutes before the tide reached its lowest point, the dry powdered root was placed in buckets or any suitable container and mixed with water to a thick chocolate malted milk consistency, allowing about 20 minutes for one man to mix 25 pounds. By squeezing and stirring with the hands, as water was gradually added the powder soon formed a thick mud. Ten minutes before low water, the distribution of the mixture began. The stupefying of a’ great variety of fishes with rotenone was most successful at the lowest fob (2 : of water. oF The success of this operation depended on determining the strength of ‘the eurrents and depth of water. A little of the mixture was tossed into the water, and the direction of movement of the small, light brownish cloud, watched. After several such tests, assistants, each with a bucket or two of the mud, were stationed in the water and the distribution began.- In water 4 to 5 feet deep, with a slow current, the mud was thrown out permitting the little soft pellets to dissolve as they settied toward the bottom, forming a light brownish cloud. Twenty-five pounds of the dry powder formed a cloud about 100 to 150 feet long by 50 to 75 feet wide. It was highly effective if. it took 10 minutes to pass any one point in water above 80° F. When used at lower temperatures the fishes must be exposed for a longer time. Usually a part of a bucket of the mud was reserved to strengthen the cloud after it had traveled a few - hundred feet. This precaution was advisable, since the currents did not always behave as predicted. Shallow Water Reef. -- It was learned through experience which shallow water habitats (to a depth of ten feet) were suitable for collecting fish with rotenone. An area: with an abundant growth of coral heads in about 3 to 4 feet of water, down to ten feet in pools, with narrow to wide channels between thé various kinds of corals, and a wind blowing the surface ies more or less shoreward, was the most ideal situation. Many kinds of fishes in the areas treated floated for a few minutes, then sank to the bottom. Some were picked up while they were violently swimming ~ more or less in circles. A greater quantity of fish appeared at the surface than were recovered immediately. Those that drifted ashore were recovered but those that got over deep water were often lost. * Schultz, COPEIA, no. 2, pp. 94-98, 1948. =9ie Immediately after introducing the rotenone, recovery of the fish started, but it was inadvisable to enter the area in which the treated water would flow, since that drove the unaffected fishes away. As soon as the water cleared, those fishes that settled to the bottom were collected. Two or three men continually wandered over the treated area, picking uv the svecimens in fine- meshed, bobbinet dipnets, 14 or 15 inches in diameter and 25 to 30 inches deep, with a 4 or. 5-foot-long lightweight wooden handle. As the water-laden cloud of rotenone drifted onward for a thousand feet or more, it spread out, gradually becoming so diluted that it lost its effective- ness. When the water appeared as a light, tan-colored cloud, it was.most effective since it retained its stupefying properties yet was not so much con- centrated as to be detectable by most fishes. Sharks, apparently able to detecu small amounts of the rotenone in the water, left the area until the cloud had passed. They then returned to feed on the sick and dead fish, sometimes becoming troublesome. With only one or two 3 to 6-foot-long sharks feeding on the sick fish, the skin diver can keep watch of them. However, when two or three of rete voracious creatures become too bold, as on one or two occasions, the ichthyologists left the water. : The searching for the demobilized fishes was done by means of a face ete covering eyes and nose, swim fins on the feet, and a dipnet. With a face mask, both hands were free to devote to picking up fishes, some of which were rather slippery. A canvas glove as an aid for holding slippery fish was used on-one hand when necessary. Some of the fishes affected appeared lifeless, but when touched were found to be very much alive and quickly moved away unless caught in the dipnet. Those fishes too small to pick up with the fingers were, with a little practice, lifted from the bottom by. inducing upward currents through | rapid movement of the hands or feet, A fish, thus suspended’ for a few moments, was scooped up in the dipnet. Desirable fishes frequently swam into the crevices of the corals ‘and - erected their. spines, making their removal difficult. With clear vision through the: face mask, these’, AOS were collected. ‘A rebber boat, tied to one of: the coral neads, peat as a base from which to work and was an added safety in case someone ran into trouble under the > rugged conditions. This boat held the preserving tank, and other. gear. Three good swimmers picked, up enough fish to ReaD one man busy preserving the specimens in the rubber boat. . hetSai Those fishes first to be affected by the rotenone were the damsel, eal butterfly, surgeon, and puffers; others such as needlefish,. -halfbeaks,. -goat— fishes, gobies, jacks, threadfins, and mullets were a little slower in reacting _to the treated water. The burrowing fishes, namely, eels, appeared last, probably because it took longer for the rotenone to diffuse into their habitats. Fish continued to appear for over 6 hours after treatment ; eels were recovered that came out 8 hours after the cloud had passed their burrow. ie ' Care was exercised in picking up eunpeccaie dead spiny fishes, and especially moray eels, since they may inflict serious wounds. Scorpion fishes, . Siganids, and other venomous sehuieihed even the stinging corals arise oc fishes, were treated with respect. —32= | The snake eels often appeared with about 6 to 12 inches of their head section above the bottom. They were grabbed firmly and quickly, then the remain- ing 2 or 3 feet of their bodies were pulled out. A light touch or a miss when grabbed usually caused the eel to withdraw into its Pee oe! and the specimen was lost. 4 The rotenone appeared to affect the fishes by constricting the capillaries of the gilis, depriving them of an adequate oxygen supply. They leave their hiding. places., for. more oxygen,.,thus ,exposing themselves ees a ee condition and simplifying vhede, capture, ae aps Uae OR ee Shallow tidal pools that trap fish at low tide are simple to work, but the © use of rotenone in the ocean surf on the ocean. reef of.a coral atoll requires special technique. Lithothamnium Ridge: -- The outer margin of an atoll rim on the ocean side usually consists of the slightly raised pink to red colored lithothamnium ridge contrasting beautifully with the deep blue ocean beyond. It is dissected by rugged surge-channels, and deep pools often 20 feet deep directly connected with the ocean. This, Yidge, creviced and pitted with holies, is about a foot or - two higher than the flat part of the reef farther inward. At extremely low. _ tides it is exposed ‘except as the surf crashes over it, then some of the water is forced back over the flat part of the reef, flowing seaward sea through the surge-channels. cn Some of the surge-channels, extending for a hundred feet or more back into the solid reef, are more or less roofed over or with perforations large and small through which the water may pour or spurt on the incoming surge of a wave. They are lined with rich green and red algae, blue, red, yellow and green corals, and a host of brilliantly colored fishes live in these clear waters. Rotenone was used successfully along the lithothamnium ridge in the ocean surf. . The "mud" was administered a few minutes after the low point of ‘the tidal ‘eycle.. An area was selected where, pools occurred but which were not completely connected” with the surge-channels.. These pools were desirable as settling basins for the sick and dying fishes. The area between two or three surge- channels, where ‘the ‘1 waves ‘flow inward across the ridge was the place where we placed ' ‘the rotefioné mixture. Big handfuls of the thick mud were thrown out as far as possible into the backwash of a wave. The next moment the oncoming breaker churned the water into, , foam and carried the water-laden cloud of rotenone, inward, spreading it over the area and into the numerous crevices, then “ait flowed out the surge-channels.. Soon the rotenone cloud was distributed. along the ocean edge of the reef, and some was, brought back again over the lithdhamnium ridge. The continual surging inward of the water brought in the sick fish. Men were stationed along the surge-channels to take fishes that were: being ‘swept out to sea and perhaps lost. After the. pools: and channels ae a the ir ie ‘were searched for’ fishes by the skin divers. Deep-water Use of Rotenone: -- Two ichthyologists who were excellent swimmers and skin divers, successfully carried on several deep-water poisonings _of fishes with the powdered root. They mixed in the usual manner about 35 pounds of the substance, then placed 5 to 10 pounds of the "mud" in desert .. water bags. Equipped with standard United States Navy shallow water 1. Tc “G3 < diving outfits they took the rotenone to the bottom, distributing it around coral growths. Down below with the usual dipnets, they recovered fishes, bringing them to a man at the surface, who preserved the specimens. This deep- water work was necessary to obtain a more complete picture of the fish fauna of Bikini and the change in kinds of fishes at various depths in the lagoon. Several fish species occurring over the shallower parts of the reefs normally are not found at depths below 10 or 20 feet, whereas some kinds below that depth are not taken near the surface. Collecting With a Light at Night A bright light suspended from a small ship at night at the surface of eH: sea attracts to it’ myriads of nocturnal organisms -- crustaceans, worms, squid,’ octopi, and numerous species of fishes. Silversides, small wrasse, round 4 herring, the pelagic stages of goatfishes, surgeon fishes, puffers, lizard and file fishes dart in and out of the field of illumination. Large flying fishes, | a foot or two long, come swimming or flying toward the light at night. Down below a few feet, larger predaceous fishes can be seen rushing about feeding on the abundant animal life. Eager collectors gathered above this light on a plat- form, with fine-meshed dipnets scooped up the animals, and preserved them for future study. PRESERVATION OF FISHES In general; the following rule should be applied: Preserve all fish that come into the net or are taken by other means. Do not throw away small specimens because there are many species of fish of which the adults do not reach an inch in total length. Large numbers of specimens of each kind are very desirable. One hundred specimens of a single species are often not too many and in some cases are not enough. a Formalin Bue ceraesalorin,: --- Formalin preservation is recommended instead’ of alcoholic preservation. The fish should be dropped alive (if possible) into a solution of formalin made up by mixing one part of commercial formalin with nine parts of water /fresh or salt- ~f4./This solution is ‘of sufficient strength to p preserve small fish up to five inches in length, in about three days, but larger specimens should be left in it for a greater length of time, depending upon their size. Fish which are allowed to die before being preserved are paled and distorted and hence are of less value. All specimens over three inches in length should be slit in the belly with a sharp knife or scissors or they should _ be injected with formalin. This allows the preservative to enter the body cavity and keep the contents from spoiling. In addition to this, very large fish, a foot or more in length, should be injected about every two inches into the muscle tissue and left in formalin from five to seven days, or more. After that time, if it is desired, they can be transferred to water for one or two days, and the formalin washed out, and then placed in seventy-five percent alcohol. One should never crowd the fish in the containers like sardines in a can. If it is desired to leave the specimeris in formalin indefinitely, they may be transferred to a weaker solution, made up as follows: One part formalin to fifteen or eighteen parts of water, to which has been added two teaspoonsful of -%,- borax to each gallon of preservative. This weaker formalin solution is usually of sufficient strength to preserve the fish indefinitely if the container is closed tightly. Always fill the containers full of liquid. Alcoholic Preservation: -- .Drop the. fish alive (if possible) into thirty- five percent alcohol and in about, Six hours they should be placed in seventy-five percent | alcohol. If the specimens are at all crowded, the alcohol should be poured off and fresh seventy-five percent alcohol added the next day. If they become soft, then another change of alcohol should be made, using seventy-five percent again. In general, formalin preservation is best at the start and should be used instead of alcoholic preservation because formalin hardens the specimens. However, after the fish have been in formalin about a week, they should be trans- ferred to seventy-five percent alcohol, after thoroughly washing the formalin out, bécause the acid in the formalin.has a tendency to soften the bones, unless pe 7 neutralized (as by adding borax - see above). Salt Preservation: -- If neither formalin nor alcohol is available, fishes may be preserved in salt. The fish should first be soaked in a saturated brine solution and when thoroughly impregnated, they should then be packed in dry salt for shipment. As with the other methods of preservation, the abdominal cavity should be opened to allow the salt solution to enter freely. It may. be necessary to open the intestinal tract of those fishes that feed heavily on vegetation and remove the vegetable matter accumulated therein. Skinning Large Fish: -- Fish too large for preservation in available con- tainers should not be thrown away but should be skinned as one would skin a rabbit. Make a slit along the abdomen and remove the skin and flesh from the bedy, but leave all of the fins in place and the head attached to the skin. This skin can then be placed in formalin or alcohol, or it may be salted. It is best to remove most of the fat and ail flesh from the skin. For moderately ee fish, it might be well to leave the vertebral column intact. Labels: -- Labels, giving all essential data, should be placed in the jar with the fish when collected. Accurate information about the locality is as valuable as the fish, for specimens without proper data are of little scientific value. These labels should have the following data: Exact locality, with reference to a town or island commonly appearing on maps; date; collector; and any other information that seems pertinent, such as depth of vatoee eS of capture, ecological data, etc. The labels should be written with a soft lead pencil on a special type of paper furnished by the U. S. National Museum, or on any pure linen ledger paper. Do not use ordinary paper, because it will disintegrate in the liquid. Do not use ink or indelible pencil, as these wash off the label. Large fish may have tags tied onto them, preferably through the lower jaw, with all essential data written on the tag, or a number may be used and the data recorded under the identical number in a notebook. Wrapping Fish for Shipment: -- After thorough preservation, fish may be wrapped for shipment in the following manner: Place the small fish in a stack (as cordwood is piled), with their heads outward, so that the tails are -95- protected, and then wrap them in cloth, with the ends secured firmly, tied up with string, or sewed. .Be sure to protect all the fins. Tie each package firmly but. not.so tightly that the. strings will cut into the fish. All con- tainers should. be completely filled with packages of fish, or the excess. space filled-with excelsior or. dry grass. Do not use paper as it softens and . vs | dissolves in the liquid and does not fill the spaces, After the container is y completely filled, then nearly all of the excess liquid may be poured off, | leaving the contents of the container wet. Be sure the container is sealed to prevent evaporation. if a metal can is used, the top should be soldered on. Shipment may be made by mail or express or other means. If the material is to be sent "Collect", please let us know prior to shipment, so that arrangements can ng. pee to take care of phe transportation charges. wy ee -< ‘hy | Borin preservation, fill all containers completely full of ‘preservative-so that there is a minimum of air-svace in the container. If the:-fish are allowed to snake around in the jar, their fins will be aS out and the rays and scales. will become worn off, thus greatly erties the ‘. value of the specimens. 2. Never overcrowd the fish in the containers, because ovei crowding causes the fish to be hardened in distorted shapes and also they are very likely. to spoil for lack of enough preservative (this is particularly true of Be baredied fish such as Suckers and of fish collected in the tropics). ope All fish over three inches in length, especially. soft-bodied fish, should have a small slit made.in the belly. The slit should penetrate into the abdominal cavity. It is best to inject all fish over one foot in ia eae oe the body cavity with preservative. ag is: a rule, eS should be left in the mihlale nt Ah for at least’ one week, depending on their size, before being wrapped for shipment. Dd. Always.place a label, with the essential data, in each jar or package of fish. Be certain that youl: have linen ledger paper for labels as this does not go to pieces in liquid préservatives. Do not use a label covered with starch, as this comes off in water. i ee Gh .. Lome \ COLLECTING SEROLOGICAL SAMPLES : pe Alan Boyden’ oe I. Introduction - The Serological Museum is an agency for developing the subject of comparative serology on a world-wide basis. It is dedicated to a simple principle, viz., that. thé proteins of the bodies of organisms are’as representative of them as any of p their othe other _c r constituent t parts, and are fully as_ worth of collection, preservation and compa comparison as their skins and skeletons, The data of comparative serology have significance for the study of the biochemical evolution of Ls hayp g for systematics, and for medicine. Suitable methods for the serological beac of proteins and other antigens are now available, The great and continuing need is for a truly representative collection of the blood sera of animals, supplemented by other tissue proteins as they become available. The following suggestions in regard to collecting procedures may make it possible to save the sera of animals being collected, or to collect sera especially when the opportunities make it possible, é II.’ Directions ms —— to ee ob rne procedures, The details in ebpara to collecting animal bloods will vary with the kinds of animals collected and with the equipment and experience of the collectors. ‘Accordingly there are described the procedures which could be used: (1) where the equipment is minimum (2). where the equipment permits the simplest procedures for collecting fluid:sera (3): where the standard serological laboratory equipment is Bee Sek In each case use the best procedure possible. Or ae Pe A. daimecedne without any: special equipment for . obtaining’ and ‘handling fluid sera,. Filter paper-=standard. grades: nowt hardened——may be ised” ‘to soak up the blood of animals .collected in the field. These’ soaked papers must be kept from contact. with all other samples ae) adn times. They should be dried in air. (but protected from sunlight or excessive heat and from visitation by insects). then packed in wax or other protective paper.and kept dry. The identification ane other: data may be written on the filter, paper in pencil. it soi filter paper is available ab the moment—~any absorbent paper will do--towelling paper oo on aie or even a piece of clean aber rpersb,, croc gail, aMenritcteies at nial a iedel mammals , reptiles, amphibia, etce - ° a, Birds and mammals © '.. “Small birds and mammals | being skinned. Soak up the blood or serum as it is liberated from the surface of the carcass. ' Then open the thorax, cut the heart and soak ‘up the blood from it on the filter paper, For larger animals (about 5 lbs. body. weight. or larger) the blood may be obtained in larger quantity from wounds as well as from the carcass and heart chamber, os Ce ia Director, The Serological Museum, Rutgers University. =e b. Reptiles and Amphibia Blood may be obtained from these by cutting into the heart chamber, as for small birds and mammals. If the specimens are to be kept with as little damage as possible, a small incision igto the heart will release some of the blood. 2.. Aquatic animals ae Sash, au . f Live fish will bleed from the haemal je if the ae are cut off, This blood can be drained onto filter paper as before. Dead specimens will supply blood or serum from the heart if it is exposed and opened. b. Crustaceae; Crabs and Lobsters Remove the joined appendages. Blood will drain from the © bases of these and can be soaked up on the filter paper. © In the case of lobsters, crayfishes or shrimp, an incision in the mid ventral membranes at the junction of cephalothorax and abdomen will yield a good supply. ¢. Mollusca; Snails and Clams, etc. : The larger snails can be bled by bringing them back to camp and propping them up off the ground with the aperture down. When the animal protrudes from the shell, jab the foot with a knife, making repeated large cuts. The blood may be ex- pelled rapidly. For the clams, the shells should be opened and the water in the mantle cavity released. Then cut the foot and allow to drain onto the filter paper. B. Collecting with containers—-(glass or metal jars or even paper cups) so that fluid samples may be obtained. The procedures here differ from the procedures described under "A" in that fluid blood is obtained from live animals mostly, and collected in clean containers. Such blood should be kept in a cool place-—but never frozen--until clotting has occurred. From this clotted blood, the serum will be expressed, which process may continue for severad hours or over night. The watery serum may be separated from the clot in one of several ways: Pour off into another container, getting as clear serum as possible. If this serum is approximately 5 ml. or more it may be preserved with formalin (see remarks on preservation). If less than this amount, it may be soaked up on filter paper and dried as described in A. If the amount of blood aad serum is quite small, the serum may be soaked up on the filter paper directly from the container holding the clot. The expression of serum from the clot may be hastened by loosening the clotted blood from the sides of the containers or cutting the clot into sectors. In case of Crustacean bloods which jell but do not express serum, the jell should be squeezed with cloth or even with the filter paper if no cloth is available. The serum can thus be expelled from the jell under pressure, ’ eee / efe "gs" ~98— Collecting. with availability of standard serological equipment: syringes and needles, centrifuge and centrifuge tubes, serum bottles, refrigerator, etce . Fluid blood may be obtained by collecting from wounds, or by heart puncture with syringe.and needle. Blood is allowed to clot and to stand ina cool place for several hours or overnight. Centrifuge and pour off serum. Recentrifuge and collect the _ clear serum. .Add aie abe and store in serum bottles. III. General eens in regard to sizes of samples, preservation, etc. kee Filter paper samples From small specimens only very small samples will be obtainable. Thus for a mouse, or sparrow-sized bird--one piece of filter paper, 2" x 4" will give plenty of room for the writing of the name on the paper and for the blood. For larger specimens up to the equivalent of a square foot of soaked filter paper is desired. Fluid samples. a. Size These may vary in size from 5 cc. up to 100 cc. or more. b. Preservative The standard field preservative is formalin. Ordinary commercial formalin (35-40% content of formaldehyde) can be used. Prepare a 2 percent solution of formalin by taking 2 cc. of formalin and adding to 98 cc. of fresh or distilled water. Add this 2 percent formalin to the serum in the proportion of one part of 2 percent formalin to nine parts of serun, The volumes may be approximated if no measuring apparatus is available, co, Pooling of samples The fluid blood of several small animals may be pooled together and treated as one sample. This should be done however only where the specimens clearly belong to the same species (or subspecies if known). d. Records The record for each sample should include the following information: -— Scientific Name: Date of Collection: Collector and Institution: Place of Collection: Sample from one specimen or pooled: Treatment of Sample--especially preservative used and amount: Ce f. ~99— Shipment ’ Filter paper samples may be sent first class as ordinary letter mail. Fluid samples preserved with formalin should be carefully packed and shipped by air mail or surface mail to: Serological Museum Rutgers University New Brunswick, New Jersey U.S. Ae Suitable filter paper, glassware, shipping containers, etc. will be supplied where needed. Acknowledgment The samples contributed to the Serological Museum will be made available to competent serologists for studies in comparative serology. When scientific reports are published, full acknowledgment will be made to the collector and his institution for the aid given. Chapter 8 -- Marine Ecology INSTRUCTIONS FOR MARINE ECOLOGICAL WORK ON CORAL ATOLLS by Robert W. Hiatt: INTRODUCTION These instructions are designed for studies of only a few months! duration and to cover the most important ecological features in a manner which permits comparison of atoll communities. The studies outlined may be made with a minimum of equipment and personnel, for only on rare occasions would an ecologist have an opportunity to take all the equipment and have all the assist- ance required to do a more intensive job. These instructions are, therefore, pointed toward meeting the problem of working in the field in the most practical manner possible with a minimum of equipment and assistance. Methods of collecting on coral reefs and of vreserving animal species are discussed in separate sections of this handbook. CHECK-LIST OF ESSENTIAL EQUIPMENT Clothing Canvas gloves (1 pair for each two weeks in the field) Canvas wading shoes (1 pair for each three weeks in the field) Swim trunks (eee pairs) T-shirts or other shirts (to be worn both in and out of the water) Hat with good sun visor Heavy shoes for hiking Other suitable clothes for travel and casual wear Drugs and first aid material ay First aid kit (include ready-made bandages, gauze, adhesive tape, merthiolate, sulpha salve, and antibiotics) Aspirin Chlorine tablets for purifying drinking water Sun-burn preventative Collecting equipment Inner tube and canvas float Face masks (at least two) Swim fins Geology pick or similar instrument Pinch bar, smail Swimming bag (canvas body, fish net Mees and heavy zippered top opening) Canvas buckets Dip nets (4 inch stretch mesh; 1 foot in diameter) Shovel, short-handled Cord, eon of small diameter cord or rope Hunting knife and scabard Kerosene mantle-type lanterns (take Spare generators and alcohol for starting them) -100- -101- Collecting equi > (cont'd) Flashlight << of spare batteries and spare bulb) Glass-bottom box (not essential if. proficient with a face mask) Shielded thermometers (three) Portable tide guage (very useful, especially where tidal data ° are lacking) Small framed mirror etched with carborundum Rotenone, freshly ground cubé root (Fish-tox, a wettable preparation from Standard Supply Distributors, Wenatchee , Washington, is preferable Hand leveling instrument Metal rule, six foot Preserving materials and equipment | Metal gallon and two gallon cans of demedednaetlLe material (make use of empty food cans temporarily in the field) Formalin, five to ten gallons, preferably in small containers Alcohol and glycerine (5%), two or three gallons Magnesium sulphate, ten pounds Chloretone, small bottle Cheesecloth for wrapping specimens, two bolts String for tying labels and specimens, one skein Parchment paper for labels Heavy manilla labels, 1" x 2" Stainless steel leader wire for tying labels to coral specimens Pliers, with cutter blades Vials, homeopathic, one gross each of l, 2 aha 8 dram Corks for viais’ India ink and pens Bakelite top bottles, 4 dram (for salinity senile bo Hypodermic syringe, 50 cc., and needle, 18 ga. Field laboratory equipment Reference books on taxonomy (see annotated list) Long forceps Dissecting instruments Dividers Foot rule, metric Glass slides and coverslips Syracuse watch glasses Stacking dishes Data books Blank cards, several hundred Pencils INITIAL SURVEY OF REEFS Land area on atolls is small but reef area is extensive and variable in composition. Since growth characteristics of coral species are closely linked to environmental conditions, especially as regards movement of water, it is to be assumed that a great variety of kinds of coral reefs, sand and muck flats, and gravel or boulder shores will be present. Naturally the complete ecology -102- of an atoll would not he understood without analyzing the interaction in and between each marine biotope. However, the time, distance from research centers, and limited personnel involved preclude such a complet e study. The investigator must, therefore, size up the entire reef and shore area so that intelligent selection can be made of working areas. Listed below on a pricrity basis are a few things to consider as criteria for selecting a special section of the reef or shore for investigation. 1. The area should be typical of a large proportion of the reefs and shores of the atoli. This means that minimally a section or sections would have to be selected from both the sea and lagoon reefs and across the island from shore to shore. 2. kxtremely wide reefs should be avoided unless they are distinctly typical. Usually the same zonation may be found in narrower reefs, thus facilitating the collection of data in a briefer period of time. 3. The areas selected should be relatively free from heavy waves during periods when collections and observations are to be made. 4. The study areas should be as close to the base camp as practi- ‘cable. Usually this is not entirely passible because of the differences in the sea and lagoon reefs. If the two study areas must be some distance apart, the base camp should be nearest the. study area on the sea reef, because transportation along the sea reef about an atoll is often impossible. whereas native canoe transportation in the paeocl is ae THE TRANSECT METHOD. ' Its Value The ecological transect seems to provide the best working procedure for the study of coral reef areas. Not only:is it possibie by such concentration in area to secure a better qualitative picture but it is the only method by which a quantitative analysis may be made. The latter point is significant in that most invertebrates closely inhabit certain coral species and these coral Species are zoned across the reef flat and down the sloping reef.edges. -More- over, the fishes are tied ecologically to particular facies in the reef biotope and such relationships are best illustrated by treatment of data.‘ gathered from particular quadrats within the transect. . Any other means of collecting data wili fall short of the critical analyses wee possible by the transect method. It is desirable, once the coilections and observations in the transects have been made, to make general observations on other reefs and shores for purposes of general comparisons with the more detailed data. In this way a better under- standing of similarities and Gotten ences from one region to another will be secured, i ran he Procedure in Laying Out Transects From a point at the upper edge of the beach where the terrestrial flora meets the sand or rocks, lay out a line down the beach and across the reef to its outer edge. Place markers at equal intervals along this line. These may be at 50, 100 or 200 foot intervals, but keep them equal, Use sticks well propped up by rocks, or use piles of rocks for markers. Now move horizontally from this line 50 or 100 feet and place similar markers along a similar line. This will segregate the beach and reef area into square or rectangular quadrats) of equal area. Give each quadrat a code number in your notes. All collections and observations should be made within these quadrats and the specimens should be labeled as to the quadrat in which they were taken. Collecting Data Physical and chemical data a ae ES HE Temperature readings and samples of water for salinity determinations: must be taken in each quadrat during both high and low water, and during morning and afternoon periods in order to measure the limits of variation. If especially low tides occur, every effort should be made to get temperature and salinity data. Tidepools in the intertidal zone should be visited for such eoljeteteus approximately hourly on especially warm cays. TEMPERATURE readings should be taken in each quadrat with a Sipelded tears meter held at the upper level of the coral heads. During low tides this» may be done by wading, but often at high tide the readings must be taken while swimming, In the latter case the best method is to wear a face mask, lower yourself to the coral and hoid there, with one hand grasping the coral and the other holding the thermometer in place. After a half to one minute read the thermometer in place. For recording in the water one may use a small framed mirror, the face of which has been etched with carborundum so that a lead pencil will mark it. Use an ordinary pencil wrapped with electricians rubber tape. Tie the mirror and pencil onto the inner tube float which should always be tied to the observer with a short rope. The code numbers for the quadrats can be written on the mirror before entering the water, then only the thermometer reading need be recorded while swimming or wading. For the quadrat which includes the outer edge of the reef and the descending reef slope, a surface reading should be taken with the thermometer held about two feet below the surface. A deeper reading down the slope should be taken because it will represent the closest to oceanic or mid-lagoon conditions that the investigator will be able to get on the transect. The simplest way is to skin-dive down as far as you can, hold onto a rock or coral head with one hand and hold the thermometer in the other. Read the temperature after a half to one minute at that level. With a little practice one should be able to lower himself fifteen to twenty feet for this purpose. SALINITY samples should be collected concurrently with the temperature records. The volume of sea water collected depends upon the method used to determine salinity. Experience has shown that the Wheatstone Bridge and similar electro-chemical indicators are not suitable for work in the field under con- ditions met on atolls, since the high humidity soon shorts out the mechanism. | | -lO4= However, it is often désirable to know something of the characteristics of the water while in the field. The only suitable method for making these determina- tions is the silver nitrate titration method, so take along sufficient 0.5 Molar Ag NO., 5% potassium chromate, and the necessary glassware to make a few determinations in the field. Use 25 or 50 cc. medicine bottles with bakelite screw tops for the water samples. Place labels containing the date, tide and quadrat code number in the bottle before entering the water. These bottles may be packed in small compact containers and returned to your institutional iheboratery for titration later. OXYGEN and HYDROGEN ION concentrations need not be recorded in ecological studies on coral reefs, since they are without significant variation. PHYSIOGRAPHIC feaeures such as levels of reef rock and tops of coral growths in certain instances are significant in relating coral zonation to environmental factors. These data may be determined by establishing an arbitrary bench mark, if an official one is not present, and mapping the con- tours with a hand-leveling instrument and a jury-rigged surveyor's staff. In certain localities tidal records will be unavailable, and in other localities such records as are available are grossly inaccurate. In order to establish the tidal information desired, and incidentally to refer the bench mark used in ascertaining the reef contours to tidal data, it is desirable to construct and take along a portable tide guagé. The type described by Wentworth (Jour. Washington Acad. Sci., 1936, 26(9): 347-352) has proved simple to construct, easy to operate and sufficiently accurate. Biological data FISH are DT ioaiee adequately only by the use of rotenone or Fish-tox.,. (See instructions for collecting fish. ) Rotenone may be obtained in . ground, powdered form in a concentration of about 5% at almost any firm, manu- :facturing insecticides. If the powdered form is to be used. it must be freshly _ground and used in a few weeks. Fish-tox is more satisfactory for longer periods in the field. Two persons can poison an area of avout 7,500 to 10,000 . square feet and do a fair job of picking up the fish. A smaller area is prefer- able. Do not poison the area if the tide is too low, as most of the fish found in the quadrat when the tide is higher will have gone over the edge of the reef. It is best to poison about mid-tide as it is receding. This will leave adequate depth of water and the force of the waves and the surge will be subsiding rather than sensing ; factors of great imvortance in alah up the dead and dying fish.. The fish should be collected quadrat by quadrat, taken back to the field laboratory and processed, and those not preserved or used otherwise should be returned to the water where they will be quickly devoured. An important point. to have in mind is that the natives will be greatly interested in this pheno- menal method of collecting fish. Great care should be taken not to poison fish indiscriminately or too near human habitation, because native subsistence depends to a great extent on the reef fishes and you must maintain happy public relations with them. Always give the natives edible fish which you do not wish to keep for later study. Local native customs will Seas how the distribution of fish will be a. . py 1 -105- Invertebrates should be collected or noted quadrat by quadrat also. Instructions for collecting in particular types of marine environments and for special groups of animals will be found in another section.of this handbook. ANALYSIS OF DATA Physical and chemical data All physical and chemical data should be analyzed and presented in the usual acceptable manner, keeping in mind that comparisons and correlations will provide the most useful data for interpretation of biological phenomena. Raw data should always be tabulated and included in an appendix to any report. | Graphic means of presentation are usually best for summary analyses in the main body of the report. Fauna and flora Two considerations should be prominent in handling these data. First, the animals present should be identified with accuracy, preferably checked at least by leading taxonomic specialists. Second, their ecological relationships should be described. The first consideration does not require elaboration, except to say that specimens should be preserved for identification in the best manner possible under field conditions. Collections made as quantitatively. as. possible in quadrats described in a transect naturally lend themselves to analysis in a number of ways. Quantitative and qualitative patterns of distribution by phylogenetic groups and by species may be made’in relation to physiographic features, physical and.chemical aspects) of the environment and special biocoenoses within the general coral reef BPO Zonation horizontaily and vertically is readily designated. Biological phenomena on lagoon reefs may be compared directly with sea reefs, and various exposures of reefs on which such transects are made may be readily compared for a better understanding of factors underlying distribution and abundance of reef | species. Useful Field References Generally speaking, positive identification in the field is difficult and should not be a prime objective of the investigator. However, it is desirable, where possible, to make tentative determinations. Unfortunately, the Pacific region is singulariy lacking in faunal and floral handbooks, so recourse must be made to original taxonomic papers. The following list is highly selected for fieid use and omits monographic works or less extensive works which would not prove practicable to carry into or use in the field. Fish Hiyama, Y. 1943. Poisonous fishes of the South Seas. (In Japanese). 136 pp. Illustrations are excellent and valuable in dealing with fishermen. A translation is available, but it is not needed. — =106- Schultz, L. P. 1943. Fishes of the Phoenix and Samoan Islands collected in 1939 during the expedition of the U.S.S. "Bushnell", Smithsonian Institution Bull. 180: 316 pp. “4 Currently a basic work for any taxonomic atde. Has good keys which incorporate descriptions of the fishes. Does not cover certain large groups such as the scarids, iutjanids, and serranids, and other literature should be used for them. Smith, J.L.B. 1949. The sea fishes of Southern Africa. Central News Agency, South Africa, 550 pp. Beautifully iilustrated and contains a key to all species listed. While it covers many fishes not found in the Central Pacific, it is nevertheless invaluable to an investigator not well-versed in Central Facific forms. Tinker, S. W. 1944. Hawaiian fishes. Tongg, Honolulu, 404 pp. Illustrations make this useful in dealing with fishermen, Umali, A. F. 1950. Key to the families of common commercial fishes in the Philippines. Research Rerort 21, U. S. Fish and Wildlife service, U. 5. Dept. of Interior, 47 pp. This key to the families should prove very useful to an investiga- tor not familiar with Central Pacific forms. Weber, M. W. C. and L. F. de Beaufort. 1911-1940. The fishes of the Indo-Australian Archipelago. Vol. I- IX, &. J. Brill, Leiden. This series of nine volumes serves to supplement Schultz in regions west of Samoa. I1t is considered basic, although its keys and descriptions are not always. adequate. Invertebrates Echinodermata Clark, H. L. 1914. Hawaiian and other Pacific Echini. Mus. Comp. Zool., Mem. 46(1): 1-78. SO 1917. Ibid., L6(2): 79-283. ‘The most concise account of Pacific Echini. Mortensen's monograph is recent and more complete, but is hardly a field reference, | Clark, A. H. 1949. Ophiuroidea of the Hawaiian Islands. B. P. Bishop Mass Wades) LOS.) 33) ps...) Includes many widespread Indo-Pacific species. “107 Fisher, W. K. 1907. The holothurians of the Hawaiian Islands. U. S. Nat. Mus., Proc. 32 (1555): 637-748. Includes many widespread Indo-Pacific species (except apodous holothurians). Fisher, W. K. 1919. Starfishes of the Philippine seas and adjacent waters. U.S. Nat. Mus., Bull. 100(3): 712 pp. The best general account of widespread Indo-Pacific species. Mollusca Hatai, K. 1941. Recent marine shell bearing Mollusca of the South Seas islands. Inst. Geol. and Paleon., Tohuku Imp. Univ., Sendai, Japan, One vol. and atlas of plates. Fine illustrations for field identification but quite incomplete. Other volumes are to follow. Hirase, S. 1936. A collection of Japanese shells. 5th ed. Matsumura Sanshodo, Tokyo, 217 pp. An excellent treatise with photos in color of many common gastropods and pelecypods of the Indo-Pacific area. The nomenclature should be checked with specialists. Robson, G. C. 1929. A monograph of the recent Cephalopoda. Part I. Octopodinae. British Museum, London. 1932. Ibid., Part II. Octopoda. (except Octopodinae). Op. ert. This is the best treatment of the Octopoda. Unfortunately there is no similar treatise available for the Decapoda. Arthropoda Kemp, Stanley 1913. An account of the Crustacea Stomatopoda of the Indo-Pacific region. Indian Mus., Mem. 4: 217 pp. Most useful taxonomic account of Indo-Pacific stomatopods. Miyake, S. 1943. Studies on the crab-shaped Anomura of Nippon and adjacent waters. Dept. Agric., Kyushu Imp. Univ., Jour. 7(3): 49-158. Excellent account of many of the porcellanid crabs of the Indo-Pacific. Pilsbury, H. A. 1916. The sessile barnacles. U. S. Nat. Mus., Bull. 93: 1-366. | Sakai, Fune. 1936. Studies on the crabs of Japan. I. Dromiacea. Zool. Inst., Tokyo Univ. Lit. Sci., Sci. Rep. (B) 3(supp. 1): 66 pp. -108— 1937. Ibid., II. Oxystomata. Op. cit. (suppl. 2): 1938. Ibid., III. Brachygnatha, Oxyrhyncha. Yokendo, Ltd., Tokyo, 193-36h pp. 1939. Ibid., IV. Brachygnatha, Brachyrhyncha. Op. cit., 365-741 pp. : These volumes are the most useful in identifying crabs in the Central Pacific area. eae Chapter 9 -- Anthropology ANTHROPOLOGY AND CORAL on FIELD RESHARCH by Alexander igooehi I. Introduction. The examination in the field of the relationship between man and his atoll environment can be focussed on the man (including his culture) term, or on the environment term in this relationship. In either case, it is understood that the relationship between the two terms is the important thing, but as a matter of field procedure it may be necessary first to concen- trate on the terms themselves and then investigate the nature of their relation- ship. In the organization of the work of an ecological field team the anthropologist will necessarily play a. primary role when the nature of the cultural adaptation of man to his‘environment is under consideration. When the focus of interest shifts to the ways in-which man's presence has affected the environment, the anthropologist will play a supporting though nevertheless essential role in assisting his colleagues in understanding the factors involved in man's behavior in altering his environment. In the ecological approach to the. aaEHropetody of coval! + Stolle, it isa reflection of the immaturity of our science that we do not ‘yet know those ‘enecific questions to ask whose investigation will bring us’ greatest returns. One phase of atoll research necessarily revolves around formulation of future problems for empirical investigation. This is true of any vigorous discipline, of course, but in human roe we are still not very far from our starting point. } However , we can aah eat’ ne: Bpanoudele within which sidnncnet betel field- work on coral atolls: should proceed.» It is my feeling that such field observa- tion should Ka) Focus | oh the present rather than the aboriginal past at the inception of any period of field observation, (b) As the work progresses, elucidate | ‘processes involved rather than. static patterns, which may mean dredging up as much relevant historical data as possible, (c) Operate with a set of problems suitable for comparative investigation, for in atoll research comparison must be substituted very largely for experiment. II. Problem areas. Rather than attempt to outline a series of questions applicable primarily to a given atoll, it is perhaps preferable to indicate a number of probiem areas within which field research will’ be fruitful. It should be emphasized that field research should in the end center on relationships rather than terms alone -- as on the relationship of technology to the amount and characteristics of natural resources rather than technology alone or resources alone. Problem areas follow: A. Population size and density in relation to food resources, given a particular type of technology. In view of the probable future population pressure on atoll resources _generally, this forms: an,important field of investigation, Ove leads a to.diet and nutrition studies. Sqo98 -110- B. Technology in relation to the amount and character of natural resources available and used. Description of tools, in the familiar context of material culture studies, are not as important in this connection as the determination of the over-all characteristics of the technology under examination--skills in relation to technical processes; time and energy expended in particular technical processes in relation to productivity, the total tool system, etc. There is also a need for re-orienting studies of technology to the modern scene rather than to the aboriginal past. C. Technology, economic organization and utilized resources. Here it may be possible to consider economics under the “conventional headings of production, distribution, exchange, and consumption to determine how the organiza- tion of human activities in these areas is related to the technology, and to the utilized resources. On the other hand, an entirely different approach may be more useful. Fieldwork in economics is very time-consuming, however, a factor that must be taken into account. An important question is the mechanics of social control of atoll resources in relation to the organization of production, distribution, and consumption. D. Social organization in relation to technology, economics, and utilized resources. Some of the most productive field research in American Indian social organization has utilized an ecological approach, such as the work done by Steward in the Great Basin, and by Hallowell and others among Algonkin hunting groups in the northeast. Comparable results shouid be obtained when a similar approach is applied to the study of the social structures of atoll peoples. In this regard, it is important to determine the essential differences in social organiza- tion between populations on atolis differing in the abundance and variety of their resources. Asa preliminary hypothesis, it is suggested that stratifica- tion, elaboration of the functions of corporate kinship groups and complexity of local organization will be more highly developed on atoils with abundant resources than on those with scarce resources. Hypotheses of this sort need to be formulated and tested. III. Planning. A fundamental problem in the planning of atoll field research is the need for selection of the kinds of data to be collected in order that a relatively brief period of fieldwork be of maximum value. It is suggested that the anthropologist who is able to spend only several months on an atoll concentrate his work on a few problems rather than to diffuse his effort more widely. The type of problem envisaged by an ecological approach demands the examination of functional relationships between variable factors; This type of problem requires more than superficial observation and there is a real danger that the fieldworker may diffuse his effort to the detriment of his over-all results. There is a distinct need for concentration of effort on a limited number of objectives. -lll- SUGGESTIONS FOR INVESTIGATING THE CULTURE OF ATOLL FEOFLES by Leonard Mason The following suggestions for investigating the culture of an atoll people are based upon work already accomplished at Arno Atoll in the southern Marshalls. They are outlined here as a guide to the kinds of information which have already been collected for one atoll group. It is hoped that in other atoll studies the same kind of information may be collected by other workers in order to provide some common basis for comparison of atoll cultures. It is recognized, of course, that some subject matter suggested here may not be emphasized at all in other cultural situations, and that some subject matter not included here may be important to concentrate upon elsewhere. I have found it convenient to group my suggestions in the following organization because investigation of man-land relationships focusses the investigator's attention upon economic activities in which material culture, social culture, and land tenure are frequently so interrelated that this kind of division of subject matter is a difficult one to follow. I have chosen to group my suggestions, therefore, into five categories, not all of which have equal importance for an ecological study, but all of which relate directly to it: (1) population inventory, (2) economic activities, (3) land tenure, (4). social and political organization, and (5) movement of people. 1: Population inventory. An inventory of the population is essential. How intensive or extensive this inventory will be depends upon the time and assistance available to the field investigator. At Arno, I tried to complete a card file for the entire atoll, a population of under 1,000. This was not feasible in the time available, and the populations of only four communities in the atoll were processed, about 450 people. For each individual we sought and received, usually with very little difficulty, certain personal history data which were recorded during the interview upon a 5 x 8 card form previously prepared for that purpose.* This information included the following: (1) age,(2) sex, (3) residence at. the time of the interview (noted by name and location of the land-holding on which, the residence was located), (4) ethnic origin (whether all-Marshallese or mixed with Gilbertese, Japanese , or Caucasian), (5) birthplace, (6) identity and birthplace of parents (useful information for determining the degree of immigra- tion), (7) children (with their names, ages, sex, and present residence, plus fact of adoption where pertinent), (8) educational background (number of years and type of school, for evaluation of effect of formal education on the man-land relationship), and (9) specific awareness of the world outside the atoll (visits or residences at other atolls and- other Pacific regions, and the reasons and dates of same). A photograph was taken of every individual interviewed for purposes of cross-checking photographic data later in the identification of particular individuals pictured in various economic activities. Additional information was secured in these interviews as to (1) clan affiliation, (2) social rank or status, (3) change of residence with adoption or marriage, and (4) religious membership. Some of this material has a direct bearing on the economic relationship of man to land, but much of it serves rather to aid the anthropologist in further understanding the composition of the community. *"Soa sampie card A, p. 115 -112— Analysis of these data provides helpful clues in tie determination of trends in the changing relationship between populations and natural resources over a period of years. Shifts in population away,from,or onto the atoll in recent generations, as indicated by individual ancestry records, obviously may have some effect on the actual utilization of an atoll's resources. The longer-term trends in population growth*have to be determined with the aid of official records or early estimates by travellers and others, where such ‘written records exist. Factors contributing to changes in the size of popula- tion could often be detected in interviews with older islanders, when informa- tion was given about infant mortality, diseases, foreign medical aid, immigration or emigration for a number of reasons, and disasters, such as typhoons, drouths, and warfare, 2. Economic activities: The investigation of economic activities was organized along two lines: (1) subsistence, which has to do largely with the direct utilization of atoll resources to meet the demands of daily living, such as the acauisition and preparation of food, the construction of housing and canoes, the preparation of fibers and other materials for handicraft used locally, and (2) commercial, which has to do with introduced economic activities, such as copra and handi- craft production for sale to traders, wage labor, and the importation of trade goods to supplement local products. In both cases, studies were made of the kinds and quantities of local resources and imported items that were used, of the techniques and apparatus employed, of the work organization (by individual effort or cooperative labor of a family, a lineage, a household, or a community) , of the distribution and use of the products manufactured or prepared or of the money derived from commercial activities. Special attention was given to the amount of work accomplished, the time expended, the individuals involved and their relationship to each other, and the location of the activity. Following this section is a copy of a project proposal for anthropological investigation of Arnoese economy during the. summer of 1951 by Mr. Harry , Uyehara (graduate. student at the University of Hawaii, under my direction). This project emphasized the collection of certain quantitative data to supplement material gained the previous summer. This project was part of the Coral Atoll Program sponsored by the Pacific Science Board bias the Office of Naval Research. To supplement the notes recorded on economic nieigeet cages an extensive photographic record was obtained at Arno, with pictures of food preparation, handicraft production, etc., in black-and-white and color stills, and in color movies for those activities better described by the cinema medium. Such a photographic record proves to be very worthwhile in checking some field notes. where for reasons of haste or misunderstanding , notes are incomplete or con-_ fused. In the analysis of economic activities, special interest should be directed toward the relative importance of commercial activities through the years, and the trend today toward greater dependence of island populations on participation in the world economy. Special inquiries were made in the field about the historical development of certain commercial enterprises, with notes about significant changes brought about in the.man-land relationship. Estimates were also attempted as to the possible effect of other changes as contemplated by the people or the administration for future years. lig 3. Land tenure: Closely related to the preceding topic is, of course, land tenure and associated matters, and especially so in atoll regions where scarcity of land frequently heightens anxiety about land ownership and use rights. At Arno, land maps drawn to a scale of about 1:5000 were produced for several islands (all islands in the atoll could not be surveyed in the limited time) . For this map- ping, hand—compass for directions, and pacing for distances proved to be accurate enough for our purposes. Generally, it was easiest to lay in the single main street first, then tie in both ocean and lagoon beaches which roughly paralleled the street. Paths and trails from the beaches provided convenient traverse lines to the street. All cultural facilities, such as dwellings, cook-huts, copra driers, latrines, pig-pens, wells, boat—houses, cisterns, public buildings, cemeteries, paths and trails were indicated on the map to scale. (A photographic inventory of these facilities provides a convenient check-list for identification and comparison of types.) Finally, with the aid of community officials and local landowners, the boundaries of each land-holding were traced from ocean to lagoon and tied into the street. In general, where the terrain had been cleared, the actual boundaries were followed (with notes on the type of markers " employed) « In some cases, however, where brushing of the area required too much time and labor, only the general location of the boundaries was indicated. Notes were made on the names of individual land-holdings, of persons associated with the land as owners, and of persons holding only a use right to the land. Inheritance of land parcels was traced back for several generations, in representative cases, and related to genealogies of the lineages concerned, Disputes about boundaries or about title to land-holdings were recorded during the field mapping, to be followed up later by interviews with interested parties to determine more Rneqisely the present methods of dealing with land problems. Information was collected in interviews on the rights and obligations of various categories of persons associated with land-holdings, about changes in attitude regarding these rights and obligations, and about the interrelation- ships of land tenure patterns and other aspects of the economic system. Special attention was devoted to the actual relationship of people to their land in daily practice, the use made of the land, the products derived from the land, and the potential hardship to be suffered if their interests in the land were either reduced or completely voided. Fishing areas, in the Marshalls, are relatively free and available for use by all Marshallese. In atolls elsewhere in the Pacific where fishing areas are regarded as properties to be owned and safeguarded, a more detailed investigation of this subject would obviously be called for. 4. Social and political organization: In the investigation of social and political organization as it pertains especially to the man~land relationship, attention was given to the data under three general headings: (1) kinship groupings, (2) territorial groupings, and (3) systems of rank. With regard to each of these, land ties were found to be a vital element. Ph a For kinship groupings, data were collected on the nuclear family, the lineage, and the clan, as pertinent social units; for territorial groupings, attention was centered on the household, the hamlet, the island community (village), the district, and the atoll. In both cases, notes were taken on the composition of the groupings (together with significant variations permitted in the culture), the organization of the members, leadership, the relation of each grouping to others of both types (kinship ana territorial), and vcsigations of members to each other (especially those involving exchanges or transfers of property of any kind). Trends were investigated as to the respective positions of each type of grouping within the total community, particularly the increasing importance attached to territorial units as compared with kinship units, and the possible effects of such trends upon the man-land relationship. In investigating systems of rank, both social and political, data were sought about the respective importance of such factors as birth, wealth, personality, education, etc. Notes were taken on the actual composition and organization of certain rank hierarchies as they existed within the atoll population, on the function and relative importance of each rank, and on the relationship of each rank to the exploitation and distribution of natural resources. 5. Movement of peopie: Finally, attention was devoted to the movement of people, within the atoll (from’ land-holding to land-holding, and from island to island) and between Arno Atoll and other atolls and islands in the Marshails. Frequency of travel was noted, as were also the purpose, persons involved, type of craft utilized, and various factors affecting time of travel and duration of trips. Through inter=- views and actual observation, daily, seasonal, and annual patterns of movement were established, especially as these related to the exploitation of natural resources. An attempt was made to note significant differences in such exploitation, since contacts between Arno and other atolls have become more frequent in recent years. SAMPLE CARD A, recto ARNO ATOLL CENSUS, SUMMER 1951 No. Name Residence Island Household Sex Age Race Clan Atoll of birth Mother Mother's birthplace __ Mother's clan Father Father's birthplace Father's clan Adopted? By whom? Change of residence? Married? Spouse Clan Date Place Spouse's mother Spouse's father Change of residence at marriage Religion - Education, Where? Occupation Class or social status Atolls visited: (German) (Japanese) (American) vere Children: Age ___Sex__ Residence (Living) Children: (Dead) Spouse deceased? When? Divorced? When? Why? Remarried? Date Place Change of residence? Spouse Clan Residence before mrriage bratel . Arete te Ht . peep ue a : 1) eee eee 3 ai , ee ee 5 yeahs newrsihn gn, i f y . 7 ¢ ‘¢ “Ss 2 - 2 a: . = [= * & at Seoten sli wet o panei Set, metab ts jaro e eeaamaeeniian sragenincintiesinneeyamenacrinerinentt inlaws one 890 € , wate 4 : a es ’ ‘ aceasta nem TgHeRTONE AP on item oe ; > r See ee Jeaigateicnan tli RON ALE IL ELI eh SI te AT TR Oe HRY OR OP EO AM ; 4 i rig Ree TO catia: pine amelie a ER ei ommnaanieel sepccuinbabbalehannaache hee snansiwrneassiigibmeisihnaneiars seth tt aN OTe oR PN Le em sinmanaia asia ihe NA) NTA NT “ ‘ \ Wherein - : th : ‘ ‘ sueoramrpenaa mean one atih a ecm hs aac pore ee ee ae siiaihalimapiaischalnenit NOELLE EI 4 x 4) 4 a fy we : ve . pakenes oe ‘tgp kee teh mes bbe RR Peer ee ere Se — wit 1 cco hd son bbce "dha iinet bs alin en fnheonf = TG A QUANTITATIVE STUDY OF CERTAIN ASPECTS OF THE MAN-LAND RELATIONSHIP IN MARSHALLESE ECONOMY AT ARNO ISLAND ~ by Leonard Mason and Harry Uyehara Quantitative data for the fiscal year 1949-50 were obtained” for atoll production of copra at Arno (by district and by season, sales value and WETEnE and distribution of money income) and for atoll importation of trade goods (by district and by season, sales value and, in the case of food supplies, specific itemization). Further quantitative studies had been intended, at Arno Island only, with special reference to individual and family participation in copra production, in préparation and consumption of food, in movement of persons within the atoll and between atolls, and in the distribution and use of other economic goods derived from co natural See: Time, however, did not permit. A feeling that generalizations about these activities, as based upon personal observation by the investigator or upon statements and estimates by Marshallese informants, were frequentiy tco vague or. unreliable, resulted from two conditions. The first was the great individual variation in work activity as noted in the few test studies on a quantitative basis that were attempted. The second was the unconcern on the part of the majority of Marshallese, in re- calling past behavior, about the accuracy of murbers of hours cxpended, numbers of people involved, and amounts of work produced. These are conditions, of course, which are not peculiar to the Marshallese situation. Other anthropologists working with other groups in the Pacific and in other parts of the world, have noted the same methodological difficulties among nonliterate peoples. Nonetheless, the value of quantitative data which can be obtained under controlled conditions’ wath statistically reliable techniques is never doubted. : : (in this case, Mr. Harry Uyehara) It was proposed,therefore, that an anthropologist /conduct an intensive quantitative study of certain aspects of the man-iand relationship in Marshallese economics at Arno Island during the summer of 1951. This study was an extension of the 1950 field work referred to above, and utilized the materials already collected and analyzed. The results of the 150 and 1951 field projects will eventually be integrated with other studies under taken at Arno Atoli within the framework of the Coral Atoll Program for the better understanding of the Arnoese and their ecological relationship with other components of the total atoll environment. In 1950 Arno Island, with a land area of 0.66 square miles or 422.4 acres, supported 201 persons who were combined in various numbers (ranging from 1 to 15, averaging 6) in 30 households. The island was divided into 63 land parcels, each averaging 6.72 acres and with boundaries from lagoon to ocean shores; only 30 of these parceis were actually occupied in 1950. The remaining 33 were either more or less abandoned although ownership and use rights were kept alive, or associated with parcels then occupied. * by Mason in 1950. ~117- In obtaining quantitative data for statistical treatment, 12 of the 30 households were selected by random sampling, thus 4 from each of three strata based upon size of household. After an inventory of economic resources available to each of these 12 households made by the anthropologist, with — special attention to numbers and kinds. of economic plants, animals, and -fishing areas;, building facilities and household goods, and personal effects, he checked each household daily during one month,and later for one week for a detailed quantitative report on a number of economic activities as listed: below. A number of simple’ questionnaire forms were prepared in English and” in Marshallese on which each household group recorded numbers, amounts, and ~ times involved in production, distribution, and consumption of economic goods .# The anthropologist lived‘ at intervals of one to two weeks in several‘of these 12 households, reviewing daily the reports of the other 11, checking for accuracy wherever possible, filling in more complete information through inter- view with members of each household, and further instructing Marshaliese - assistants in a more reliable reporting of economic activities. The difficultie of employing questionnaires, among an essentially untutored people are realized, but it is believed that the carefully conceived forms were feasible in this case, since most Marshallese at Arno Island are literate in their own language, and many have become accustomed to reckoning in numbers and by weight through _their transactions over half a century with foreign traders and tax collectors. Within the three-month period, the anthropologist became acquainted with each household, providing thus a better basis: for evaluation of the daily reports, and had adequate opportunity to take his own observations on economic activities of the people. The three summer-months, roughly from- mid-June to mid-September, coincide with the main breadfruit season, the most productive period of-the year in terms of food, The results of the study ~~ have primary value for this season of the year; similar studies should be | carried out at: other seasons for a more complete understanding of the eerie | round relationship between Arnoese and their natural environment. Subjects for Quantitative study gee The following subjects were basic in the daily reports from each of - the households selected for this study... ‘In general, the questions He be anisweared were: How much work ‘is Geconplished? Where is the work undertaken? How many persons are involved? Who are they? What is their social Prue How much time is’ consumed in each activity? How much leisure: time remains?. How is it spent? . 1. Food collection and. preparation: Kinds and ationbtg of food materials assembled (plant , oe” marine, store); time consumed in different stages of preparation; artifacts and accessories required; persons employed; location of activities. * See gample cards attached (B and C, pp. 119 and 120) -}118— 2. Food distribution and consumption: Food exchanges (raw, cooked, store); persons involved in exchanges; their relationships; quantities and kinds of food exchanged; occasions for exchange; quantities and variety of food consumed by individuals; number of meals, times, and composition of eating groups; visitors, at mealtime, their relationship with host; their treatment by host; foods consumed by visitors. 3. Copra production: Number and size of nuts husked; weight of meat cut from nuts and dried; weight of copra sacked and stored; weight and value of copra sold to trader; location of each activity; time involved; persons employed; distribution of income from sale of copra. 4. Production of artifacts and handicraft: Amounts and kinds of raw materials collected, prepared, and processed; types of articles produced; value of articles sold to trader; use of articles retained; ownership in relation to producer; location of each activity; persons employed; time involved. 5. Clothing production, laundry, and repair: Items of clothing produced or bought; value and quantity of clothing; materials employed in laundry and repair; time involved in work; persons employed ; location of each Sigil 6. Construction, and repair of housing | and water ee Soae: Quantities of raw materials for use in construction of housing units or canoes and boats; processing of raw materials (time consumed, persons employed, location of activity); ownership and use of completed _ structures and craft. 7. Weather conditions and events competing with economic activities: Frequency and duration of weather conditions impeding work progress; Sundays and holidays; festivals and visiting; recreation and other leisure-time activities; frequency of these, degree of disruption of work ia: det persons involved. Se Penne of versons. Travel from one tanesnotasae to another; from Arno Island to other islands in the atoll; from Arno Island to other atolls; kinds of transport used; occasions for travel; length of journeys; duration of stay; persons involved. 9. Participation in money economy: Receipts of money; source of income; distribution of income; purchases and other transactions requiring money; persons involved. -119- SAMPLE CARD B. INVENTORY OF LANDHOLDING ~ Arno Project - 1951 Name pa | Informant Lot No, Iroij Iroij erik Alab CASEEEEES NOS SE ee Ce Occupants Buildings a SD SEES LDL, Sail canoes Paddle canoes Bicycles Other transport Land acreage pate sae Pigs Chickens Ducks Others Breadfruit Coconut Pandanus___ Banana Papaya Mangrove. Others . Taro “Arrowroct as Sweet potat ss Cisterns | Wells Household effects Money ~120- SAMPLE CARD C. QUANTITATIVE SURVEY = ARNO ATOLL PROJECT ~ 1951 Household ___ 1. Re 17. Noon: Evening: _ Informant _._Date How much food was eaten here today? Morning: How much of this food was prepared before today? How much food was prepared here today, but was not eaten today? -Who helped to make the food today? Where did-the-food.come. from that was eaten or prepared today? Who ate the food here today? How much food was sent to people elsewhere? What work was do Tip SIP ER SIS Sate SES ai) GG! ots ne on houses or canoes today? Waerdid the work? aa When? What materials were used? . Where did the. materials come from?.. How oh money was earned by Rati here today? HH KH HH HH HH HE. \ How was the money - earned? Who worked for the ‘none? CE Who received the money? How ase money was apent here geasee To whom was the money paid? For what was the How: many coconut Who.did the work? money spent? ‘Who spent it? | RRR HHH HHH H s were collected today: ¢o, husking? _Where? 9 (Sample Card C - Conttd) | a 19. 326 33.6 Zhe 356 36% -121- How many coconuts were husked today? __. Who did the How many coconuts were cut today? shel Who did the How many bags of dried copra were sacked today? Who did the work? work? work? How many bags of copra were sold today? TOs WAOM Zivsncsrvitne Who sold the copra? How was the money divided? How much money was received? che Ein Sy thyocie ei are eke Sie ae What new clothing do you have here today? What clothing was washed here today?_ What clothing was ironed here today? What clothing was mended here today? _ ee a oe What handicraft was made here today? Where did the materials come from? Who did the work? Where is the handicraft now? Who owns the finished products now? % % eH HKU HK HK KH How much water was used here today? For what purpose? Cistern: Well: Lagoon; He KH He HH HH What work was stopped today because of bad weather, or for other reasons? What other things did you do today besides work? Whom did you visit today? Did anyone from here leave today for another place: house, island, or atoll? . How did he travel? How long will he be gone? Did anyone arrive today to stay How did he travel? When was he here the last time? Where? When? Where did he go? Where did he go? here awhile? Where did he come from? Why did he come? Leen EEE iranian nEnnEnnnnnRennnnnnann -122- INVESTIGATION OF MATERIAL CULTURE by K. P. Emory In the determination of what in the culture is due to the particular environment of the atoll being studied, and what due to the cultural influences brought to the island, historically, it will be necessary to record details of structure and technique in the material culture, native terms, and what the people themselves have to say about origins. For content and terminology of material culture investigations on an atoll one should have at hand Dr. Buck's "Material Culture of Kapingamarangi", Bishop Museum bulletin 200, designed to set a pattern for such recordings. What has been known of canoes is summarized by Hornell, "Canoes of Oceania," Bishop Museum special publication 27, vol. 1, pp. 345-361. Time would be saved by |. having a copy of this and simply noting anything different, new or additional. | Of special importance to record step by step in every detail is the preparation of pandanus flour, pandanus preserve, breadfruit preserve, prepara- tions of arrow-root, etc. A photograph record of these in color as well as in black and white would give us materiai comparable to that we now have from the Marshalls and from Kapingamarangi. Always inquire what the ancient methods were like, - if they differed from the present and how and why. An effort should be made to collect shell or stone adzes, ancient fish hooks and ornaments that the people may have found. Adults and children could be encouraged to hunt for them during the stay of the expedition. For the island of Onotoa, in the Gilberts, it would be interesting to know if they also made the coconut fibre armor and if any now living know the - techniques. Also, do they know how the shark's teeth were bored to be lashed to weapons. Tamana, next atoll south, seems to be the island mentioned as the homeland of the Kapingamarangi people. If so, it was once inhabited by Polynesians. Is there trace of an early occupation of Onotoa by Polynesians? Arthur Grimble's, "The Migration of a Pandanus People," Journal of the Polynesian Society, Memoirs no. 12 (1933-1934) is our main source for the Gilberts, revealing how little we know of them. Chapter 10 — Miscellaneous HINTS ON TROPICAL PHOTOGRAPHY by C. J. Lathrop Fear and uncertainty are the photographer's worst enemies when it comes to photography in the tropics. The photographer who comes from a temperate zone into the warm humid tropics finds his formulas and cal- culations, based upon temperate conditions, upset. Temperatures in the tropics are generally no hotter than summer in many places in the United States, and the other factor, humidity, may be just as high. New York when it is summer and in the :"90's" can be just as humid and hot as most tropical localities in the Pacific, but, of course, not as continuously. Selection of film (black and white) If you plan to use a miniature or 35mm size film in your work be sure to select a film of the finest grain available. Either (Eastman) PANATOMIC X or (Ansco) FINOPAN are excellent. While it is true that these. films are rated as slow films the results in enlargement will more than compensate for the loss of speed. I enlarge all my shots to 8"xl0" and without exception they are better shots than 4"x5" shots on fast (100-125 speed) film. Insist on all film being tropical-packed. If your film is ordered or purchased:in Honolulu it will be tropical—packed. (KODAK HAWAII LTD.) For cameras using ordinary roll films my selection would be (Eastman) VERICHROME, Care of film and: equipment in the field Two conditions that are not favorable to. films are heat and moisture. If your film is tropical—packed the problem of moisture, before the package is opened is not a factor for consideration. After the tropical package is opened and the film loaded in the camera the best thing to do is use it up as soon as possible. The film should not be returned directly to the tropical package. (in most cases the tropical package is a screw type metal can.) All film must be desiccated before it is resealed in the tropical can. Desiccating is a simple process whereby the moisture that the film has picked up after leaving the tropical container is absorbed by a desiccating agent. Common dried rice may be used but silica gel is easier to handle and can be re-used again and again. A large friction top or screw top can or jar is used as the desiccating container. The dried silica gel is placed in the bottom of the container. The film should never come in contact with the silica gel. One should purchase a chamois skin bag for the film to be placed in while desiccating. An indicator should be put in the container. These indicators change color when the moisture content increases to a danger level. When the air is dry the indicator will be blue, as the moisture increases the color will change to a pink. As an easy way of remembering the color, Red means Danger. The indicator should be in the container at least 4 hours before a reading is taken, as the moisture in the air will turn the indicator pink before the silica gel has time to pick up the moisture. The film may be resealed in the tropical cans after being in the desiccating can for 48 hours (and the indicator is blue). ~123~ “lhe, ) « St * To reactivate the silica gel it must be heated in an oven at about 300° to 400° F for 30 minutes, Some times it is done by placing the silica gel in a metal can or container and securing the can over the gasoline lantern during the evening.:: After the light is turned out the silica gel is placed in a container: with 4 screw or friction top. The top is not’ put in placed tightly, as, on cooling,.the container would collapse with the contraction of the air within, forming a vacuum. WARNING: Do not put the plastic indicator plugs in the sir silica gel or in the oven as - they melt at a low temperature, In resealing the tropical film containers after desiccating I -find the SCOTCH MASKING TAPE 3/4" that is used by auto painters to be excellent. Adhesive tape: can be used but electrical tape (black) is not satisfactory. Cameras should be fitted in cases and packed in moisture tight cans when not,in use. The metal parts should be treated with wax. I find that transparent (clear) Esquire shoe polish’ is ideal for this use. I have experienced no film discoloration from. fumes with this preparation. Some types have an objectionable compound that will spoil Kodachrome due to the volatile | chemical content, eyen though used sparingly. Clean your lenses frequently with a lense brush or tissue. As a fungus inhibitor, a small wad of cotton is placed in a glass in the camera case or can and a small amount of formalin added. The formalin gas in many cases will keep your camera free from mold and fungus. This may be done several nights a week.with*the camera empty (no sia oe processing Y ehie £m in the field I find that the early morning is the best. time as the temperature is.at its lowest point. Eastman Kodak has produced tropical developer, tropical hardener, and tropical fixer that” give a fine Sore fog-free negative at, a vies oe to GeO Rie: HINTS ON LIVING ON A BOAT by J. E. Randall, Jr. Living on a small vessel, either when underway or while in port, presents problems quite new to the uninitiated. Some of these problems are concerned with the operation and maintenance of the vessel and will not be considered heré Others, however, are personal or individual in nature and arise largely from the closeness of the association of all persons on the vessel and the necessity for a strict, scheduled sort of an existence. ¢ The following, in brief, are suggestions for the carrying out of a cruise on a small vessel. Some of these suggestions may not be pertinent if a large boat is utilized. ; One, and only one, man should be chosen as captain of the vessel. He should be capable and experienced. He should be placed in a position of total responsibility for the operation of the vessel. He will make all. important decisions which concern the vessel or its disposition at sea, ard such decisions should not be disputed. The stowing of gear before departure should receive very careful thought.--- and planning. All items which could become detached when the boat is in motion should be lashed securely. Canned goods or any items which could be damaged by contact with salt water should be stowed in a place which is known to be dry or should be wrapped in waterproof or water repellent bags. It would be wise to place all extra clothing and bedding in such bags even when it does not seem likely that they will get wet. Also noteworthy of mention is the covering of mattresses with zippered plastic covers. All items of equipment necessary to the handling of a vessel must be kept consistently in the same place. This is especially true of things like tools which will often be needed in a hurry and at some unexpected time. This point is sufficiently important to warrant explanation to all members of the crew at a time before the vessel has set sail. When underway, a strict schedule should be enforced as to the duties which must be performed. The delegation of duties must be carried out before the vessel has left port. With the exception of the cook, every man should stand watch if capable. The duration of the time.at the helm should be talked over and decided upon beforehand and provision should be made for a rotation in time so that one man will not have watch at the same period or periods every day. Great care must be taken to see that vital jobs like the winding of chronometers, inspection of engines and other items of equipment are executed at the correct time each day. One man should be held responsible for each task of this nature throughout the voyage. The greatest danger at sea is falling overboard. All steps should be taken to minimize such a hazard. An adequate railing should be erected all around the vessel. Life rings should be readily accessible and never lashed in such a manner as to delay their prompt procurement. Water lights should be available. These are waterproof lights which should be thrown overboard at night with the life ring. They contain a mercury switch and become lighted when floating vertically in the water. They serve the purpose of guiding a person to the life pe and, in turn, enable the vessel to locate the person who has fallen over- -126- It may be wise to trail a sturdy line astern which a person can try to grasp if he has fallen overboard. Such a procedure is vital if a person is standing watch alone. In heavy weather the helmsman should always fasten a short line from himself to the boat. Fire hazard on a boat is a very real danger. This is accentuated if a gasoline engine is used or if gasoline is stored anywhere on the boat. Certain safety regulations concerning the location of tanks, etc. must be met by all boats with gasoline engines, but a hazard still exists. If gasoline is stored on the boat the fittings on the lines should be inspected every day. Such fit- tings have a way of coming loose from vibration and this results in a dripping of gasoline into the bilge. It is just this sort of thing which causes the _disastrous explosions on power boats so often read about in news items. No gasoline-burning stoves or lanterns should ever be used on a boat. Other potential sources for a fire at sea often exist in butane, kerosene or alcohol stoves, kerosene lanterns, electrical wiring, and smoking. No one should attempt to light.a stove or lantern without proper instructions. Electric wiring should be frequently inspected. Smoking should be eliminated. Since this is practically impossible to achieve, the best alternative is to designate a specific safe place on the Oe a smoking - Every vessel must. be saan with a suadg rida number of fire extinguishers. These should be inspected. before departure. CO9 fire extinguishers are superior to the pyrene type, and at least one of these.should be on board and very accessible. A very desirable type of fire extinguisher to have is a built in CO5 system in the engine room sane Creates a 6h when a certain high temperature is reached. : wd The restricted use of fresh-water by all members"of the crew must be emphasized. Unless a very large supply.is present, water should be utilized only for drinking, cooking, shaving, and limited washing purposes. It may be necessary to wash clothes and dishes in salt water.’ Ordinary soap under such conditions is useless. Salt water: soan. in general,.is not too good. A deter- gent such as Dreft seems to-be the pest’ agent for washing with salt water, An emergency water supply should be maintained in a tank separate from thé regular water line. ae of water ae for possible leaks should be made periodically... The scourge of any cruise is seasickness. In a rough sea and in a small vessel very few persons are completely free of nauseous sensations. Each. individual usually finds a set of conditions which best alleviates his distress- ing symptoms. Generally, if sickness has come on while below in a confining cabin, going on deck in fresh air is very helpful. A recent drug, dramamine, has proven to be amazingly effective in preventing seasickness. It must be taken before sickness ensues or it is not effective. If a person knows that he is very prone to sickness at sea he should take dramamine a half hour before departure, and continue dosage every six hours. The drug has a pronounced somnolent effect. For.this reason, the dosage. recommended by the physician may be experimentally lessened so that the minimum amount necessary to prevent sickness is obtained. Dramamine can be procured only by dactor's prescription; it is considered very dangerous to persons with low blood pressure. eye Sleeping in a vessel which is underway is an art in itseif, especially in rough weather. Most individuals will sleep only intermittently during their first few nights at sea, but gradually will adapt themselves to the unusual motion. Narrow bunks are often preferable and the sides of the bunks should be equipped with bunk boards or some similar arrangements to prevent the sleeping individual from falling out. The sides of the bunks should be well-padded with extra bedding. A cruise at sea, especially if it is a long one and on a small.vessel, presents an environment of close quarters and unpleasant living conditions which is usually quite unexpected to thoge who have not experienced it before. And the knowledge that there is no escape from this environment until the cruise is terminated may make these conditions even: more difficult. It there- fore behooves all members of the crew to make constant effort to be congenial, to do at least their share of the work, and to be as considerate as possible of their fellow crew members. ; There are a few points worthy of mention regarding living on a vessel which is lying at anchor or tied to a dock. First, there are often very dis- tracting noises which arise from movements of the vessel against fenders on a dock or alternate tightening and slackening of mooring or anchor lines. Many of these noises can be eliminated, and effort should be made to do so before turning in. A breast anchor can be taken out from the side of the vessel away from the dock and pulling in on this line will prevent contact with the dock. On anchor or mooring lines pieces of elastic cord or inner tube can be incorporated to reduce the sudden jarring effects which the boat transmits to the lines, This saves wear on lines as well as eliminating the noises wien occur when the lines tighten on boat fittings. Adequate ventilation presents another problem. Frequently one can create a current of air through the boat by erecting a funneling arrangement on the back side of hatches facing the wind. An elevated hatch cover may serve the purpose or a more elaborate affair may be constructed from a large piece of canvas. Usually there is no provision for preventing the entrance of flies, mosquitoes, or other insects into a boat. Pieces of mosquito netting over hatches and the cutting of round pieces of screen wire to fit into port holes will help materially. Aerosol bombs should be on hand as well, however. Since space on a boat is so limited, cooking or bathroom odors are generally a source of annoyance. Airwick is a big help. While in port and in hot weather, decks tend to leak due to the drying out and consequent opening up of deck seams. The best preventative is the wetting of decks at least once a day. Salt water should be used for this, for constant wetting with fresh-water promotes dry rot in a vessel, Megiceriy t ei on Pte He 0 Ra eB ihe fs sche eeite a care mp fet cin apbipte NO telat auienwe oft od qovlpamane engine ”E. spt > yas: x sett eee ay ‘en Ea: = LTT oaths eae > ~~ = ‘ Dich sea aw shyt eer Teh Le: ‘a ve" (oe pret Teh icc agin pebbaqed at set sihneit ie iat: we ao 8 anise dai! er “Oh + Ramaol ideale a! i fas @sva) ashi 8 we pat: tf x nha pmkritnos yabrhe Araawiger Tyre bine! etidanup op ba Ba ached t4 Sagoy- PE sos ead opin epard: ot Laks etd Dey? tapas toate" elie a ch Senin. rere 3h "Nae ER: ee ghee ane eaes < ne jis IVE’. poker ited her + goer ie irepites ad OF: 34 ko ales eat on at stato 6 ae ée aa em: maid ter * Re : Biupe) r OB, Crea aN, F a * 4 v Tacs Piast Rae ae iene? ‘a 347 Nabe a BY eter nev eek > # ee. vies! aad * wg ebeoit Ke? omit wn a) Yeiarensiy aS sgh Yaes Peed wonk ont # pay | , ad’ #80 ise apoenett ao ieres ohne) ; a* oy frantest send mes fis tier ho joatte y anal, tt doi ae. nee Le owe al ii oN oe a eye coe ‘doit? mn ‘a Way 9% 4 re thi a i sit we *Snmoasas eaten et @ antéoste “yl dane abies: ea “pence (eogart betters : Elin bee ee pao, hesisuradun . e Saal : Ler Caran 8 ektt te pelsanieie ‘ed gnibane (att wobebewag aa ei shades ev yoid tage pion page tw Beco rt sont «ode eabeath poo HO. 4 Ae obit ane oR ALI Ta Te Ps betegtent: See geet = eetod 3% : wnestie ot ie adi aanil acl an s vibe seme arty no sagiwtas, Ce ahha Erin batts ifs, eh sot ee ol hay ‘yathee ite a a RENEE, a gaterh sas os ‘i so abi ‘fant inbme pela: a ei wwiaginoe ony heads nant Qa : iv ‘et ot freaiey’ gut! BE ponte ghia hed Lanoh it . tao? o we i set moe i igs awe dont i te 4 mate ain fae -128- HINTS ON LIVING UNDER RESTRICTED CAMP CONDITIONS C. K. Wentworth Small groups of people living in remote places with limited supplies, communication and transport often become acutely aware of differences in tastes and habits between individuals. While possibly confronted by less potential danger than a party in a small boat, a small land party has nearly equal need for a single leader who must make final decisions on camp dis- position and major procedures, The writer has been as guilty as any in failing to recognize those of his own habits that were or might have been noisome or distracting to others. The text below is an attempt to suggest possible varieties of offense. Each must, of course, be weighed in the balance between its importance to the individual or the work of the expedition and its possible discomfort to other members, The camp menu must depend on wise planning, successful transport, preservation, and preparation, often under difficult conditions. At the best it will be a kind of average in taste with little support for individual whim. If you have a small whim and will balance the needed transport and care against some small personal sacrifice in another direction, by all means make your own provision. But don't expect to pilfer the group supply to satisfy your own improvidence. If you have a large whim that involves transport don't go on such a trip. Individual tastes must in some degree be met by rotation, both in the menu and perhaps in the preparation itself. If you can't give and take in eating what other people adore, or what the existing larder provides, at least keep quiet about it. The camp leader may have to ration certain supplies and may have to invade private supplies of drugs and similar articles in emergency. Persons requiring special drugs should make adequate provision. It may not be superfluous to mjoin orderly, perhaps assigned or rotated care for wood and water supply, stoves, and lamps, outside the obvious need for such handling of cooking and k.p. duties. A good bit of fussiness is needed in care and placing of both personal effects and group articles, flashlights, matches, perhaps firearms and the rest of the gear that may have been brought along. Some people are more gifted at this than others; the less gifted should at least help up to the point where the more zealous fusser is clearly a nuisance. Differences in taste for ventilation and other sleeping conditions may be met by little choice, but a bit of negotiation can be a wise medium between suffering in an explosive mood and exploding in an endless tirade. People who snore may not be able to do much about it, but should be willing to accept the serious testimony of their fellows and accept such segregation as conditions permit. Personal laundering, caring for specimens, dressing and preserving of organic materials, can be slightly or quite obnoxious to others of the party. While continuing necessary operations, try to be aware of such possibility and make any adjustment that is feasible. The practices of your craft, even those approved in this manual, may ad nauseam seem silly to someone in another craft; a great deal of fine interchange can take place but try not to let the operations get rubbed in on the other fellow because of your own enthusiastic inattention. > as ‘ “") soca wt 16 aya Sohn,’ 30 7 . e yor wert Sued, Pred en s pours ¥, ae ti od Sh. BHDL AE biti wy sare xBkb ‘pe baine £6 dto yak daieqene, sant “tauthire Lon “i iponmmse_¢ foeih adi sides ed Rta aaarees. 3 sop eteo SRB Saoratterss ce ” rene eck panna tLe ye fo bias cor pretties oF pis Paes Hob outers? eovlorci: ipttetes ee: dagad eerrnel,. i ; ote aS ge: BOK 1 at oats ‘Abbaye wed. ke hives ~ gebrel aniteies odd dar to 4e2ebe akgaor: ; Atsdtey (pobie 6d oved ‘yaa iene Cptiont. Line waLios ore a. ie i oie asa ode, @ WHOLE ivy, etapobs wale. bition wc SYED © MOTE) htod 4g pois sake bre oreo of oa bos aareerehs. .6¢ gd qaoe ri Petts 8 coon kins quota baw oe vrei: rie 29 Oe eld od al gine. push a y inode er sisanlbere a vino al. vee ih pualses. Lape rere: bar “merece, fits a a WOU? 4 4 idee YtReB Ao 28 dap & bei | . oat 3 Et a) qi mamas Preto rediein ot oppemoe OF: Sahai ean spi hn tan vregnns tigi edt dat os gos ; enki "to foek Jeong & | ot dap tions vitaahelsne ae 4 pee bas paket mee bo. sore a bop paneer end phiesve 2am BOG ROWS. a whee Af MOTTIGNOD WAL ORM ATE 95. nd ogee wo 1 emcees te o, ese i al aoa: bed tints Hiby esonkg osomon st “gabved ine i) noone VEE, "bo eaawe qLossoe Pasuiueriih nooo dmonaaert’ a beste nae yidhorey CR wa ea bik bith “pranirat heeeh Lies - n OS gitar Risks) pl med. 0 BAIS betooh Lani® eciva domi oriw sabark ed, ah 7 an, Yilleg as. ned suet Raw eft «tom at wend a oe shoal ren ak ie ‘et. wood. dove ete te. , toate: see ‘ Seegiirs ato doom iay oat 0 out thot ed pone) 7 pia cnt herabee a8 Sear vt ywakromba eebe a0 ak Pee —— ——— Se peered Litianaboue x wth apekod LED nibh re ONL, ak 90: wid awit od. bu | ; chaste mate agg. 3. aoe Ch Tt ‘omen’. Lal cytes et aint | Setoore, gunner 88 ee | Larabar gieind fae One pleco ® “ea weet % a em neubihovep, 20 ‘ tether "hor fein tesa Bey yeid'ne q oe bal peor ak ohankeiaat: 48 ake boea. iis rc “in oa ete Siac aor. oe dieapond pened oval) Yea sere “e088 bovtig: weal ole qoredtdo | (ot ben ytte mea Liy ‘1: wide: bots 60 1! tabi jetery “ect phone: cia eunaeehtil). : ors by ao oe ne mobtah rapt: Lo, tide Sane ype tol, olson ye eet wtan 4 ce bee ENO avdaoleen we of gobnetion: neewted | net Rukaseeln’ ry at ri thvaday Cneage ok. ot. alain, ee PORT YRS vom ote algqoe , awolie? send. 20: 2 ener caakves oft. dqeone 7 Abe kews Zz t ire foe a at 126 Ew C9 ria deine gels y CTRL ayant ye ei bee LOR «gchno bean £ Lope. ‘ als "te sanietd i ad Biro. LO wit . Sb ituty thes ac ie tae “Snare 3e 4 ihidhaee aoon heyunen ae Od CL etek ama yinaeaoen jrctsinikenoo gov ahs o WO, Ne gat Ee nrg gett 4 eed maa gk Sarid Jovoust.cat| Eas Eee ok nam atid OF paieeinanoy | uid hid emblq cope sn Sf, macht | ‘ner, hes wonnond ad fat ant oat), ne “th beddire fo8 | bf . -129- Turn about with him as much as possible in operatioy:s that are within your own capacity, when you are in camp and he is not, and’ je like. On the other hand it must be borne in mind that these activities ev if disagreeable, are the work of the expedition and the reason for it, anr hat tolerance of them may augment the success of the enterprise. Some of us talk too much, at times, others in compensation are reticent to the point of impeding group success. Neither extreme is best. The heavy talkers can be very boring, I have many times been told} But the quiet person owes it to his fellows to leave adequate daily word which way he went and about how long he might be gone; in such situations it is of concern perhaps for his own safety. Also some sharing of knowledge of home addresses and location of valuable or emergency articles in equipment is an obligation in the group interest. Some things can be needed in a hurry, with or without categorical permission. Recreation has its place even on a hard-working expedition, but should yield priority to work around the available light, and to sleep. Noisy game playing can be an unwarranted intrusion at times. Singing and whistling may be among the N freedoms but in long doses may also be far less inspiring to the hearers than to the producers. A balance may have to be struck between cheer leaders and joy killers. teed ALO NA Via ge MAS ~ No. 18 | July 31, 1953 ATOLL RESEARCH ~ BULLETIN 18. Ichthyological ree Data of Raroia Atoll, Tuamotu Archipelago (THSOWWS AUG 111953 LIBRARY by ROBERT R. HARRY Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences—National Research Council Washington, D. C., U.S.A. ATOLL RESEARCH BULLETIN ee ee ee 18. Ichthyological Field Data of Raroia Atoll, Tuamotu Archipelago by Robert R. Harry Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences—-National Research Council Washington, D. C. July 31, 1953 ACKNOWLEDGMENT It is a pleasure to commend the far-sighted policy of the Office of Naval Research, with its emphasis on basic research, as a result of which a grant has made possible the continuation of the Coral Atoll Program of the Pacific Science Board. It is of interest to note, historically, that much of the fundamental information on atolis of the Pacific was gathered by the U. S. Navy's South Pacific Exploring Expedition, over one hundred years ago, under the commend of Captain Charles Wilkes. The continuing nature of such scientific interest by the Navy is shown by the support for the Pacific Science Board's research programs, CIMA, SIM, and ICCP, during the past six years. The Coral Atoll Program is a part of SIM. The preparation and issuance of this Bulletin is assisted by funds from Contract No. N7onr-291, Task Order IV. The sole responsibility for all statements made by authors of papers in the Atoll Research Bulletin rests with them, and do not necessarily represent the views of the Pacific Science Board or the editors of the Bulletin. ee ee ee ee Editorial Staff F, R. Fosberg, editor M. H. Sachet, assistant editor Correspondence concerning the Atoll Research Bulletin should be addressed to the above c/o Pacific Science Board National Research Council 2101 Constitution Avenue, N. W. Washington 25, D. C., U. S. A. Ichthyological Field Data of Raroia Atoll, Tuamotu Archipelago SCIENTIFIC INVESTIGATIONS IN MICRONESIA Pacific Science Board Naticnal Research Council Robert R. Harry Academy of Natural Sciences Philadelphia, Pennsylvania May, 1953. PREFACE In the last decade, almost all zoological (and particularly all ich- thyological) investigations.by the Office of Naval Research and Pacific Science Board in the Tropical Pacific have been in the Trust Territory and neighboring island groups, centering in the Gilbert, Marshall, Caroline and Mariana Islands. The choice of Raroia Atoll in the heart of the Tuamotu Archipelago for the third low island in the Coral Atoll Program (1952 SIM Project of the Pacific Science Board of the National Research Council) provided the first time that ichthyological research had been sponsored in French Oceania by the U. S. Government since the Wilkes Expedition of 1838-1842, and the "Albatross" Expeditions of 1899-1900 and 1904-1905. The resumption of this interest in a little-know region of the tropical Pacific should prove very worthwhile, and the author, who participated in the Raroia program as ichthyologist, respectfully submits the following report of his field re- sults. Other reports that have a direct bearing on the fish study at Raroia are the general introduction and geological report by Dr. Norman D. Newell, the list of Raroian names for fishes by Bengt Danielsson and the author, the report by Bengt Danielsson on native fishing methods and his general report on the people of Raroia, and the general report on the ecology of Raroia by the entire field team under the editorship of Dr. Maxwell S. Doty. Le Acknowledgements . ...4.s. Titustrauions . sess 5 8 » Introduction ..... » Scope of Report ..... List of Equipment .... Station Records. * * == s Ecological Survey . . Coralliferovs Outer Bench Surge Channels Coraline Hides. . . +s Outer, Reet Flat «). « :s » Inter-islet channels ... Lagoon shore reef flat . se e e e e Lagoon shore reef face . Lagoon coral heads (patch Systematic Account Histegiez! Introduction ...... +» « Tentative check list of fishes previously CONTENTS reefs). . es es es s 2 e, s recorded s e e from Tuamotus e . e es © s e ° é ° se es e e e e i e Field Data of the Fishes.collected at Family a suridae e ° e e » e e 2 e e " Galeorhinidae . i Orectolobidae . , Aetobatidae .. ' Mobulidae ... i Albulidae ... n Dussumieridae . " Synodontidae .. “i Echidnidae ... . Echelidae ... m Ophichthyidae . a , CORPEIOAC | a) sue oe Moringuidae .. Belonidae ... my Hemirhamphidae . io Aulostomidae . . * Fistulariidae . eo ‘ e e e Raroia es e 2 e eo « es e e e e e CONTENTS (cont!d.) Fanlly Synemethides . 066 e kee eae 5 si’ uli \Ady ea al Nn Holocanteidae a )a ea ale eg eal he ies " WMugilidae opi sae te athe cee ees We ctimea hme Cae ie ne en ® Gepreemddae i ae ie les es ee eos aes a " Polynemidee . . «+» ss » » Bo ae Oe te Re Ae 71 Oo: SOPPEnsdse: cies see wee ee wl ee Gere ee ree mM. “Poeudoenronsdae 5 cise. 6 ww at seem mt oa ms) oan ts te cae MY; GSRREIOSGBS alike: ia ew: Cathe ie eo 60 8, ha Da a a tt mn) BVGSAs ERS Celie eel le woe le ee ee Ver ae EN, tan en aes Be) RAO we arse te ce ce te de ee) rae) ie li aes ee 82 WM. Adseontdne ff f2 2 2 oe dd Sobel ee Rm Kyplosidse Sf ec ree hee oS ah eee eee eee 1 ORE RENEE ea ea le Be wie ee mw, Re mL) ha i oe a te 85 o> Pemoperiedet fo of Ff SPS ee eee eee eee OS Carepitadae «ee oR ot eS a ee ae ee Ree ee 86 PY Cictodentsde ee ee ce ek ee eee we Pandeeopeddate 6.2 2 se Se Ae 2 oe Se See eee 99 Pe Lapras PS 8 2 ee A MS Se Shee ae ee C SRSESSRE a WS ks ee ee ee ee ee ae ee eee pe ib es 22 ed ar ae Me gi a RR a ea ce St Ch ach 2. BA Caronsidue 2g oe 2 SE ONKS PS, DEM POLO, SPUR Poe oe ieee T\ Smimidesst oe oe owe ee ew Re a eee FAP mee \ Paraperesdac o « ow bh oh we et ot et ot ot co Mee 130 | Bleaniidee s\n a wn we wo a no eNO 238 i Bs Plegtrigne s,s, ss 0; os) Oe eh ay Gh we op eR 1 Ls GODUA GAG say ay by Ss) by ay Sy hee ee) Sy tek we! oe ab 8k ae ee Me GODIPSOCTAAG) =, 0, sy. o5 ey ss my ee oy my Sy oy oy SS) ee CO a M.. Brotmaadse . sass oe, 6 8 5, 9 ak @ pieetiatnen Hommel Lame N BENSAGHGTORE ble eel 6) e616!) we whey ol 8) eae palaces Deana 1. REARUNP IGS ie eho ia 6) ie ae a lal ec het ah Neues eee ae es Mo PEMA POB ig! ie a athe’) ww: wie. bie! elidieh) Ver Lise) se a ae ae ORR PECs SI 3) Ik EO ag aR create MENGE RD VV OUR ESHA (91S 2.2 w PBE ACETONE a) elite.) 6 ee) wwe ta ig ie ieee ee wie el pes ae WO BOMGE CAS Mia sol ee wt sel bi, hie ve ceaws eary a alt eo arta abiaeei bam ace nea . WT BOM POEIARS: a ia ia ae ek aa ia aria iat (anor ey Vei Nleeh Calieaey ie, ea ea W MOMECOREDTAGE eel a ee iat ecrin eee ee ee ULneiels enue PCLPASAOMTAGEE Gig ela) hile ee ek aT Oe ie er aetna ; CONTENTS Family Diodontidae o 28©- © ©» © © @» 6. ‘s' " STOEL T NARS wa dil ali iain tlw usin ih Anternnariidae oi (elie le ire Check list of the Islands of the Tuamotu Native Uses of Fishes at Raroia Bodog Resources 1 « s = 6 «6 «J CakehnStabistaes eS Si Fishing Methods and Localities . Poisonous Species Tuna Bait Resources e e e 9 s (cont'd. ) Archipelago e e ° es e e e SMart ee el re i aie Raroian Vocabulary for anatomical features Conclusions e e e of fish Notes on Spear Fishing Equipment .......s.es e e s s s es es s e e e ° 2 s Reterentes oss sis ss «3 6. @ ee 68 2 am 8s e Page 167 168 169 170 176 176 176 176 7 ain 17 ‘Aye, 179 | 161 184 186 ACKNOWLEDGMENTS This section would be almost endless, if everybody were acknowledged that helped with the fish study at Raroia. The fieid work was made possible by the Coral Atoll Program of the Pacific Science Board of the National Re- search Council and was supported by funds granted to the National Academy of Sciences by Mr. George Vanderbilt and the Office of Naval Research. The fish program was also under the auspices of the Academy of Natural Sciences of Philadelphia and the California Academy of Sciences. Generous cooperation was received from the enthusiastic team of scien- tists on the Raroia project, from the administrative officials of French Oceania, from Miss Ernestine Akers, Mrs. Lenore Smith, and Mr. Harold J. Coolidge of the Pacific Science Board, from Dr. Earl S. Herald of the Cali- fornia Academy of Sciences, from Miss Odette Vernaudon of Société Franco- Océanienne, Papeete, Tahiti, from Miss Aurore Natua of the Polynesian Museum in Papeete, from Mr. Edwin H. Bryan, Jr. of the Bernice P. Bishop Museun, from Dr. W. A. Gosline and his students at the University of ae iene and from the Office of Naval Research. Words cannot express my gratitude to the people of Raroia. ‘They were generous and cooperative beyond all expectations. And to Huri and Rago, who spent many days patiently teaching me what they know about their fishes, I am deeply thankful. All the publications on fishes together could not give the insight they imparted to me of the fish life on coral reefs. Special credit shovld be given to Mr. George Vanderbilt, who generously made the fish study possible as a contribution to his over-all program of research on Pacific fishes. Mr. Vanderbilt provided almost all the fish equipment as well as additional funds necessary. And none the less important has been his sincere personal interest and inspiration to the author in his fish research. Grateful thanks are also given to Miss Albertine Talis and Mrs. Aldona Mosinski for extensive aid in assembling this report, and to Miss Florence Sprague for preparing the illustrations. To all the remaining people who helped in the fish study I wish to express my gratitude and appreciation. aiken ILLUSTRATIONS Figures Field stations of the ichthyological survey. Raroian names Pig la essnene of a fish. Ecological seh of fishes on outside reefs. Ecological zonation of fishes on snore reef face west side of Raroia Atoll. Ecological zonation of fishes around islets of Raroia Atoll. Daytime ecological zonation of fishes in surge channels, Raroia Atoll. Ecological zonation of fishes on small lagoon ccral heads, Raroia Atoll. Tables 1&2. Tentative check list of fishes previously recorded from the 3. Tuamotus. Index to families of fishes collected at Raroia. ae INTRODUCTION Scope of Report The purpose of this report is to present the ichthyological field ‘data’ prepared at Raroia Atoll and subsequently assembled at the Academy of Natural Sciences of Philadelphia. It has been prepared to show the type and scope- of information gathered, the general extent and size of the collections, and the field results obtained. It is hoped that this contribution will aid fu- ture Coral Atoll teams to correlate further the fish research program of the Pacific Science Board and to profit from the information gained on the Raroia study. While this paper should prove of some value to other ichthyologists, it is recommended that it not be cited until the final reports are published. The aim of this contribution is entirely for the benefit of the coral atoll studies of the Pacific Science Board, and definitely is not a formal system- atic account, or a definitive study. The studies of fishes have formed an important part of the Coral Atoll Projects of the Pacific Science Board. Large ichthyological collections and considerable data were obtained at Arno Atoll, Marshall Islands and Onotoa Atoll, Gilbert Islands in 1950 and 1951, réspectively. As originally planned, the fish program was not to have received as much attention in the Raroia in- vestigations. Dr. Morrison and Dr. Doty’were to have given part of their time to prepare a fish collection, and to assemble data pertaining to the team study. It was also intended that. the George Vanderbilt Pacific Equa- torial Expedition, which was to have coli@cted fishes in the Tuamotu Ishands. at the same time the Coral Atoll Project was being carried out at Raroi t, would come to Raroia and aid the atoll team as much as possible. When a was learned that the Vanderbilt Expedition was postponed, it was suggested to Mr. Vanderbilt that one of the ichthyologists, who would have gone on his expedition, should be sent on the Coral Atoll Project. Mr. Vanderbilt whole- heartedly agreed and generously provided the funds for the author to carry out the ichthyological survey. The aims of the ichthyological research at Raroia were dual: (1) to contribute to the Raroia team survey by studying the ecology and economic relationships of the fishes; and (2) to contribute to the Vanderbilt program in Polynesia by making as extensive observations and collections of the fishes as possible. Unfortunately neither program could be carried out as successfully as anticipated. Practically none of the ichtzvotogical equip- ment was received because of the Pacific coast shipping ctrike. Some of the most important gear was repurchased in Honolulu, but it did not reach Raroia until the last two weeks available for collecting. As a result, the team was heavily dependent on the cooperation of the natives for transportation, assistance and fishing equipment. At various times sailing canoes, outrigger canoes, outboard motor boats, an outboard motor, and a small cutter were made available by the natives for ae -trensportation. Generally, three native assistants helped with the handling of the boats and the collecting. Dr. Newell made available his shallow water diving equipment for four stations in 20 to 40 feet of water. Many types of hooks and lines were tried and the natives were most helpful in providing this type of fishing gear. About three-fourths of the time at Raroia was spent in the field by the fish crew, and the remainder was used in preserving and packing the fishes and taking notes. List of Basic Fish Equipment The gear that was actually available for fish collecting is listed be- low. The items received during the last weeks of the survey are preceded by asterisks. The equipment that was strike-bound on the Matson ship "Sierra" is nct listed. Transportation: *5 horse outboard motor; anchors. Spear fishing gear: %*Champion rubber sling gun; “Hawaiian slings (2); *spear shafts with simple hinged barbs (6); simple spear loaned by native. Swimming gear: Depth gauges (2); Swim fins (2 pr.); Snorkels (1); Tabe shoes (2 pr.); Squale face masks (2); *Diving underwear (1 pr.). Ichthyocide gear: Rotenone-powdered (app. 385 1bs.); Cloth sacks (6); Inner tube floats (2); Galvanized buckets (3); Dipnets--small (6); Dipnets—— large (6); Shark repellent (20 pkgs.). Fishing Tackle: Navy survival Pishane kit; Assorted nylon lines; As- sorted hooks loaned by natives. | Preservation: Fommel dehyde 9) gals. pt Washtubs (3); Porcelain and - plastic trays (3); Museum jars 8 oz. (24); Screw top vials (50); Canning machine and 50 cans; cone soft string; Sodium borate (2 qts.); Cheese cloth 50 yard bolts (4) 5 Burlap sacking 3 x 100" 5 Numbered tin tags. Steel Drums: 54 gal. (5); %*35 gal. (4). Records: Collection mel, tary labels (400); Field data sheets (100); Record Books (3); Higgins Eternal ink and pens, Fish Books: Tinker: Hawaiian fishes (1944); Schultz: Fishes of the Phoenix and Samoan Islands (1943); Fowler: Fishes of Oceania (1928). Photographic Equipment: 35 mm. Kodak Signet, accessories, and film; 4.x 5 Speed Graphic, accessories, and film. STATION RECORDS During the two month survey at Raroia from June to August, approximate- | ly 60 different fish stations were studied. The most interesting collections and observations were made (1) in surge channels and on the outer reef flat on the leeward (west) side of the atoll; (2) on the windward lagoon shore | reefs; (3) in inter-islet channels and the ship (Garue) Pass; and (4) on the : southeastern and southwestern lagoon coral heads. Approximately, 9,000 fishes of 400 species were collected at Raroia. All of them were preserved in 10% formalin and packed in five 54 gallon and four 35 gallon steel drums, and in 50 small tin cans and jars. At each sta- |. tion a United States National Museum Fish Division ecological data sheet was | filled out and additional notes were taken when possible. A section of this data is presented below with each station outline. Notes on the fishes are | not given, but can be traced back from stations listed for each species in the Systematic account. Fig. 1 shows the location of each field station on a / chart of Raroia kindly supplied by Dr. Norman D. Newell. The locality names are clarified in the report of Dr. Newell. At the suggestion of Bengt Daniel- sson all native "NG" sounds are spelled with only a "G" (e.g. the word pro- nounced "Nengo Nengo" is spelled "Nego nego"). During the return from Raroia to Tahiti, the natives at Takume pre- sented me with a small local collection of reef fishes. This material was not examined in the field and is not incorporated in the present report. The same is true of several collections of market fishes obtained in the Papeete market, Tahiti, in early September. Station 1.--Tahiti I., Society Islands; Shore reefs at Punaavia in Taapuna's pass area; 94 kilometers west of Papeete, June 18. Vegetation: encrust- ing algae on live and dead coral. Time: 3-4:00 P.M. Bottom: sand and coral. Shore: coral sand. Current: app. 2 knots. Distance from | shore: 10'. Tide: between high and low. Depth of capture: 8'. Depth of water: 8'. Method of capture: Tarzan spear gun. Collected by: John Vernaudon and R, Harry. One specimen of one species. Station 2.--Tahiti; 6 kilometers west of Papeete; coral reefs bordering "Les Tropigques" Hotel, next to lagoon, June 20. Vegetation: encrusting al- gae. Time: 10-11 P.M. Bottom: coral and coral rubble. Shore: beach- rock. Distance from snore: O-100 yards. Tide: high. Depth of capture: O-2'. Depth of water: O-2'. Method of capture: dipnets. Collected by: John Byrne and R. Harry. 45 specimens of 10 species. Station 3.--Tahiti; shore reefs at Punaaiua in Taapuna's pass area, 93 kilometers west of Papeete, June 20. Vegetation: encrusting algae on live and dead coral. Time: 9-12 A.il, Bottom: coral sand, live and dead coral. Shore: coral sand. Current: app. 2 knots. Distance from shore: app. 300 yards. Tide: between low and high. Depth of capture: 0-6'. Depth of water: 0-6'. Method of capture: rotenone (app. 8 lbs.) and dipnets. Collected by: John Vernaudon and R. Harry. 250 speci- mens of 47 species. PQ) Station 4.--Tahiti; He eran reefs of Motu Uta Islet in Papeete Harbor, June 21. Vegetation: encrusting algae on live and dead coral. Time: 8:30-11 A.M. Bottom: coral sand, live and dead coral. Shore: coral. Current: very slight. Distance from shore: BPR 50 UREGR Tide: be- tween low and high. Depth of capture: 0-15'... Depth of water: 0-15'. Method of capture: rotenone (app. 8 lbs.) ana dipnets. . Collected by: J. Vernaudon and R. Harry. App. 125 specimens of 25 ‘species. Station 5,--Tahitis, shore reefs at Punaavia in Taspunat s Pass area. 94 kilometers west of Papeete, June 22. Vegetation: encrusting algae on live and dead coral. Times '7:30-11 A.M. Bottom: coral sand and coral. Shore: coral sand. Current: app. 1 knot. Distance from shore: app. 100 yards. Tide: low. Depth of capture: O-1Z'. Depth of water: O-12'. Method of capture: rotenone (app. 8 lbs.) and dipnets. Collected by: J. and O. Vernaudon, R. Harry. App. 150 specimens of 30 species. Station 6.--Tahiti; purchased in Papeete Central Market, June 22. Time: 6 A.M. Obtained by R. Harry. 50 specimens of 15 species. Station 7.--Raroia Island, Tuamotu Archipelago (app. 169 1' S. 142° 26' W., H. 0. Chart 2004); coral head off Garumaoa Village jetty, west side of lagoon, June 30. Vegetation: encrusting algae. Time: 2-5 P.M. Bot- tom: coral sand and 100' wide flat—topped coral head rising straight from flat sand bottom 20-25'. Shore: coral sand. Current: practically none. Distance from shore: app. 250 yards. Tide: high. Depth of cap- ture: 0-20'. Depth of water: 0-20'. Method of capture: rotenone (app. 15 lbs.) and dipnets. Collected by M. Doty, B. Danielsson, Tetohu, R. Harry. App. 300 specimens of 60 species. Station 8.--Raroia; Channel between Kumekume and Garumaoa (Korere) Islets, near Garumaoa Village, July 1. Vegetation: encrusting algae on dead coral and beach rock. Time: 9-11:40 A.M. Bottom: Coral sand, gravel, Beach rock, dead coral. Shore: beach rock, dead coral. Current: o-+ knot. Distance from shore: 0-25'. Tide: between low and high. teeth of capture: 0-23'. Depth of water: 0-24!. Method of capture: rotenone (app. 8 lbs.) and dipnets. Collected by: M. Doty, W. Newhouse, R. Harry. App. 450 specimens of 35 species. Station 9.--Raroia; coral shelf near jetty next to lagoon shore between Ohue coral head and Garumaoa Village, July 1. Vegetation:.encrusting algae on live and dead coral. Time: 2:15-5:15 P.M. Bottom: . coral sand and dead coral. Shore: coral sand. Current: slight toward: shore. Dis-~ tance from shore: app. 100 yards. Tide: between high and low. Depth of capture: 0-20'. Depth of water: 0-20!. Method of capture: rotenone (app. 12 lbs.) and dipnets. Collected by: M. Doty, W. Newhouse, R. Harry. App. 450 specimens of 60 species. Station 10.—-Raroia; outer reef near Garumaoa village on west side of atoll on shallow tidel flat next to shore, July 3. Vegetation: sparse coralline algae. Time: 3-5:40 P.M. Bottom: coral sand, live and dead lS coral. Shore: beach rock. Current: app. $ knot along shore. Distance from shore: 0-100 yards. Tide: extreme low. Depth of capture: 0-6". Depth of water: 6". Method of capture: rotenone (4 lbs.) and dipnets. ’ Collected by: M. Doty, W. Newhouse, R. Harry. App. 325 specimens of 35 species. Aug. 25. Collected by: J. Morrison. 2 specimens of 2 species. Station 11.--Raroia; channel between Takeke and Temari Islets, near Garve _ North Passage, west side of atoll, July 4. Vegetation: encrusting algae. Time: 8-11:15 A.M. Bottom: coral sand, dead coral, some live coral. Shore: dead coral and beach rock. Current: app. 1 knot.. Distance from shore: 0-100'. Tide: just after low tide. Method of capture: rotenone (6 lbs.) and dipnets. Collected by: R. Harry. App. 220 specimens of 35 species. Station 12.-—-Raroia; in lagoon off pier at Garumaoa Village on top of reefs, west side of atoll, July 5. Vegetation: encrusting algae. Time: 8:30- 9230 A.M. Bottom: coral sand, dead and live coral. Shore: coral sand. Current: none. Distance from shore: C-100 yards. Depth of capture: O-3'. Depth of water: 3'. Method of capture: rotenone (app. 10 lbs.) and dipnets. Collected by: J. Morrison, Raroian natives, HK. Harry. 40 specimens of 18 species. Station 13.--Raroia; "Chava" coral head in lagoon off Garumaoa Village, west side of atoll, July 5. Vegetation: encrusting algae. Time: 2:30- 4:15 P.M. Bottom: coral sand, live and dead coral. Shore: coral sand. Current: very slight. Distance from shore: about 3/4 mile. Tide: between low and high. Depth of capture: 0-20'. Depth of water: 1-20'. Method of capture: rotenone (7 lbs.) and dipnets. Collected by: M. Doty, N. Newell, natives, R. Harry. 100 specimens of 35 species. Station 14.--Raroia; channel between Geogeo and Kukina Islets, near Garumaoa Village, west side of atoll, July 6. Vegetation: encrusting algae. Time: 7:30-11:30 A.M. Bottom: coral sand, live and dead coral. Shore: gravel, beach rock. Current: none to 1 knot. Distance from shore: O- 100', Tide: extreme low. Depth of capture: 0-14'. Depth of water: 6"-14!, Method of capture: rotenone (app. 6 lbs.) and dipnets. Col- lected by: R. Harry. 200 specimens of 28 species. Station 15.--Raroia; North Pass, near Garumaoa Village on west side of atoll, July 7. Vegetation: encrusting and low green algae. Time: afternoon. Bottom: sand, coral. Shore: beach rock. Current: various, usually up to 5 knots. Distance from shore: 100-400'. Tide: various. Depth of capture: 10-35'. Depth of water: 25-40'. Method of capture: hook and line. Collected by: native for KR. Harry. 2 specimens of 2 species. Station 16.--Raroia; near Garumaoa Village on outer reef flat, July 8, Aug. 20, Sept. 2. Vegetation: encrusting algae, lithothamnion ridge. Bot- tom: coral, beach rock, sand. Shore: beach rock. Current: 0-3 knots. Distance from shore: 0-150 yards. Depth of capture: O-1'. Depth of water: 6"-1', Method of capture: dipnets, spears. Collected by: J. Morrison, K. Harry. 13 specimens of 10 species. S18. Station 17.--Raroia; first transect in open channels on edge of outer reef, near Gerumaoa Village, west side of atoll, July 9. ‘Vegetation: encrust— ing algae, lithothamnion ridge. Time: 9-12 A.M. Bottom: coral, beach- rock, algae. Shore: beachrock. Current: strong surge. Distance from shore: App. 200 yards. Tide: extreme low. Depth of capture: O-15'. Depth of water: 1-15'. Method of capture: rotengne (8 lbs.) and dipnets. Collected by: Tetohu, R. Harry. 500 specimens of 55 species. Station 18,.--Raroia; first transect on outer open reef flat near Garumaoa Village, west side of atoll, July 10. Vegetation: encrusting algae. Time: 2-3 P.M. Bottom: beachrock, coral, algae. Shore: beachrock. Current: surging. Distance from shore: app. 50-100 yards. Tide: be-. tween low and high. Depth of capture: 6". Depth of water: 6". Method of capture: dipnet. Collected by: R. Harry. One specimen, one species. Station 19.--—Raroia; channel and leeward outer reef north of Garue Pass be—- tween Kahogi and Fakatomo Islets, July 11. Vegetation: encrusting algae. Time: 8:30-11:30 A.M. Bottom: algae, coral, beachrock. Shore: beach- °. rock. Current: strong with surge. Distance from shore: 100-200 yards. Tide: between high and low. Depth of capture: 0-5'. Depth of water: 1-5'. Method of capture: singie and three pronged spears. Collected by: natives and R. Harry. 5 specimens of 5 species. . Station 20.--Raroia; first transect on outer open reef flat near Garumaoa Village, west side of atoll, July 11. Vegetation: encrusting algae. Time: 1-3 P.M. Bottom: sand, coral, algae. Shore: beachrock. Current: O-4 knots. Distance from shore: 0-100 yards. Tide: just after low. Depth of capture: O-1'. Depth of water: 1'. Method of capture: rote-" none (app. 6 lbs.) and dipnets. Collected by: R. Harry. 50 specimens _ of 23 species. is he : % Station 21.--Raroia; Tahiti islet,.app. 2 miles south of Garumaoa Village, westside of atoll, July 12. Vegetation: encrusting algae. Time: 8:30- 11 A.M. Bottom: sand, gravel, beachrock. Shore: beachrock. Current: 1-5 knots and surge. Distance from shore: 50-100 yards. Tide: between high and low. Depth of capture: 0-4'. Depth of water: 4'. Method of capture: 1 and 3 pronged spears. Collected by: B. Danielsson, natives, R. Harry. 3 specimens of 3 species. Station 22.--Raroia; Mataira Islet, north of Garve Pass, west side of atoll, July 17. Vegetation: encrusting algae. Time: 6:30 A.M.-11 P.M. ~Bot- toms beachrock, coral sand, coral. Shore: beachrock. Current: 0-4 knots. Distance from shore: 0-200 yards. Depth of capture: O-3'. Depth of water: 3-6'. Method of capture: rotenone (8 lbs.) spears, dipnets. Collected by: natives, R. Harry. 440 specimens of 65 species. Station 23.--Raroia; first transect on outer open reef flat near Garumaoa Village, west side of atoll, July 18. Vegetation: encrusting algae. Time: 9:30-12:30 P.M., 1:30-4:00 P.M. Bottom: coral, coral sand. Shore: gravel, beachrock. Current: O-l knot. Distance from shore: 50-100 ep ig yards. Tide: mid-low to high to mid-low. Depth of capture: 0-25'. Depth of water: 2-25'. Method of capture: rotenone (app. 12 lbs.) and dipnets. Collected by: Tetohu, Peni, R. Harry. 410 specimens of 63 species. Station 24.--Raroia; Tetou Islet, largest islet on east side of atoll, chan- nel on lagoon side, July 19. Vegetation: encrusting algae. Time: 9:30 A.M.-1 P.M. Bottom: sand, coral, Shore: sand, gravel, beachrock. Current: 5-8 knots. Distance from shore: 50-100 yards. Tide: low to high. Depth of capture: 0-10'. Depth of water: 8~-10'. Method of cap- ture: hook and line. Collected by: natives, R. Harry. 5 specimens of 4 species. Station 25.--Raroia; channel between Temari and Kumekume Islets, immediately south of Garue Pass, July 20. Vegetation: encrusting algae. Time: 9-11:30 A.M. Bottom: beachrock, sand, coral. Shore: beachrock. Dis- tance from shore: 0-20'. Current: none. Depth of capture: app. 6". Depth of water: app. 6". Method of capture: by hand. Collected by: J. Morrison for R. Harry. 12 specimens of 3 species. Station 26.--Raroia; first transect at excurrent channel, and ridge channels of the lithothamnion ridge, July 21. Vegetation: encrusting algae, lithothamnion ridge. Time: 9:45 -12A.M. Bottom: beachrock, gravel. Shore: beachrock. Current: surf and turbulent surge. Distance from shore: app. 200 yards. Tide: extreme low. Depth of capture: O0-25'. Depth of water: 5-25'. Method of capture: rotenone (app. 15 lbs.) dip-— nets, spears. Collected by: Peni, Tetohu, R. Harry. 840 specimens of 56 species. . Station 27.--Raroia; outer reef surge channels at lithothamnion ridge at Garumaoa Village, west side of atoll, July 22. Vegetation: encrusting algae. Time: 11 A.M.-1:30 P.M., 8-11 P.M. Bottom: coral. Shore: Beachrock. Current: surf. Distance from shore: 150-200 yards. Tide: low. Depth of capture: 0-3'. Depth of water: 1~3'. Method of capture: dipnets, spears, lanterns. Collected by: Tetohu, Peni, Paulo, R. Harry. 136 specimens of 19 species. Station 28.--Raroia; "Ohava" coral head in lagoon + mile east of Garumaoa Village, west side of atoll, July 23. Vegetation: encrusting algae. Time: $:30-11 A.M., 2:30-4 P.M. Bottom: coral, sand. Shore: sand. Current: app. 1 knot. Distance from shore: $ mile. Tide: between low and high. Depth of capture: 0-50'. Depth of water: 2-50'. Method of capture: rotenone (40 lbs.) and dipnets, spears. Collected by: Tetohu, Peni, Paulo, Kehea, Vaia, R. Harry, Approx. 500 specimens, 65 species. Station 29.--Raroia; sand channel, vicinity of Tetou Islet, east side of atoll, July 23. Vegetation: none. Time: morning. Bottom: sand. Shore: sand. Current: app. 1 knot. Distance from shore: 50 yards. Depth of capture: 4'. Depth of water: 6'. Method of capture: spears, Collected by: expedition party. One Caranx armatus. iy ---Raroia; outer reef flat at Kon-Tiki Islet, east side of atoll, July 23. Vegetation: encrusting algae. Time: morning. Bottom: beach- rock, coral. Shore: beachrock, gravel. Current: app. 2 knots. Dis- tance from shore: app. 75 yards. Depth of capture: i'. Depth of water: 1'. Method of capture: spear. Collected by: expedition party. One EBulamia melanoptera. Station 30.—-Raroia; in lagoon near jetty of Garumaoa Village, west side of atoll, July 23. Time: 4-5 P.M. Bottom: sand, coral. Shore: sand. Current: app. 2 knots. Distance from shore: app. 200'. Tide: high, ' Depth of capture: 4'. Depth of water: 4'. Method of capture: spears. Collected by: natives, ®. Harry. 2 specimens, 2 species.. Station 31.--Raroia; Garve Pass, west side of atoll, July 24. Time: 9 A.M.- 5 P.M. Bottom: coral, gravel, sand. Shore: beachrock. Distance from shore: = mile. Depth of capture: 20-40'. Depth of water: 30-40'. Method of capture: hook and line, spears. Collected by: Paulo, R. Harry. 29 specimens of 10 species. ante Station 32.--Raroia; in lagoon at end of pier, Garumaoa Village, west side of atoll, July 22-24. Vegetation: encrusting algae.. Time: continuous 2 days. Bottom: sand. Shore: sand. Current: none. Distance from shore: App. 200'. Depth of capture: 15'. Depth of water: 15'. Method of capture: plastic screen trap. Collected by: R. Harry. 2 Aulostomus chinensis. ne Station 33.--Raroia; Garue Pass and immediate vicinity, west side of atoll, - July 25, Aug. 1, Aug. 14, Aug. 15. Bottom: coral, gravel, sand. Shore: beachrock, Current: 1-8 knots. Distance from shore: 4-4 mile. Depth of capture: 10-45'. Depth of water: 10-100'. Method of capture: hook and line, spears. Collected by: natives, R, Harry. 100 specimens of 48 species. Station 34.--Raroia; outer reefs of Namunamukona Islet on northwest side of atoll, July 26. Vegetation: encrusting algae. Time: 12-1 P.M. Bot- tom: beachrock. Shore: beachrock. Current: O-10 knots. Distance from shore: 50-100'. Tide: low. Depth of capture: O-2'. Depth of water: 2'. Method of capture: spears and hands. Collected by: Hure, Rago, Ri Harry. 3 specimens of 3 species. Station 35.--Raroia; lagoon and outer reefs of Hava Islet, northeast side of atoll, July 27. Vegetation: encrusting and free growing algae. Time: 10-12 A.M. Bottom: rock, gravel, coral. Shore: gravel. Cur- rent: none. Distance from’ shore: 15-30'. Tide: between low and high. Depth of capture: 0-15'. Depth of water: 5-15'. Method of capture: rotenone (app. 16 lbs.) and dipnets. Collected by: Huri, a Rairai, Kore,’ fh, Harry, 52 ee aie of 14 species. - Station 36.--Raroia; coral head in lagoon channel of Tapoki- ‘Islet, northeast side of atoll, July 28. Vegetation: encrusting algae. Time: 1-2:30 P.M. Bottom: gravel. Shore: gravel. Current: app. 2 knots. Distance a5 from shore: 100'. Tide: low. Depth of capture: 4'. Depth of water: 7',. Method of capture: dipnets. Collected by: Hure, Rairai, Kore, R. Harry. One Ostracion lentiginosun. Station 37.--Raroia; large submerged coral head 15' across in lagoon bay of Tevriamote Islet, north end of atoll, July 28. Vegetation: encrusting algae. Time: 2:30-4:30 P.M. Bottom: gravel, coral. Shore: gravel, sand. Current: very slight. Distance from shore: 0-20'. Tide: high. Depth of capture: O0-18'. Depth of water: 10-18'. Method of capture: rotenone (10 lbs.) and dipnets. Collected by: Huri, Rairai, Kore, R. Harry. 118 specimens of 43 species. Station 38.--Raroia; large enclosed pool 100 x 40' on lagoon sand flat of Tikamiti Islet, north end of atoll, July 28. Vegetation: encrusting algae. Time: 6 A.M.-4:30 P.M. Bottom: thick layer of fine sediment. Shore: gravel, beachrock. Current: none. Distance from shore: O-20'. Depth of capture: 10-15'. Depth of water: 10-15'. Method of capture: rotenone (5 lbs.) and dipnets. Collected by: Kore, Rairai, R. Harry. 82 specimens of 19 species. Station 39.--Reroia; outer reef flat between lithothamnion ridge and shore of Teuriamote Islet, north end of atoll, July 28.. Vegetation: encrust- ing algae. Time: 8:30-11 P.M. Bottom: beachrock, coral. Shore: beachrock, gravel. Current: 2-6 knots. Distance from shore: 50-100 yards. Tide: between low and high. Depth of capture: 1-2'. Depth of water: 1-24'. Method of capture: dipnets and Coleman lanterns. Col- lected by: Kore, Rairai, R. Harry. 30 specimens of 2 species. Station 40.--Raroia; enclosed brackish pond 100' diameter, lagoon side of Tikaheru Islet, north end of atoll, July 29. Time: 9-10:30 A.M. Water: murky, brackish, with marked thermal layering. Bottom: thick layer of fine sediment. Shore: beachrock, gravel. Current: none. Distance from shore: 0-50'. Depth of capture: 10-20'. Depth of water: 20'. Method of capture: spears. Collected by: Hure, 8. Harry. 4 specimens of 1 species. Station 41.--Raroia; "Tomogoru" coral head, 75' across its flat top 1! below the surface, near Tomogoru Channel, north end of atoll, July 29. Vegetation: encrusting and abundant free growing algae. Time: 10-12 A.M. Bottom: coral, sand. Shore: sand, beachrock. Current: none. Distance from shore: app. 1 mile. Depth of capture: app. 2-4'. Depth of water: 2-25'. Method of capture: dipnets. Collected by: R. Harry. 150 Spratelloides spp. Station 42.--Raroia; Geogeo lagoon reefs at inner edge, south of Garumaoa Village, west side of atoll, July 31. Vegetation: encrusting and free growing algae. Time: 3-4 P.M. Bottom: sand coral. Shore: sand. Cur- rent: very slight. Distance from shore: app. 3/4 mile. Depth of cap- ture: 5-25'. Depth of water: 25'. Method of capture: rotenone (21 lbs.) dipnets, diving gear. Collected by: Rago, Rairai, Kore, J. Byrne, R. Harry. 140 specimens of 42 species. 1-16- . Station 43.—-Raroia; "Motoko" coral head in lagoon 14 miles east of Tetataga Islet, west side of atoll, Avg. 2. Vegetation: dense free growing al- gal growth. Time: 12-3:30 P.M. Bottom: coral, sand. Current: app. 1 knot. Distance from shore: app. 13 miles. Tide: high. Depth of cap- ture: 12-30'. Depth of water: 2-40'. Method of capture: hook and line, spears. Collected by: Hure, Rago, R. Harry. 11 specimens of 3 species. Station 44.--Raroia; outer reef flat near Oneroa Village, southwest side of atoll, Aug. 2. Vegetation: encrusting algae. Time: 9:30-11:45 A.M. Bottom: rock, gravel, coral. Shore: beachrock. Current: surge, strong. Distance from shore: 10-20'. Tide: high. Depth of capture: 0-2'. Depth of water: 2'. Method of capture: spears. Collected by: Rago, R. Harry. 6 specimens, 6 species. Station 45.--Raroia; outer reef flat and surge.channels near Oneroa Village, southwest side of atoll, Aug. 4-8. Vegetation: encrusting and some free growing algae. Time: day and night, app. 6 hours per day. Bottom: coral, beachrock. Shore: beachrock. Current: O-strong surge. Distance from shore: app. 100 yards. Tide: low. Depth of capture: 10-30'. Depth of water: 10-30'. Method of capture: rotenone (app. 75 lbs.), dipnets, spears, diving gear. Collected by: expedition personnel. App. 1000 specimens of 75 species. Station /46.--Raroia; large coral head app. 2’ kilometers from the east and south sides of the atoll, Aug. 5. Vegetation: abundant encrusting and great clumps of free growing algae. Time: 9-10 A.M. Bottom: coral. Shore: sand, coral. Current: app. 3 knot. Distance from shore: 2 kilometers. Tide: high. Depth of capture: 1-15'. Depth of water: 2-30'. Method of capture: spears. Collected by: Huri, Rago, R. Harry. 6 specimens of 3 species. | vu ' Station 47.--Raroia; gravel and sand channels at extreme southern end of atoll, Aug. 5. Vegetation: encrusting algae. Time: 11 A.M.-3 P.M. Bottom: coral, sand, gravel, beachrock. Shore: gravel. Current: 0-8 knots. Distance from shore: 1-4 kilometers. Depth of capture: 5-10'. Depth of water: 10-15'. Method of capture: spears. Collected by: Huri, Rago, Kore, R. Harry. 39 specimens of -17 species. Station 48.--Raroia; shore reefs in lagoon on submerged coral heads, in region of Rotava Korereka Islet, north end of atoll, Aug. 5-7. Vege- tation: practically none. Time: day and night, app. 15 hours. Bot- tom: sand, coral. Shore: sand. Current: none. Distance from shore: 50-100 yards. Depth of capture: 5-20'. Depth of water: 10-20'. Method of capture: rotenone (app. 20 lbs.) and dipnets. Collected by: Kore, Rairai, Huri, ®. Harry. 160 specimens, 30 species. Station 49.--Raroia; "Vera Vera" coral head in lagoon app. 3 kilometers east of Oneroa Village, south west side of atoll, Aug. 8. Vegetation: abundant encrusting and free growing algae. Time: 9:30-11:30 A.M. Bottom: coral, sand. Shore: sand, gravel. Current: none. Distance Aiea from shore: app. 3 kilometers. Tide: low. Depth of capture: 5-10'. Depth of water: 2-40'. Method of capture: rotenone (app. 15 lbs.) and dipnets. Collected by: Kore, Rago, R. Harry. 68 specimens of 23 species. Station 50.--Raroia; lagoon reefs at Oneroa Village, southwest side of atoll, Aug. 9. Vegetation: encrusting algae. Time: 9:30-10:30 A.M. Bottom: coral. Shore: sand. Current: app. $ !mot. Distance from shore: app. + mile. Tide: low. Depth of capture: 0-4'. Depth of water: 3-4'. Method of capture: rotenone (app. 10 lbs.) and dipnets. Collected by: Rairai, Kore, R. Harry. 17 specimens of 7 species. Station 51.--Raroia; lagoon reefs, Teputaiti Islet, south of Garumaoa Vil- lage, west of atoll, Aug. 9. Vegetation: encrusting algae. Time: 11:30-1:20. Bottom: coral, sand. Shore: sand. Current: none. Dis-— tance from shore: 0-100 yards. Depth of capture: 6-10'. Method of capture: spears. Collected by: Etienne for R. Harry. 2 specimens of 2 species. Station 52.--Raroia; outer reef channels of Kahogi Islet, north of Garue Pass, west side of atoll, Aug. 10. Vegetation: encrusting algae. Time: 3:30-5 P.M. Bottom: beachrock, coral. Shore: beachrock. Cur- rent: 1-4 knots. Distance from shore: 50-100 yards. Depth of capture: 5-15'. Depth of water: 10-18'. Method of capture: spears, hook and line. Collected by: Rairai, Kore, R. Harry. 13 specimens of 10 spe- cies. Station 53.--Raroia; vicinity of "Oreti" coral head, app. 3/4 mile northeast of Garumaoa Village, west side of atoll, Aug. 12. Time: morning. Bot- tom: sand, coral. Shore: sand, gravel. Current: none. Distance from shore: app. 3/4 mile. Depth of capture: app. 40'. Depth of water: app. 40'. Method of capture: deeper water dredging in lagoon. Collected by: J. Newhouse for Rk, Harry. One specimen of Eviota sp. 2. Station 54.--Reroia; shore reefs in lagoon at Takeke Islet next to Garue Pass, west side of atoll, Aug. 12. Vegetation: encrusting algae. Time: 8-12 A.M., 1:30-5 P.M. Bottom: coral, sand. Shore: gravel. Current: none. Distance from shore: 5-100'. Depth of capture: 2-40'. Depth of water: 4-40'. Method of capture: spears, dipnets. Collected by: Rairai, Kore, Kehea, Vaia, 2. Harry. 75 specimens of 6 species. Station 55.--Raroia; lagoon and outer shore reefs at Tenukuhaupapatea (Fakatomo) Islet, immediately north of Garue Pass, west side of atoll, Aug. 12. Vegetation: encrusting algae. Time: 9:30 A.M.-—3:30 P.M. Bottom: coral, beachrock. Shore: beachrock, gravel. Current: 0-3 knots. Distance from shore: 10-100 yards. Depth of capture: 3-10'. Depth of water: 3-15'. Method of capture: spears, dipnets. Collected by: Rairai and Kore for R. Harry. 18 specimens of 10 species. isan Station 56.--Raroia; Garve Pass, west side of atoll, Aug. 13. Vegetation: abundant encrusting and free growing algae. Time: 5:30-7:30 P.M. Bottom: coral, sand. Shore: beachrock, Current: 2-10 knots. Distance from shore: 300 yards. Depth of capture: 10-20'.. Depth of water: 30-40'. Method of capture: hook and line, trolling. Collected by: Kore for, R.. Harry. 2 SP HeaAmeHe of 2 species. ‘Station 57.--Raroia; lagoon vent petals of Kumekume Islet south of Garue Pass, west side of atoll, Aug. 13. Vegetation: encrusting algae. Time: 9:30 A.M.—3 P.M. Bottom: coral, sand. Shore: beachrock, ae Cur-— rent: slight. Distance from shore: 50-100 yards. Depth. of capture: 2-6', Depth of water: 2-6'. Method of capture: rotenone (app. 10 lbs.). Collected by Kore and Rairai for R. Harry. 70 specimens of 35 species. Station 58.--Raroia; outer reef channels next to lithothamnion ridge next to Garumaoa Village, west side of atoll, Aug. 15. Vegetation: encrust- ing algae. Time: 9:20-11:45 A.M., 1:30-6 P.M. Bottom: beachrock, gravel. Shore: beachrock. Current: surge. Distance from shore: 100 yards. Depth of capture: 0-20!. Depth of water: 4-20'. Method of capture: rotenone (20 lbs.) and dipnets. Collected by: Kore, Rairai, _Tetohu, R. Harry. 600 specimens of 75 species. Station 59.--Raroia; first transect, outer reef surge channels at litho- thamnion ridge, immediately south of excurrent channel, Aug. 17. Vegetation: encrusting algae. Time: 9-11:30 A.M. Bottom: beachrock, gravel. Shore: beachrock. Current: surge, turbulent breakers. Dis- tance from shore: app. 150 yards. Tide: low. Depth of capture: O-20'. Depth of water: 4-20'. Method of capture: rotenone (22 lbs.) and dipnets. Collected by: Tetohu, Rairai, Bone Re Hee 350 specimens of 40 species. Station 60.--Raroia; lagoon reefs and channels of Teremu Islet, east side of atoll, Aug. 20-22. Vegetaticn: barren encrusting algae. Time: app. 18 hours. «Bottom: sand, beachrock, coral. Shore: sand, beach- rock. Current: 0-1 knot. Distance from shore: 0-3 mile. Depth of capture: 3-35'.. Depth of water: 0-35'. Method of capture: hook and line. Collected by: Rairai, Kore, Huri, R. Harry. 91 specimens of 32 species. Pah Station 61.--Raroia; coral head 4 mile off Teremu in lagoon, east side of atoll, Aug. 21. Vegetation: barren encrusting algae. Time: 1-4 P.M. Bottom: sand, coral, staghorn prominent. Shore: sand. Current: none. Distance from shore: # mile. Tide: low. Depth of capture: 0-25!'. Depth of water: 2-25'. Method of capture: rotenone (30 lbs.) and dip- nets. Collected by: Kore, Rairai, Huri, R. Harry. 448 specimens of 54 species. Station 62.--Raroia; app. % mile off outer west coast of atoll between Garue Pass and Garumaoa Village, Aug. 25, 27. Time: 8:30 A.M.-3:30 P.M. each day. Bottom: coral. Shore: beachrock. Current: slight. gis Distance from shore: app. ¢ mile. Depth of capture: 2-40'. Depth of water: 40-100'. Method of capture: hook and lines, trolling, hand- lines. Collected by: Kore, Rairai, for R. Harry. 24 specimens of 11 species. Station 63.--Raroia; inner lagoon reef next to Kahuruna Islet, ‘southeast Side of atoll, Aug. 26. Vegetation: sparse encrusting algae. Time: 3:30 P.M. Bottom: sand, coral. Shore: gravel. Distance from shore: 100'. Depth of canture:.8'. Depth of water: 10'. Method of capture: spear. Collected by: Kehea for R. Harry. One specimen of Lutjanus bohar. Station 64.--Raroia; lagoon shore reef of Rare Islet, west side of atoll, Aug. 27. Vegetation: encrusting algae. Time: 11 A.M. Bottom: coral, Shore: gravel. Current: none. Distance from shore: 50'. Depth of capture: 10'. Depth of water: 12'. Method of capture: spear. Collected by: Huri for R. Harry. One specimen of Siganus sp. Station 65.--Raroia; channels at south end of atoll, Aug. 28. Vegetation: sparse encrusting aigae. Time: 1-3:30 P.M. Bottom: sand, coral. Shore: sand, gravel. Current: app. 1-2 knots. Distance from shore: app. 100 yards. Depth of capture: 0-10'. Depth of water: 10'. Method of capture: rotenone (app. 10 lbs.) and dipnets. Collected by: Kore for R. Harry. 150 specimens, 45 species. Station 66.~-Raroia; reefs in Garve Pass next to Takeke Islet, west side of atoll, Sept. 1. Vegetation: considerable encrusting and free growing algae. Time: 10 A.M.-1:30 P.M. Bottom: coral. Shore: gravel. Cur- rent: slight. Distance from shore: app. 100 yards. Tide: low. Depth of capture: 1-25'. Depth of water: 1-25'. Method of capture: rotenone (40 lbs.), dipnets and spears. Collected by: Kore, Rairai, Huri, Rago, and R. Harry. 340 specimens of 50 species. Station 67.--Raroia; lagoon reef at Teputaiti Islet, south of Garumaoa Vil- lage, west side of atoll, Aug. 25. Vegetation: encrusting algae. Bot- tom: coral, sand. Shore: gravel. Currents: none. Distance from shore: 50 yards. Depth of capture: 12'. Depth of water: 15'. Method of capture: spear. Collected by: Huri for R. Harry. One specimen of Naso annulatus. LaOe ECOLOGICAL SURVEY Figures 3-7 While the general report of the Raroia team will consider the marine ecology of this atoll on a broad biological basis, specific notes on the zonation of fishes are included in this section. A large proportion of the field time was involved in preparing the collection of fishes, taking koda- chrome slides, and compiling notes on the colors of living fish. To a lesser” extent habitat data were also recorded at the end of the time in Raroia and | part of these notes are given in the systematic account. The present section was also prepared at the end of the Raroia study and is an attempt to give a general view of fish distribution on Raroia Atoll. Far more notes were taken than included here. So mich was observed that was not understood that the confusing observations were deleted. It is hoped that more of the field data can be incorporated into the repart of the next coral atoll project. For convenience the marine environment of the atoll has been divided into eight major zones which are as clearly delimited ecologically as they are physically. These zones can be considerably subdivided on the basis of fish associations and habitats and most of this phase will be considered in the general ecological report. Coralliferous Outer Bench: This region is defined by Cloud (1962). It ex- | tends from the steep outer slope to the surge channel buttresses. The greatest growth of coral at Raroia is in this region and a large percentage of the total fish fauna lives on this shelf. Sharks constantly patrol this area, apparently continually circling the atoll looking for food. Tuna, bar-~ racuda and jacks also scout these waters, but normally keep nearer the sur- face. Larger sea basses are in greater abundance in this zone (as are also the previous fishes mentioned) than anywhere else om the atoll and seem to hide in every hole in the coral. Naso and Balistes are evident immediately above the coral and various species of the demoiselles that penetrate to deeper water (e.g. Dascyllus) form clouds around the coral. The extent of the fish fauna can best be seen by dropping balls of rotenone mud onto the coral. — About twice as many fishes as would be seen around a lagoon coral head boil out of holes in the coral when rotenone takes effect. Several times I talked myself into chancing encounters with sharks to collect in this region, but the natives forcibly refused to let me go in the water, talking as if it would be sure death. Without a doubt the finest fish collections at Raroia could be taken on this shelf, and very easily on the west side of the atoll, if it were not for the sharks. While this zone seemed to be relatively the same on the east and west side of the atoll, it seemed definitely shorter*on the east side. Practically all observations were made on the west side. + This region has a very characteristic fauna, typified by several sharks and carangids not observed elsewhere around the atoll. At the surface are particularly Sphyraena, thunnids, and schools of Hemirhamphus. At mid-—depths are small schools of carangids. Immediately above the coral sharks wander about, passing from one surge channel to the next; none larger than 12 feet were seen and they would average about 5-6 feet in length. Remaining rela- mes tively motionless above the coral are mostly Balistes, Naso and often Acan- thurus. Only very small fishes, such as Dascyllus, Labroides and Halichoeres, would be seen about the coral. When this region was poisoned great numbers of small coral reef fishes less than five inches long (such as blennies, gobies, a few cirrhitids and a large number of wrasses) came out of hiding, along with large sea basses one to three feet long. Conspicuous by their complete absence were chaetodontids, and holocentrids. Parrot fishes were also uncommon in this region. ; Surge Channels: This term is defined by Cloud (1952), and comprises the next region inward toward the shore between the coralliferous outer bench and the coralline ridge. This zone consists of deeply grooved troughs that have wide outer mouths approximately 25-35' deep and narrow dead ends at the ridge. It is continuous around the atoll and is complexly developed on the southeastern side in the region of Oneroa. The entire lengths of these troughs were col- lected at several localities on the west side of the atoll. Since it was dangerous collecting in the open mouths because of the sharks, less is knowm of this outer area. Twice I was rushed by sharks in the outer zone and the Raroians completely refused to dive there. The surge channels contain the most distinctive fish fauna of any zone at Raroia. Many species are completely limited to them and many fishes found just about everywhere else on the atoll do not occur in these surge channels. The zone is completely dominated by cirrhitids and blennies (especially the genus Cirripectes), although none can be seen when casually observing the channels. In fact during the day they look practically barren except for sharks, scarids and carangids schooling in the outer mouths. However, when rotenone is thrown into the troughs, the waves swirl it quickly from one end to the other and the fishes immediately succumb and are swept out of their hiding places from under rocks and from cracks in the buttress walis. The fishes in this region apparently have a high respiratory rate because of the heavy surf action with high oxygen concentration. As a result the entire fish fauna of the troughs completely succumb to rotenone laden waves in a few min- utes. No where else did we obtain such complete collections of a particular habitat. Also, wave action tended to sweep fishes out into the open and into pockets on the bottom, making observations on relative abundance and collect- ing much easier. Sharks were constantly attracted to collecting stations, although they never paid any attention to the fishes, living or dead. They were, however, quite interested in the swimmers. Aside from the blennies and cirrhitids, the narrow end of the channels have a large population of small echidnid eels, most of which are limited to this region and the sea urchin holes on the coralline ridge. Two pomacentrids are confined entirely to the FAKARAVA I. GAMBIER IS. APATAKI I. = Se a = Lycodontis picta (Ah1l) " meleagris (Shaw & Nodder) tal ra tal * u" ruppelliae (McClelland) " flavo marginata (Ruppell) " undulata (Lacépéde) u favaginea (Schneider) Enchelynassa canina (Quoy and Gaimard ) Uropterygius concolor Ruppell " macrocephalus (Bleeker) " marmoratus (Lacépéde) Synodus japonicus (Houttuyn) ae | x Saurida gracilis (Quoy & Gaimard) : Fe x | x x Belone platyura Bennett x | x Strongylura indica (Le Sueur) x Hyporhamphus acutus (Gunther) x x Rhynchorhamphus georgii (Valenciennes ) | Hemirhamphus erythrorinchus | x ‘Le Sueur | Cypselurus atrisignis Jenkins n angusticepts Nichols x and Breder ~30 ie — a Rava MARINEA I, TAKAROA I. VAUTTAHT 1. Tikit 1. * fi ne "TUAMOTUS" bry 2 : EMO) je ‘MANGAR RANGIROA I. AKTAKT I. ANAA I. | GAMBTER IS. Cypselurus simus (Valenciennes) Bregmaceros mcclellandii Thompson Bothus mancus (Broussonet ) " _ gonstellatus (Jordan & Goss) " pantherinus (Ruppel1) 2 1 Malacosarcus macrostoma (Gunther) Anomalops katoptron (Bleeker) Holocentrus caudimaculatus Ruppell " ruber (Forsk31) n erythraeus Gunther " lacteo-guttatus Cuvier microstomus Gunther Me spinifer (Forskal) " sammara (Forskal) — " opercularis Valen- ciennes Myripristis murdjan (Forskal) Wu undecimalis Herre ne violaceus Bleeker | | " | diadema Lacépéde x x rit a microphthalmus Bleeker ae VALRAATEA I. Myripristis pralinius Cuvier Corythoichthys conspicillatus (Jenyns) Aulostomus chinensis (Linnaeus) Fistularia petimba Lacépede Atherina temminckii Bleeker " vaigiensis (Quoy & Camara) Mugil tade Forskal és longimanus Gunther va vaigiensis Quoy & Gaimard ° macrolepis Andrew Smith ° crenilabis Forskal Myxus leuciscus Gunther Neomyxus chaptalii (Eydoux and Souleyet ) Sphyraena barracuda (Walbaum) “ obtusata Cuvier Parathunnus sibi_ (Schlegel) Acanthocybium solandri (Cuvier) Gempylus serpens Cuvier Coryphaena hippurus Linnaeus Nomeus gronovii (Gmelin) AKIAKT I. ANAA Pe iA Dsy APATAKT 1. tal : : fy ———— PINAKT I. RANGIROA I. cane Aaa EY er ran) oi eS Oe ee "TUAMOTUS" oo reer soy i VAHITAHI I. x VAIRAATEA I. ANAA I. Scomberoides toloo-parah (Ruppell) " sancti-petri (Cuvier) Decapterus pinnulatus (Hydoux & Souleyet ) - Selar crumenophthalmus (Bloch) Caranx stellatus Eydoux & Souleyet | Carangoides ferdau (Forskal) t " gymonstethoides Bleeker Trachinotus baillonii (Lacépede) Amia bandanensis (Bleeker) " aroubiensis (Hombron and Jacquinot ) " fraenata (Valenciennes ) ‘" hypselonotus (Bleeker) ic ; deythibtind (Snyder) Apogonichthys auritus (Valenciennes) Ghatledipharas lineatus (Linnaeus) - quinquelineatus Cuvier Dules rupestris (Lacépede) it marginatus Cuvier . sandvicensis (Steindachner) ~33— ae) Lo | on = a 3 nn ER HITI I. MANGAREVA I. PINAKI I. rad ra ra tad tal a RANGIROA I. ba TAKAROA I. . TIKEI I. > "TUAMOTUS" us KH a eH i a > Plectropomus maculatus. (Bloch) Variola louti (Forsk&l) Cephalopholis argus Schneider n urodelus (Schneider) " sexmaculatus (RUppell) Serranus fasciatus (Forskal) Ms gilberti Richardson “s corallicola Valenciennes * merra (Bloch) " socialis Gunther n summana (Forsk31) fusco-guttatus (Forskal) Grammistes sexlineatus (Thunberg) Priacanthus hamrur (Forsk&l) Lutjanus kasmira (Forskal) " bohar (Forskal) vaigiensis (Quoy & Gaimard) i gibbus (Forskal) ! Pterocaesio tile (Cuvier) Lethrinus mahsena (Forsk31) my es ‘APATAKI.I. ————EEE | LT | monostigma (Cuvier) x FAKARAVA 1. GAMBIER iS. HITLI 1. e S = Hi g a < 3 4 ale o1s Ar 3 ee ml a E 4 213 kK | ac 3\3 ale "TUAMOTUS" VAHITAHI i. VAIRAATEA I. Lethrinus reticulatus Valenciennes am miniatus (Schneider ) Pentapodus aureo-lineatus (Lacépede ) Monotaxis grandoculis (Forskal) ‘Mulloidichthys auriflamma (Forsk&l) " samoensis (Gunther) Pseudupeneus trifasciatus (Lacépede ) " barberinus (Lacépede) " crassilabris Valenciennes Paracirrhites hemistictus (Gunther) " polystictus (Gunther) Chaetodon ornatissimus Cuvier “i pelewensis Kner at | bennetti Cuvier ? trifasciatus Mungo Park My ephippium Cuvier citrinellus Cuvier " lunula (Lacépede ) a reticulatus Cuvier quadrimaculatus Gray auriga Forsk&l “AKIAKI Le ANAA I. ~35- APATAKI I. FAKARAVA I. GAMBIER IS. S. MARUTEA I. pai Ls alg! mils e La Ale Z\s S| a H 3 = ie) Zz fs IK 3 S =x cS "TUAMOTUS" VAHITAHI I. VAIRAATHA L. Chaetodon mertensii Cuvier " falcula Bloch . lineolatus Cuvier n trifascialis Quoy & Gaimard Heniochus permutatus Cuvier " monocerus Cuvier Forcipiger longirostris (Broussonet ) Pygoplites diacanthus (Boddaert) Pomacentrus imperator (Bloch) Centropyge flavissima (Cuvier) Zanclus cornutus (Linnaeus) " canescens (Linnaeus) Hepatus triostegus (Linnaeus) " guttatus (Schneider) " fuliginosus (Lesson) " elongatus (Lacépede ) lineatus (Linnaeus) " bleekeri (Gunther) olivaceus (Schneider) nigricans (Linnaeus) RS e—eeeee——————eeeeeeEee—e—eEeEeEeEeEeEee————— EEE AKIAKI I. ANAA I. APATAKI I. FAKARAVA Le GAMBIER IS. Tig HIDE te MAKATEA Ie MAKEMO I. HANGAREVA ie RANGIROA I. TAKAROA I. ts =) 5 3 E ° tf KH a —4 tH = 1 : : 3 we H C+ ® 4 le Ale S| a e She < a4 o BIS ale 4 Q = = S FAKARAVA I. MAKATEA I. GAMBIER IS. | HIT LL. AKIAKI 1. ANAA I. MS APATAKT I. Hepatus glauco~pareius (Cuvier) " achilles (Shaw) bal tad ” Ctenochaetus flavicauda Fowler | x " _ gtrigosus (Bennett) . x Zebrasoma veliferum (Bloch) | | x x Laephichthys rostratus (Gunther ) Naso lituratus (Schneider ) x Fal ral "' eoume (Lesson) | " previrostris (Valenciennes) x|x Siganus punctatus (Schneider) x " rostratus (Valenciennes) x |x Scorpaenopsis gibbows (Schneider) — Sebastapistes byuocusis (Richardson) x Nemapterois biocellatus Fowler x Pterois antennata (Bloch) x " radiata (Cuvier) xix x x Synanceja verrucosa Schneider | x Caracanthus maculatus (Gray) e Wie Sie: " unipinna (Gray) : x Dactyloptena orientalis (Cuvier) Dascyllus trimaculatus (RUppel1 ) : | x{ 4 37 "TUAMOTUS" VAHITAHI f. VAIRAATEA I. Dascyllus aruanus (Linnaeus) Chromis caeruleus (Cuvier) Pomacentrus ahh (Bloch) " nigricans (Lacépede ) Abudefduf saxatilis (Linnaeus) " coelestinus (Cuvier) " sordidus (Forskal) uv septemfasciatus (Cuvier) " luecozonus (Bleeker) " biocellatus (Quoy & Gaimard) " glaucus (Cuvier) " leucopomus (Cuvier) " amabilis (De Vis) Lepidaplois axillaris (Bennett) " hirsutus (Lacépede) Epibulus insidiator (Pallas) Anampses geographicus Valenciennes T diadematus Rilppell Stethojulis strigiventer (Bennett) " axillaris (Quoy & Gaimard) Halichoeres centriquadrus (Lacépede) — Hi ‘4 ES a < Ss a S re a q = oo = S 3 MAKATEA I. MANGAREVA I. PINAKI I. RANGIROA Ie. TAKAROA I. - tad e = j|H wa 2 iH Sale QO js a EY (rs = |g = j= VAIRAATEA I. Halichoeres sdtendeulatnd | (Quoy and ® i Gaimard ) ~Coris gaimard (Quoy and Gaimard) Thalassoma quinguevittatum (Lay and Bennett ) ot - hardwicke (J. W. Bennett ) eee (RUppel1 ) purpureum (Forsk&l) " trilobata (Lacépéde) _ lutescens (Lay & Bennett) | Gomphosus tricolor (Quay & Gaimard) Pseudocheilinus hexataenia (Bleeker) ' Cheilinus trilobatus Lacépede " ‘undulatus RUppell " diagraminus (Lacépede ) oe chlorourus (Bloch) Geliyodon perspicillatus (Stein- dachner ) n sordidus (Forsk&l) | ea microcheilos (Bleeker) agit microrhinos (Bleeker) " *-‘brevifilis (Gunther) ee; ° A ee; - eit} ri] = ti KH ej/—aj es tH rep ra HiSiHig =) eae | ies | O| = el fx ie) tH {ato oe} {ali Slat ity] a a Sisiie A/S jis|eis ra ee 2|2|4|a|= = eS eo) S(Slaiala | = x x x x x x x x x x x pia x x re x x x x] xX x x x x x x x x x x x x x Bh Callyodon harid (Forsk$1) " pulchellus (Ruppell) " fasciatus (Valenciennes) " janthochir (Bleeker) oviceps (Valenciennes) " blochii (Valenciennes) x " rubroviolaceus (Bleeker) " macrocheilos. (Bleeker) nuchipunctatus (Valen- ciennes ) Asterropterix semipunctatus (Ruppell ) Eviota afelei Jordan & Seale - viridis (Waite) Ptereleotris microlepis (Bleeker) x Diaphoroculius rangiroae Fowler Gobiodon rivulatus (Ruppell) - Paragobiodon echinocephalus (RUppell ) Glossogobius biocellatus (Valenciennes) Gobius fuscus Ruppell Echeneis remora Linnaeus Leptecheneis naucrates (Linnaeus) «0 FAKARAVA I. GAMBIER IS. MAKATEA I. MANGAREVA I. PINAKT I. RANGIROA I. ne a TAKAROA Ie TIKI I. "TUAMOTUS" VAHITAHI I. VAIRAATHA I. Callionymus cookii Giinther Petroscirtes filamentosus (Valenciennes ) " taeniatus (Quoy and Gaimard) Blennius tonganus Jordan & Seale Salarias edentulus (Schneider) " periophthalms (Valenciennes ) " biseriatus Valenciennes " “meleagris Valenciennes n caudolineatus Gunther mM lineatus Valenciennes Enchelyurus ater Gunther Brotula multibarbata Schlegel Balistes vidua Richardson " | capistratus Shaw n bursa Schneider " flavimarginatus RUppell " aculeatus (Linnaeus) Balistapus rectangulus (Schneider) n undulatus (Mungo Park) Melichthys buniva (Lacépede) te e << be | << | APATAKI I. yale FAKARAVA I, ” A tH (am) 2 a 3 MAKATEA I. MANGAREVA I. | PINAKI I. RANGIROA I. TAKAROA I. TIKRI I. ep) ra & Q = as) a e tH IH = 4 tH E: > : : 3] (4214) Je] alslalc)4) Jelts ala} S|) SIS |S) SSS) 21 S/SS/214 =| al S| £|3)2/3) SS lala] Si alE Sis Cantherines pardalis (RuUppell) . x Amanses scopas (Cuvier) | | i Ostracion cubicus Linnaeus x x " cornutus Linnaeus | | x p 4 Canthigaster bemnettii (Bleeker) | x | " margaritatus (RUppell) | | x! x Tetrodon hispidus Linnaeus x id % meleagris Schneider x x|x xX} x|x Antennarius bigibbus Lacépede x LT x | Table 2 Islands from which fishes have been Number of species of fishes previously recorded . recorded Akiaki a. Anaa 2 Apataki Th Fakarava 48 Gambier Is. 3 Hiti 3 Makatea 23 Islands from which fishes have been Number of species of fishes previously recorded recorded aes 1 dvi cas 52 biereene el 62 Marutea’ 1 Nengonengo al Pinaki 4 Paneiros 50 S. Marutea al Takaroa 106 Tatakoto AL Tikei 2 "Tuamotus" _ 101 Vahitahi 2 Vairaatea 3 wih Bus Field Data of the Fishes Collected at Raroia Approximately 400 species of fishes were collected at Raroia Atoll, | but information of only about 300 was prepared in the field. The smallest specimens could not be identified or examined for lack of a microscope so that such groups as Blenniidae, Gobiidae, Labridae and Apogonidae are very incompletely recorded. Whenever possible, the species were identified, life colors recorded before preservation, and habitat and distribution noted. This information should be of value to future coral atoll teams and is pre- i sented in this section. The scientific names are all field identifications ; from inadequate literature, and must be confirmed; this section merely pre- } sents the extent of the work done. When the final report is prepared the i deficiencies will be corrected. The subheading "stations" for each species presents the stations where it was collected (providing this was recorded in the field), and the number of specimens taken is indicated in parentheses after each station numbér. It was originally intended to use Ridgway's color standards for the color descriptions, but it was among the gear held by the shipping strike and no standards were available. INDEX TO FAMILIES OF FISHES COLLECTED AT RAROTA Table 3 Number of species Number of species recorded at predicted to be Raroia 1/ | at Raroia 2/ Acanthuridae Aetobatidae Albulidae Antennariidae Apogonidae Aulostomidae Balistidae Belonidae Blenniidae Bothidae 1 Brotulidae ze Canthigasteridee | i Caracanthidae 2 Carangidae Chaetodontidae Table 3 (conttd.) aoe ihe SMe sip ett ey Number of species . “|” FAMILY recorded at. predicted to be PAGE Raroia 1/- at Raroia 2/ Carapidae - a2) 126 Cirrhitidae : 8 9 86 Congridae 2 a 57 Coryphaenidae : - 2 - Dasyatidae - 2 was Diodontidae 1 al 167 Dussumieridae Ai 3 50 Echelidae 2 3 51 Echeneididae 2 4 144 Echidnidae I te Eleotridae 6 Ingraulidae - Exocoetidae - Fistulariidae 1 Galeorhinidae 3 Gempyl.idae : - Gobiesocidae u al Gobiidae ; che ees Hemirhamphidae | aL Histiopteridae ee Holocentridae als) Isuridae 1 Istiophoridae - Moiidesy ek 4 Kyphosidae ae anes i Labridae DoE sei? lyric Lethrinidae pes] Lutianidae 7 Mobulidae 2 Monacanthidae 3 Moringuidae 2 Mugilidae 4 shiSnn Table 3 (cont'd.) Number of species Number of species FAMILY recorded at predicted to be Raroia 1/ at Raroia 2/ / Mullidae 6 12 Ophichthyidae 2 5 Orectolobidae i uf Ostraciidae 2 3 Parapercidae a. 2 Pempheridae uf x Polynemidae 1 a Pomacentridae als 17 Priacanthidae - a - Pseudochromidae 2 3 76 Scaridae 1Z 30 117 Scorpaenidae 10 15 156 Serranidae : 12 ay ves Siganidae al 2 85 Sphyraenidae 2 2 70 Sphyrnidae ~ eh = Syngnathidae 1 i 60 Synodontidae 1 x en! _ Tetraodontidae 3 3 165 Thunnidae ul 6 130 Xiphiidae - Ja Ee: Zanclidae 1 i A55 | 1/ Not including estimates of unsorted material. 2/ This list includes only those families confirmed tc be at Raroia by col- lections, sight records, or identifications by natives as occurring in this immediate region. I believe that the total number given is sub- stantially short of the fish fauna existing at Raroia, because all fami- lies that probably occur at Raroia are not included. -L6- FAMILY ISURIDAE Man Eating Sharks According to the natives this family is abundant around Raroia Atoll; only a few small individuals were seen, approximately 4-6' in length, and one about 12' long. The natives fear this group and will not spear them. Carcharodon sp. one.-~Black above, white below. Teeth with single slender points. Stations: 33(1). Habitat: One individual approximately 5) in length taken while trolling in Garue Pass. One other slightly smaller example observed swimming in Garve Pass. Raroian name: GAKEGAKI (small), RURUKI. (large). FAMILY GALEORHINIDAE Gray Sharks This group is the most abundant family of sharks occurring at Raroia, although only three species were taken. Additional species were observed but adequate collecting gear was not available. Triaenodon is speared for food, but the black-tipped sharks (Eulamia) are not eaten. Eulamia melanopterus (Quoy and Gaimard) .~-General color grey to white below. fil fins with solid black tips. Stations: 11(1), 22(1), 29(1), 31(1), 47(2). Habitat: This shark is as abundant at Raroia as all other sharks combined and occurs everywhere around the lagoon and outer reef. It is not at all abundant in the lagoon, but great numbers come over the outer reef flat to feed at high tide and sey rane henery patrol the outer surge channels day and night. Raroian name: KEIKEI (small specimens), VAKI (larger specimens). Fulamia sp. one.--Same as melanopterus in coloration except first dorsal without black. Life colors of specimen from station 45: Brown overall, pale yellowish-brown on belly. In life black edges to all fins ex- cept first dorsal distinct. Specimen 2 weeks after preservation still with same coloration as alive except black distinctly paler. Iris gold. a a Stations: 33(1). at | Habitat: This species is common in Garve Pass and along the outer edge of surge channels in 20-40' depths. The natives fear it and say it will attack people. It is a considerably larger species than melanopterus and adults approximately 5~6' long were observed. Raroian name: KOKIHE. Triaenodon obesus (Ruppell) .--Overall coloration light brow. Head broad. Dorsal fins tipped white. Stations: 45(1). Habitat: Occasionally observed over sand flats near shore in the lagoon. The natives are not afraid of this shark and spear it for food. Raroian name: ARAVA. FAMILY ORECTOLOBIDAE Carpet Sharks Apparently there is only one species of this family at Raroia, accord- ing to the natives. They say that occasionally they are seen in the lagoon. The natives spear them for food. -Ginglymostoma sp. one.-~Nasal cirrus barely reaching teeth. Upper lobe of caudal fin somewhat elongated. General color a sandy brow. Stations: 30(1). Habitat: Occasionally observed in lagoon on shore sand flats. Raroian name: KOHUEHUE (small), ROHOI (large). FAYALY AERTOBATIDAE Eagle Rays Only one eagle ray was observed during the entire summer, although the natives say that they are normally more common. Also, the natives describe sting rays as usually being fairly abundant and we were given their native names; but none were found, despite repeated search where they were supposed he. to occur. On their own initiative they pointed to drawings of sting rays in fish books and said they occur at Raroia. Aetobatus sp. one.~-This species has the teeth completely flat without mosaic arrangement. Snout somewhat elongate. Spots on underside of body as well as top. Life colors of the specimen at station 45: Solid black above with white spots. Dark markings below all black. Stations: 45(1). Habitat: According to the natives this species is found only on the lagoon sand flats at the southern and eastern sides of the atoll. Raroian name: TAPERETA. FAMILY MOBULIDAE Manta Rays Two species are common in the lagoon, particularly preferring the pro- tected east side of the lagoon. Several mantas were speared from small boats but were lost. se The natives call all mantas FAFARUA but were aware that two species occurred there. Mobula japonica (Muller and Henle).--One approximately 8-10 feet in width was watched for approximately an hour in Garue Pass, August 29th. Coloration consists of two white V's on the back; the anterior was broader and more distinct than the second. Posterior white band ex- tended far down the side to near the tail. Inside of head flaps white. When first seen it was resting on the bottom in 25 feet of water with the flippers folded out, so that the white was evident. Later it made several circles close to the boat. Region around eyes white. Underside of body pure white. Underside of tail pure white. Eyes wide apart. Tail was fairly short, approximately one-half body length, and appeared to be normal, not broken off. Observations were made ‘through a look—box. Habitat: The natives say this species is often abundant in the © Lagoon. Raroian name: FAFARUA. Manta birostris (Walbaum).--Many different specimens observed along east shore of lagoon. Back solid black. Tails long and thin, approximately equal to body length. No white bands across body. fGen Habitat: Very abundant on eastern and southern shallow sand flats in lagoon. The mantas observed appeared to be 6-15! in width. Often two and three specimens were seen at one time, Raroian name: FAFARUA. FAMILY ALBULIDAE Bone Fishes Albula occurs in only one region at Raroia--the brackish enclosed ponds of the northeast side of the atoll. A drawing of Dixonina was shown to the natives, but they did not ever recall seeing it. Albula vulpes (Linnaeus) .--Iridescent Silver to blue above. Examples over 12" are rare, most being about 6-8" long. Stations: 37(1), 38(8), 40(1). Habitat: Storms keep enclosed pools on the northeastern islets filled with brackish water, forming an ecological condition unique to this part of the atoll. The dominant species are muilet and particu- larly Albula. The floor of these pools are thickly covered with siit and detritus, and the water is usually murky. No Albula were seen in any other habitat. Raroian names: PATI AND KIOKIO for adults, NIFA for young. FAMILY DUSSUMIERIDAE Herring The number of clupeoid species occurring around Raroia is apparently slight and only one genus and species was collected. Spratelloides sp. one.--Life colors of specimens from station Al: general color on body silver to bluish-silver. Stripe along side narrow anteriorly, broad posteriorly, silvery—blue. Eye also bluish-silver shading to black on eyeball. Stations: 41(100). Habitat: Very abundant over certain lagoon coral heads in large close-packed schools. Hemirhamphus were observed chasing them. The ~50- natives say that during certain seasons, great shcools of them form masses along the western shore of the lagoon. At these times the na- tives scoop them out with nets and buckets and have community feasts. Raroian name: PIHERERE. | FAMILY SYNODONTIDAE Lizard Fishes One species of Synodus of various color patterns was found to be abun— dant in the sand around the base of lagoon reefs and coral heads. No other synodontids were found. The natives do not use this family for food. Synodus variegatus (Lacépede).—-Palatine teeth in two rows. Inner pelvic rays longer than outer. Adipose fin with a large brown spot. Body and fins crossed by brown bands. Life colors of specimen from sta- tion 57: Ground color brom to reddish-brown above fading to silvery- white below; body crossed by approximately ten vertical irregular brown bands. Head mottled brow, lighter below. Lower jaw crossed by six suffuse lines. Interorbitei and tip of upper jaw pale black. Breast and belly pure white. Dorsal fin membranes hyalin; rays with five or six horizontal series of irregular brow lines. Caudal fin crossed by six or seven vertical brow bands that are often divided; otherwise fin pale. Anal fin with four oblique brow bands along . length of fin; membranes otherwise hyalin. Pelvic fins with approxi- mately 6 faint browm bands. Pectoral fins with vertical brown bands that fade out on lower rays. Stations: 7(1), 12(1), 22(1), 23(11), 28(6),; Satueiuetl), 57(2), 61(12), 65{1). Habitat: Abundant in lagoon on sand bottom around coral heads and shore reefs. These fishes hide in the sand or in holes during the day and come out to feed at night. Raroian name: KARAFBA. FAMILY ECHIDNIDAE Moray Eels At least 15 species of this family were recorded in field notes, but the data was prepared from only approximately half of the total collections, Various species seemed to have different color phases and many were extremely difficult or impossible to identify. As a result the following notes are in- adequate. -5l- The family Echidnidae is abwndant wherever there is active coral growth or hiding places under rocks. None were found in the sand, but the family dominates the fish fauna of the outer reef coralline ridge and shallow tide pool region on the reef flat next to the beachrock shore. They are also very abundant in the inter-islet channels that sre continuous at high tide between lagoon and outer reef. Most species are distinctly restricted to definite habitats and the greatest number of species and individuals occur on the outer reefs. The natives are quite confused over names for most of the species and although they know a great number of names for eels, they are often uncer- tain about using them correctly. The Raroian names given with the follow- ing species are the ones Bengt Danielsson and I repeatedly checked with different natives. The natives eat only the large moray:eels over three feet long, roasting them whole, wrapped in palm leaves. The largest moray seen at Raroia (Gymnothorax javanicus) was slightly over four feet long. Echidna leucotaenia Schultz.--Color plain brow with white dorsal and anal fins. Head white, striped with brom. Some teeth molariforn. Stations: 17(8). Habitat: Rare at Raroia; in the outer reef surge chahtnets and in the sea urchin holes on the bncamicna ridge. Raroian name: MAKIKI. i Echidna sp. one.--Distinctive characteristics consist of molariform teeth, 30. broad vertical brown bands with pale brow interspaces; irregular outline of bands. Tail approximately as long as body. Similar to E. polyzona but markings different. tg Stations: 28(1). Habitat: Only one specimen recorded from a lagoon coral head. ~ Raroian name: none. Enchelynassa canina (Quoy and Gaimard).--Distinctive colors solid brow without markings. Anterior nostril with an expanded lobe. Posterior nostril with a raised.flaring rim. Life colors of specimens from station 58. Head and body solid brow; head distinctly lighter, par- ticularly behind eye onto occiput. Pores above and below bordering jaws cnetered in white circles. Snout and lower jaw tip dark brown. Fins with dark brow base shading to white on outer half. A large specimen, approximately 2' long, (very likely another species) is dark brovm to blackish-brown overall, without any marking or light areas. Nostrils somewhat different than on smaller specimens. Stations: 17(6), 26(1), 45(1), 58(4), 59(1). et -Eve Habitat: Outer reef surge channels dow to 15' at least and over entire coralline ridge living in the sea urchin holes. One of the dominant moray eels in the surge channel region. Raroian name: MAKIKI. Uropterygius thopterus Hipaleeies Send and body dark brow to black Bpeesee with aiite and bluish white dots. Midcaudal rays long. Stations: 10(8), 17(7), 45(1). ‘Habitat: Moderately common in outer reef surge channels and next to shore on reef flat. Burrows deep into cracks in the coral and is very difficult to collect. Raroian name: none. Uropterygius fasciolatus (Regan).--Vertical fins confined to tail. Posterior nostrils in a raised tube. Head and body with about 42 dendritic stripes. Ground color browmish white. Dark stripes narrower than light inter- spaces. Stations: 17(1), 26(1). Habitat: Rare at Raroia. Taken in outer reef surge channels. Raroian name: none. Uropterygius marmoratus (Lacépéde).--Tail longer than snout. Color pattern mottled. Gill operning on level with eye. A single pair of nostrils present. Life colors of specimen from station 11: ground color pale browm. Body mottled with black and brom Spor s superimposed on each other. Iris golden brow. Stations: 11(1). Habitat: Recorded specimen taken from under rocks in an inter- islet channel near Garve Pass. Raroian names: none. Gymnothorax fimbriata (Bennett).--Distinctive characters consist of a green. edge to fins, gill opening not in black spot, body and fins covered with round black spots, and ground color solid brom. Life colors of specimens from station 49: Ground color overall a solid browm tinged with green, somewhat lighter below. Head with tips of jaws slightly darker. Small black spots on upper part of head behind eye. Black spot at rictus of upper jaw. Gill opening not in black spot. Vertical fins same color as body, edged in white in smaller specimens, and edged in green in larger. Dorsal fins with oblique spots along the entire -53- length of fin; posteriorly they are continuous with those on hody. Anal fin without entire spots, only lower edges of spots on body ex- tending onto fin. Body with approximately 3 series:of black spots | along its length; anteriorly the spots are smaller than the inter- spaces, posteriorly they are approximately the same 'size or somewhat smaller. Caudal fin with continuous color pattern of dorsal and anal, also light edged. Pores bordering upper and lower jaws in white spots. Stations: 49(5). Habitat: Lives in the Ulva-like Microdictyon on lagoon corai heads. Raroian name: KIART. | Gymnothorax monostigma (Regan).--A large black blotch bordering eye poster- iorly. Head and body dark brown without any other markings. Pores on lower jaw surrounded by white. Posterior nostrils in iong white tubes. Iris red. Stations: 17(1), 45(12). | . | Habitat: Outer reef surge channels and sea urchin holes in the coralline ridge. This species was observed only at a few stations. Raroian name: none. Gymnothorax gracilicauda Jenkins.--Body brow, with 45-50 vertical dendritic darker brown vertical stripes. Lower jaw white with brom tip. Upper jaw with one maxillary row of teeth and one or two inner teeth. No black spot on gill opening. Head long, occiput not gibbous. Anal fin white. Dorsal fin same color as head and cp except edged in white. Stations: 17(2), POMS 2543) i. Habitat: Moderately common in outer reef surge channels. Raroian name: none. Gymnothorax picta (Ahl).--Intermaxillary teeth reduced. Color pattern con- sists of black specks or tiny black spots on a white background. No teeth molor-like. In-.specimens below 100 mm. the spots on the body occur in 4-6 irregular rows along body. In specimens 150-200 mm. the spots have pale centers. In specimens 200-300 mm, blackish rings are broken into small spots and arranged in a ring of spots. In the very large specimens the body becomes speckled with ey black aoe Stations:G(10)»5 1O(1) 5. 11(12), 14(5); 22(6). Habitat: This species is very abundant in the shallow inter-islet channels and next to shore on the outer reef flat. It lives in holes EI in the beachrock and dead coral blocks. This species comes out at high tide to feed. Raroian name: KUIRU. Gymnothorax flavimarginata.—-~Vertical fins edged in green. Gill opening black. Body black, mottled with brown and black. Stations: 8(11), '10(S), 12(3), 13th); 121) eeataenesta) , 26(9). Habitat: One of the most abundant eels at Raroia, occurring in a wide variety of habitats in the outer surge channels, reef flats and on lagoon coral heads and shore reefs. Raroian name: MAKIKI. Gymnothorax javanicus (Bleeker).--Black spots along body in 3 irregular series. Ground color solid brow. Gill opening in a black spot. No inner teeth on maxillary. 4). Habitat: Moderately abundant on lagoon coral heads and shore reefs; uncommon on outer reef although occasionally observed in large pot holes on coralline ridge. This is the largest moray eel at Raroia and is commonly eaten by the natives. The natives fear this eel more than all others and kill them whenever possible, whether they want to eat them or not. Raroian name: HAMURENGA (small), TIOHU (medium), TAVERE (large). Gymnothorax buroensis (Bleeker).--Three rows of maxillary teeth present. General coloration with large black blotches on light brown body. Small specks and dendritic spots of darker coloration over head and body. Anterior nostrils white. Gill opening not in black spot. Stations: 17(1). Habitat: A few specimens observed in outer reef surge channels. Raroian name: KIARI. Gymnothorax thyrsoidea (Richardson).--Ground color solid brom. Darker brown dendritic spots over body, tending to form vertical rows pos- teriorly. Tiny darker specks and spots overall. Maxillary with two : rows of teeth, the inner series consisting of approximately 8-10 teeth. Stations: 17(8), 28(7). Habitat: Taken in outer reef surge channels and on lagoon coral heads. Raroian name: KIARI. ~55=— Gymnothorax petelli (Bleeker) .--Characterized by. black vertical stripes on a white body. Stations: 17(6), 22(2), 26(i), 45(2). Habitat: This is one of the dominant moray eels ‘on. the outer reef coralline ridge in sea urchin holes and in the surge channels. Raroian name: TAKATAKA. FAMILY ECHELIDAE Worm Eels © According to field notes, several species of echelid eels were collected but only one was recorded. They were quite hard to collect, only a few being obtained after several hours exposure to rotenone. The natives know nothing about them and referred to them by the general eel term KOIRO. They do not eat. then. cat Muraenichthys laticaudata (Ogilby).--Characterized by the dorsal fin origin over anus, the eye situated over rictus, the posterior nostril in the upper lip, and a light brown coloration without distinctive markings. Life colors of specimens from station 58: yellowish—light brown overall. Iris light blue. Fins same color as body, pale. Stations: 17(2), 20(1), 58(2). Habitat: Taken only in deep surge channels next to the coralline ridge of the outer reef flat. All specimens were collected in 12-20! of water. we ain foie FAMILY OPHICHTHYIDAE _onake Eels At least two species were collected of this family, but only one was recorded in field notes. Since this group burrows in sand and never was killed at Raroia by the rotenone, it is very likely that East species were not collected. The family was found only in the lagoon in shallow water on tops of coral heads and shore reefs. Some natives gave the name MATIROHE to this group but they might have made up this name for the occasion. Sometimes the natives would be embarrassed at not having a:name for some common fish and invent a name for it. -56- Leiuranus semicinctus (Lay and Bennett).—-Dorsal fin origin over gill open- ing. Twenty-eight BpCte alorig head and body, separated by broad white interspaces, Stations: 9(1), 12(1), 28(1), 61(1). Habitat: Moderately common burrowed deep in the sand in lagoon, in shallow water on top of shore reefs -and coral heads. Most abundant on. west side of lagoon. Raroian name: MATIROHE- ? FAMILY CONGRIDAE Conger Eels The conger eels are of minor importance at Raroia and only two species seem to occur there. This family is almost entirely limited to a shallow sandy environment on top of coral heads or shore reefs in the lagoon. Sel- dom were more than one specimen seen at the same locality. These eels burrow in the sand similar to Moringua, and were taken entirely by poison. The natives know very little about them and do not eat them; they give them the general eel name KOIRO, but only the older natives use this name ete any certainty. ; Conger cinereus Ruppell.--Characterized by very small pectoral fins, uniform coloration on head and body, and a submarginal black edge to the verti- cal fins. Stations: 7(3),.8(1), 9(1), 23(1), 28(2), 45(1). Habitat: This species lives in the sand in: shallow water less than 4! in depth on the shore reefs and tops of coral heads in the lagoon. It would usually take an hour or more before the rotenone would drive them out of hiding. Actually this eel may be more abundant than the collections indicated. The rotenone might not have affected adults, since only young specimens were taken. Also adults may only be in deeper water than our collecting permitted. Conger sp. one.--Characterized by a blunt short body and an all white colo- ration. _ Stations: 9(1). Habitat: One specimen taken on west coral shelf in lagoon. a FAMILY MORINGUIDAE Worm Eels This family was almost completely unknown to the natives until our ichthyocide drove them out of the sand. Some individuals are buried over a foot deep in sand at the base of coral and often it would be several hours before rotenone would affect them. This group is confined to water less than six feet deep, and very seldom was noted outside the lagoon. Some natives said these eels are called HOEA but others said there is no name for them. Moringua abbreviata (Bleeker).-~General color orangish-red. Fye minute. Pectoral fins lacking. Lower jaw longer than upper. Veitical fins confined to extreme tip of tail. Stations: 8(3), 9(3), 10(4), 28(7), 61(2). Habitat: In shallow water burrowed in sand. Common on lagoon west shore reefs and sand—topped coral heads near shore. Most examples observed near bases of coral growth. Largest concentrations on lagoon west shore reefs in region of Garumaoa Village to Garuve Pass. Raroian name: none. Moringua bicolor Koup.--Characterized by lower jaw longer than upper, fairly large eye, well developed pectorai fins, a general silver coloration, and by vertical fins interrupted on tail. Stations: 8(3), 11(1), 28(1). Habitat: Same as Moringua abbreviata except not as common. Raroien name: none. FAMILY BELONIDAE Needle Gars One species present at Raroia scattered in open surface waters in moderate numbers. The natives are not familiar with any other species oc- curring at Raroia. Belone platyura Bennett.--Life colors prepared at station 38 iridescent sil- ver with blue above. JVhite below. Stations: 27(12), 35{1), 38(1). -58- Habitat: Adults come onto the outer reef flat around the atoll during high tide at night to feed. Small schools are moderately com- mon in the open surface water of the lagoon. The natives use this species for bait, but do not eat then. Raroian name: KAGU. FAMILY HEMIRHAMPHIDAE Half Beaks Only one species known at Raroia that is very abundant in the lagoon and along the outer reef. Hemirhamphus pacificus Steindachner.--Teeth minute in jaws. Preorbital as long as eye. Pelvics slightly nearer caudal fin than head. 46 pre- dorsal scales present. No stripes present on body. Tip of lower jaw reddish-orange in life. Life colors of specimens from station 58: General color iridescent silver, blackish above. A median silver stripe present down side that turns blue or black in formalin. Belly silvery white. Head iridescent silver on sides, blackish or bluish black on top of snout and head. Iris iridescent silver. Dorsal fin edge hyalin. Caudal fin dusky black on upper lobe; lower lobe hyalin with scattered chromatophores. Anal fin hyalin except for scattered melanophores near tips of first rays. Lower jaw with fleshy tip red. Pelvic fins hyalin except for dusky bases to the first rays. Stations: 7(6), 9(11), 12(1), 17(1), 22(1), 26(4), 27(36), 28(2), 38(2), 39(5), 45(6), 58(4), 61(1). Habitat: This ‘species is one of the most common fishes at Raroia and occurs in large schools in the open water. At night at high tide, great numbers come onto the outer reef flat to feed. The natives catch them with nets for food and bait. They troll for larger open water fish with them. Raroian name: FANEA. FAMILY AULOSTOMIDAE Trumpet Fishes Aulostomus is very prominent around coral in the lagoon and inter-islet channels. It was not observed in swifter moving water, in the Passes or around the outside of the atoll. This form is one of the main scavengers of the coral reefs, and is easily taken with baited traps. The natives uninten- tionally take them on hook and line and seldom eat them. 59- Avlostomus chinensis (Linnaeus).--This species is remarkably variable in color. Some specimens are yellow to yellow-orangg overall. Other in- dividuals vary from a solid brown to a dark grayish-brom. A few ex- amples were mottled with white and black and dark browns to black. 66( eon 71), 9(3), 2301), 28(1), 3202), Zeus hot ays ay tty s 2). Habitat: Most abundant on the lagoon west shore reefs and neigh- boring coral heads. This species was attracted to rotenone stations where they would gorge themselves on the smaller dead fishes while they in turn succumbed to the effects of the derris root. While spear fish- ing around coral heads we would often see lines of Avlostomus head to tail winding in and out of the coral. Also, large adults would often be seen right next to large sea basses closely following them along as they moved about. Apparently, this peculiar association was for pro- tection and for feeding on the sea basses prey. Raroian name: KAKAVERE TUPOUPOU. FAMILY ®ISTULARIIDAE Cornet Fishes The relative abundance and distribution of the single species observed at Raroia is the same as for Aulostomus. Fistularia petimba Lacepede.-—General color brown to greyish-brown. Stations: 7(1), 8(1), 9(2), 10(1), 22(1), 28(2), 33{1), 35(1), 45(2). Habitat: This species appeared to have almost identical habits as Aulcstomus, although they were more often observed in the inter- islet channels than the other genus. Also this species comes onto the outer reef flat at night to feed and the natives kill them with long knives and use them for bait, particularly for night fishing. The natives do not eat them. Raroian name: KAKAVERE. FAMILY SYNGNATHIDAE Pipe Fishes Despite intensive search in habitats where pipefishes normally occur, only one rather common species of Corythoichthys was collected. This pipefish -60= congregates around particular tide pool coral patches and individuals are not found between these areas. They feed in the tide pools at low tide, keeping within the immediate vicinity of their coral patch. At high tide they hide in the coral, just the opposite of most outer reef flat fishes. Corythoichthys SP. one.—~Snout slender, sharply pointed. Tail and snout bright red. Dorsal fin covered with red spots. Approximately 16 spots down side of body. Two dark longitudinal lines on side of head. Dorsal fin with 32 rays. | Stations: 10(10), 11(1), 22(1), 25(10), 57(1). Habitat: Entirely confined to restricted coral patches in tide pools next to shore on outer reef flat and occasionally in mouths of inter-islet channels on outer reef flat. Noted only on west side of atoll. Raroian name: none. FAMILY HOLOCENTRIDAE Squirrel Fishes Without a doubt the most confusing group of fishes at Raroia is the Holocentridae. There seem to be species in both Holocentrus and Myripristis that are very similar in appearance even in fresh material and it was dif- ficult to do an adequate job with field determinations. A large number of kodachrome slides were taken, copious color notes and field sketches pre- pared, and it is hoped that this data will prove of value. Practically all of the specimens taken are incorporated in the following notes. Fifteen species are recognized in the field records, but it is suspected that fur- ther study will expand the number. It is believed that all species normally occurring at Raroia were taken. The family is abundant in shallow water to 15~25' wherever there is — coral or hiding places. Few specimens are observed during the day, but re- main hidden in holes. At night they come over the reef flats in great nun- bers to feed. They are specially abundant at night in the outer reef surge channels and on the outer reef flat. af The natives are hopelessly confused over the names of most species, and there was definite agreement for only KIOA, TIKEIKEFI, RUKERUKE, PETI and PETIMU. The remaining names have been given by the more reliable of the older natives. Apparently at one time each species was named, but now the natives are not sure to which species most of those names belong. The natives eat RUKERUKE, PETI and PETIMU, but usually not the other species. moi. Holocentrus sammara (Forskal).--Dorsal fin with a distinct large black blotch between first four spines. Upper and lower edges of caudal fin black. Anterior edge of soft dorsal and anal black. Life colors of specimens from station 61: General coloration red above to silver below. Head black to reddish-black on interorbital. Top of snout dusky red. Iris silver, tinged with red around lens. Cheeks silver with brownish black spots. Lips white. Underside of head white. Overculum brassy reddish- brown. Eyebali black above with narrow horizontal red line. Body with dusky browmm lines along scale rows. Body above dusky red. Lateral-line forming a conspicuous dusky red line that is the most prominent of the markings along the side. Scales above anal fin with dusky brow centers. Breast and belly silver white. Spinous dorsal fin with large black spot on first three membranes; spot edged in red. Outer tips of all membranes . pure white. Neer base of each membrane of spinous dorsal is a large white spot, except last which is all red without white edge. Middie of fin pale dusky black fading into red on last membranes. Soft dorsal an- teriorly edged reddish-black, fading to yellow and then hyalin poster- iorly. Caudal fin reddish-black on upper and lower most rays, middle of fin vale yellow; postericr edge nyalin. Anal fin with third spine and first soft ray pale red. Remainder of soft fin hyalin. Membranes between second and tnird spine white, hyalin at outer edge. Pelvic and pectoral fins hyalin. Upper pectorai rays tinged with red. Fine pale brown spots immediately before pectoral fin but not below. Stations: 7(}), 14(2), 22(10), 23(1), 27(1), 42(1), 47(1), 60(1), 61(6), 65(2). Habitat: Moderately abundant in shallow water around coral heads and shore reefs, occurring in greatest numbers on the outer reef flat at night. Found in association with Holocentrus laevis. Raroian name: TINGITINGIA. Holocentrus laevis Gunther.-—Dorsal fin plain. Upper and lower edges of caudal fin black. Black spots on preopercle. Life colors of speci- mens from station 61: Coloration very similar tc sammara except paler red, and spots on cheek and around pectoral base more distinct. Head reddish-black to reddish-browm above, darkest on interorbital. Iris silver, a vertical reddish-black band immediately behind iris; it is sometimes anteriorly tinged with red. Cheeks with a distinct brownish- black spot on almost every scale; otherwise silvery white. Jaws white, except for dusky tips. Opercular edge light reddish-brown, a row of brown spots bordering preopercle edge. Body reddish-brown above to silvery—white below. Each scale row with a longitudinal dusky brow line; these are faint below lateral-line. Lateral-line recdish—brom, forming the most prominent marking on the body. Breast before and below pectoral fin with 3-4 longitudinal rows of brownish—black spots; lowest with seven spots. Spinous dorsal hyalin; there may be a faint red spot on first membrane near tip. Soft dorsal with first rays red; remainder of fin tip pale yellow to hyalin. Caudal fin with outer ~62— rays reddish-black to solid red posteriorlysiuMiadie’ rays pale yellow. Posterior edge of middle rays hyalin. .Analsfinshyalin except third ih spine and first soft ray pale red... Membranes ‘ tbe tiiéen ‘second and third... spines white. Pelvic and pectoral . fins ieccoteatiie a pectoral 5 ae tinged with red. a SN a Vt ee Stations: 8(2), 92) 5 130), 23(3) »: 28(2), e1(a0) Habitat: Confined to the. lagoon ws olen and coral heads’ amt ‘shallow water. Found in association with Holocentrus sammera’ ‘bat: not, as abun- Mg dant. Raroian name: TINGITINGIA. ; Holocentrus opercularis ee eee dorsal with a wide black band; tips of membranes white; dorsal third of membranes white with a black spot anteriorly between each spine.. Second dorsal fin with first ray red, remainder orange. Caudal fin with outer rays reddish-black. Inner rays orange. Anal fin with third spine and first soft ray red. Re- mainder of fin orange. Pelvic fins pale, slightly reddish. Fectoral fins pale red. Head red to reddish—black; operculum darxest. Body red to brownish-black dorsally. Each scele row icnds to'have dark longi- tudinal lines ventraily. Bottom of head and body white. Above color notes prepared from preserved specimens from station 31. - Stations: 33(1), 61(2), 65(2). Habitat: Rare at Raroia; taken on lagoon coral heads and shore | reef in Garue Pass. A few additional observations record them from widely scattered coral heads in lagoon. Raroian name: KIOA. Holocentrus microstomus Gunther.--Characterized by a black ‘blotch between first to third dorsal spine, then a weak greyish blotch between each spine on membranes. . Posterior spinous rays with black. spots basally; tips of all membrares red; basal portion white. Seven broad longi- tudinal stripes on body. Stations: 7(1), 9(3), 28(3), 38(1), 4949). Habitat: A nondescript species that is moderately abundant in shallow water around lagoon corel heads and shore reefs. Specimens - were also taken in enclosed brackish pools on northeast islets. Raroian name: TIKEL. a a dl ed edge and caudal fin base.. Head: cand body ‘bromish-red with faint brown mottled specks. No. distinct longitudinal stripes. Spinous dorsal -63- with basal third of membranes pale red; middle third pale black; outer | third brilliant solid red. Soft dorsal, anal and caudal fins orangish- i red. Pelvic fins hyalin. Life colors of specimen from station 58: Head and body brownish-red above, shading into silvery white below. | Head reddish-brown above, specked with brown over side, including cheeks | and lips. Underside of head mostly white, with scattered brown pigment cells. Operculum with a brown spot above, silver-brown over most of operculum. Body specked with brown over sides to level with pectoral fins.. No distinct stripes on body though the brom specks tend to clump on the scale rows, forming vague longitudinal bands. Anus black. Spinous dorsal fin edged in a solid bright red band. Middle of fins suffuse black, solid black dorsally; basal portion solid red, varying in distribution and intensity among specimens. Soft dorsal reddish on first rays, dusky black for remainder of fin (tinged with red) with hyalin edge. Caudal fin pale red, the upper and lower most rays hyalin, the edge of the fin hyalin; upper and lower rays brightest red. Anal fin soft rays pale red basally, hyalin at tips; first 2 soft rays bright red. First two spines white; third spine red; spinous membranes and soft ray membranes hyalin. Pelvic and pectoral fins hyalin. Upper | pectoral rays tinged with red. Stations: 8(1), 9(1), 11(1), 14(1), 20(1), 23(1), 26(2), 27(61), 45(16), 58(1), 59(5), 65(1), 66(5), 10(1). Habitat: The most abundant holocentrid at Raroia in shallow water on shore reefs and coral heads. Great numbers come over the outer reef flats at night at high tide to feed. Raroian name: TEMU. Holocentrus diadema Lacépede.--Characterized by horizontal black stripes, and i a solid black spinous dorsal fin with oblique white stripe. Life colors of specimens from station 61: General colors solid bright red with white lines. Head solid red except for white line between eye and upper jaw which continues to opercular margin, and a vertical, white line on preopercle that includes large spine. Top of head red, slightly darker on interorbital. Lips red. Underside of head pale red except for gill membranes which are pure white. Iris red, irregularly edged in silver. Body with nine longitudinal white lines, that extend the length of the body; two short white lines on breast as well as a median | line dovm breast and belly; also a median dorsal line before dorsal fin. White lines become broader ventrally; on midlateral region they are edged by brown lines. Body is browmish-red dorsally. Spinous dorsal fin black with tips of all membranes except last hyalin. Re+ mainder of fin solid black except for an oblique longitudinal white line that is disrupted on middle rays. Soft dorsal red across middle of rays, remainder of fins including membranes hyalin. Outer caudal rays red, middle rays pale red. Outer edge of middle rays hyalin. Anal fin pale black between third and fourth spine. Fourth spine and first soft ray red. Remainder of rays pale red and membrane hyalin. igi Pelvic fins with first two soft rays bright red. Remainder of rays pale red and membrenes hyalin. Pectoral fins hyalin, upper rays pale red. oa) tations! 71)» 9(2), 22(2), 23(2), 2316), 2822), 4912), 6218), Habitat: Common at Raroia, but almost entirely confined to lagoon coral heads. Raroian name: TIKEIKEI. Holocentrus binotatus Quoy and Gaimard.——Characterized by a solid red dorsal fin with two small black spots at the base of the first two membranes. No silver area on caudal peduncle. Life colors of specimen from station 49: Head and tbody solid red without dark markings. Interorbital dusky black. Iris red posteriorly, golden anteriorly. Body solid red above, shading into light red indistinct stripes below. Caudal peduncle solid red. First dorsal brilliant solid red with a suffuse black spot at the base of each of the first two membranes. Second dorsal with anterior rays all red; remainder of rays red basally. Caudal fin rays red at base and on tip of upper lobe; edge of lower lobe hyalin. Anal fin soft rays red basally and on all of first ray membrane; remainder of soft fin hyalin. Anal spine red. Pelvic fins with soft rays dusky red across middle of fin, membranes hyalin. Pectoral fins pale red, lower rays hyalin. Stations: 49(1). Habitat: One specimen taken on a lagoon coral head. Raroian name: TIKEKE. Holocentrus spinifer (Forskaél).--Life colors of adult from station 55: Head and body uniform brilliant red. Head with a bright red vertical patch imnediately behind eye. Iris golden red. Lens black. Body uniformly red. Spinous dorsal fin brilliant solid red that breaks down to yellow when preserved in formalin and finally becomes pale white. Soft dorsal pale red on rays; membranes hyalin except for pale red on middle of anterior rays. Caudal fin red shading to hyalin at posterior edge. Membrane between second and third anal spine orange- red; anterior soft rays red shading into yellow on last rays; mem- branes pale orange-red. Pelvic fins uniform reddish-orange. Pectoral fins orange shading to hyalin on outer one-third. Standons: 20(1)), 23(8), 28(7). 25112), 55(1), 60(3) 6107) « Habitat: Solitary individuals live in holes around lagoon coral heads and shore reefs. This species is abundant at night in the outer reef surge channels but does not come onto the outer reef flat. It bites on a hook at night. Uncommon in surge channels during the day. Raroian name: TIKEI (young), RUKERUKE (large). Wee Holocentrus sp. one.——-Characterized by a brilliant solid red coloration overall. Tiny black specks on head and horizontally on scale rows. Spinous dorsal with large biack spots on first two membranes. Suffuse pale spot at posterior edge of each membrane for other spinous rays. Remainder of spinous membranes reddish-black. Other fins solid red. Stations: 17(2). Habitat: A few small specimens observed and taken in outer reef surge channels. Raroian name: TEMU. Holocentrus sp. two.--First dorsal without black pigment. Large red patches dorsally between first two spinous membranes; reddish-orange streaks near tips of other membranes, particularly last three; otherwise spin- ous dorsal hyalin. Body without distinct dark streaks, but a great number of black specks cover head and body. Fins other than first dorsal pale reddish-orange. Outer rays of caudal fin slightly darker at base. Head reddish-black on interorbital and eccinat black spot on upper region of operculum. . Stations: 22(55), 57(6). Habitat: Taken only north of Garve Pass in inter-islet channels and lagoon shore reefs, Raroian name: TEMU. Holocentrus sp. three.--Head without distinct markings. Spinous dorsal with red band in upper membranes with a narrow white tip. Last mem- branes with red spots at base; middle of fin pure white. Life colors of specimens from station 57: Head reddish above, silver and white below. Opercle with a vertical dusky red band anteriorly. Lips pale red. Branchiostegal membranes and underside of head pure white. Body with 8 distinct longitudinal red bands extending to caudal fin base; some of the dorsal bands are very short. Soft dorsal red anteriorly, shading to pale red on last rays and hyalin at edge. Caudal fin with outer rays red and inner rays pale red. Anal fin bright red between second and third spines, pale red on soft rays shading to hyalin distally. Pelvic fin with first ray red, shading to hyalin on last ray; membranes hyalin. Pectoral fins pale red. — Stations: 13(1), 23(9), 27(1), 42(1), 57(2), 65(3). Habitat: Taken on lagoon reefs and coral heads on west side near Shore and on outer reef flat. Raroian name: none. -66- Holocentrus sp. four.--Similar to RUKERUKE but has white edge to spinous dorsal and lacks preopercular red patch. Life colors of specimens from station 66: Overall brilliant red, Head solid red above (not at all black). Iris silver around lens, remeincer red. Body solid red above shading into suffuse longitudinal red stripes below. Belly and breast white, with scales basally white posteriorly edged in red. Spinovs dorsal brilliant solid red, the outer edge pure white. Soft dorsal and soft anal anteriorly solid red: to posteriorly only red on rays; mem- branes posteriorly hyalin. Caudal fin solid red. Anal fin solid red between third and fourth spine except pure white distally next to third spine. Pelvic fins red, brightest anteriorly, with a pure white anterior margin. Pectoral rays red, membranes hyalin. Stations: 26(1), 66(4). Habitat: Outer reef surge channels and shore reefs of Garue Pass. Raroian name: TIKEI. | Myripristis murdjan (Forskal).--Characterized by the second dorsal and anal fin having an anterior white edge followed by a dusky black vertical line. Life colors cf specimens from station 49: Overall red, darker above. Head red, reddish-black above. Eye black above. Iris with a large rectangular silvery’spot behind eye ball. Iris golden red shad- ing into purple anteriorly and golden below. Posterior edge of eye brown. Opercular flap brom as are scales bordering it on body. Body Silver, each scale edged in red for ventral scales, edged in browmmish- red for median scales, and-above lateral line each scale purplish-silver with a dark brown edge. Spinous dorsal edged in white; a broad red band longitudinally dow the’ middle of each membrane which may have patches of yellow in it; basal part of membrane pale yellow. Soft dorsal with first ray white; remaining rays red. Dusky black region behind white almost lacking in specimens here described. Distal region of second to fifth soft rays bright red. Caudal fin bright red. Anal fin same as dorsal. Pelvic fins with first soft ray white, next rays red, shading to hyalin for last ray. Tips of all soft rays hyalin except eceo Pectoral fin pale red, hyalin distally. Stations: 7(2), 9(2), 12(2), 13(2), BHA), 28(5), 424), 25018). 49(4). Habitat: Abundant on lagoon coral heads, shore reefs and outer reef flat. t. Also observed in outer reef surge channels at night. Raroian name: PAPURAGO (small), PETI (large)... Myripristis adustus Bleeker.--Characterized by broad black tips to soft ver- tical fins. Fourteen soft dorsal rays. Life colors of specimens from station 28 one month after preservation: General color black above to dull reddish brovm below. Head black on top of head. Lips blackish. ~67~ Black spot on opercular margin. Iris black. Spinous dorsal edged in solid black; base black shading to white on middle of fin. Soft dor- sal and soft anal broadly tipped black; remainder of these fins dusky 1 yellow. Caudal fin edged in black, anteriorly shading to yellow for | most of fin. Pelvic and pectoral fins pale yellow. Stations: 28(2). Habitat: Apparently restricted to a few large lagoon coral heads. | This species is prized for food by the natives but they seldom find | them. At station 28 approximately 15 were collected but were dispatched © by the natives before they could be preserved. Raroian name: PETIMU. | Myripristis sp. one.--Characterized by no black markings on dorsal or anal fins. Opercular margin black. Pectoral fin axis black. High number of gillrakers on first.arch. Life colors of specimens from station 61: Head and body solid red. Iris golden red anteriorly and poster- iorly, black above, and golden to reddish-golden below. Spinous dorsal solid red, pale red or whitish-red centrally on each membrane. Soft dorsal and anal bright solid red; posterior rays shading to pale red at tip. Caudal fin bright solid red, median rays fading to pale red at tip. Pelvic fin rays pale red anteriorly, shading to lighter red pos—- - teriorly. Membranes between spine and first soft ray pale red. Pec- toral fins pale red on rays and membranes, fading to hyalin at tip... Stations: 23{25), 28(1), 61(10). Habitat: Abundant around lagoon shore reefs and coral heads. Raroian name: PETI. FAMILY MUGILIDAE Mullet Four species of mullet are recorded from Raroia and probably additional species were taken. All species noted were not collected so that there ap- pear to be at least eight species occurring at this atoll. The family is most abundant schooling along shore in the lagoon and on the outer reef flat at night feeding at high tide. It is also the dominant family besides Albulidae in the brackish enclosed pools on the northeast islets; the young are especially abundant in these pools. The natives use this family for food, but prefer. many groups of fishes before eating mullet. All the species have specific Tuamotuan names. Bara yd uni, baleen Biathes (mach jaw with one row of fine teeth pointing downward. Jaws strongly oblique. Approximately 35.scale rows between opercular margin and caudal fin base. No spot at.pectoral fin base. Lips without crenvlate flaps. Edge of caudal fin black. Stations: 16(2), 22(1), 27(2), 37(1), 38(8), 58(1). Habitat: This species has a surprisingly varied distribution. It is dominant mullet in the northeast brackish ponds and on the outer reef flats. The natives use them for food. Raroian name: VOA. Mugil crenilabis Forskal.—-Upper lip with several rows of fleshy papillae. Lower lip crenulated. Black spot present on pectoral fin base. Life -- colors of adult from station 60: Head,and body iridescent silver overall, dusky black above. Head darkest on snout. Lips and lower half of head 1 EE sade list Iris golden. Spinous dorsal dusky black on spines, membranes ‘hyalin. Soft dorsal and anal scaled over most of fins, rays dusky black. Caudal fin dusky black. Pectoral fin with a black spot at upper edge of base, otherwise brownish black. Stations: 60(1). Habitat: A small school observed. along shore of Teremu Islet, over sand bottom. Raroian namie: KANAB. sant engeli mideher Depo gtsea de! adepose eyelid present. Mouth almost level, ithout projecting teeth. Lip edges smooth. Life colors of specimens pr station 60: Head and body iridescent silver, darker above. Head grayish on top; upper lip dusky black. Iris silver, ringed with yellow- orange. Body blackish above shading to pure silvery white below. Dor- sal fin ’ with dusky black rays, scattered melanophores on membranes near ‘rays. Caudal fin dusky black, with an abruptly blacker posterior edge. Anal fin same as dorsal. Pelvic fins hyalin. Pectoral fins dusky black on basal two-third of rays, membranes hyalin. Stations: 60(4). Habitat: Abundant in small schools on west side in lagoon next to shore. They swim at surface feeding, bobbing the tops of their heads out of water. Often in ass Gi AP aen with Mugil vaigiensi : Raroian name: KAVATA. iN \ Mugil vaigiensis Quoy and Gaimard.--Characterized by black pectoral fins, by a pale caudal fin, and bythe pectoral fin being shorter than the head. Stations: 22(1), 38(3), 60(12). -69- Habitat: Common in lagoon along shore and around coral heads near shore. Small schools present in brackish ponds of northeast: islets and scattered individuals on outer reef flats, particularly -. at high tide. Raroian name: TUGOUGOU (small), HOPIRO (large). FAMILY SPHYRAENIDAE Barracuda Two species are abundant around the outside of the atoll and are con- monly taken by the natives when trolling. The Raroians say that only these two kinds of barracudas occur at Raroia and that TUATAU (Sphyraena helleri) are sometimes poisonous to eat. More definite information on toxicity could not be obtained. Z : Sphyraena snodgrassi Jenkins.--Maxillary extending to below front of eye. | Scale rows about 80. Life colors of specimens from station 62: General color iridescent silver, blue above, with vertical bands along upper Side. Top of head blue shading to brow on sides of head. Jaws iri- descent bluish-brom. [ris iridescent gold to blue. Spinous dorsal black. Soft dorsal black with tips of anterior rays white. Basal two- thirds of anal fin white; edge of fin with a broad black band. Caudal fin black; tips of upper and lower lobes white; middle of base silvery with a black spot in center. Pelvic fins pale red and hyalin. Pec- toral fins dusky white at base shading to black at edge. Stations: 33(3), 62(2). Habitat: Common at Raroia.. Open water in Garue Pass and over coralliferous outer bench of west side of atoll. Said to be poisonous at times but the natives ate them during the period we were at Raroia. Raroian name: ONO. Sphyraena helleri Jenkins.--Maxillary not reaching eye. Distance from pec- toral fin origin to pelvic fin distinctly more than postorbital distance. Life colors of specimen from station 56 immediately after capture: General color iridescent silver; blue above. No distinct markings on head. Top of head and tips of jaws black. Iris metallic brassy. Body blue shading to black above. A gold stripe present dom side midway between lateral line and ventor. Dorsal fin dusky black distally on anterior rays. Caudal fin dusky black overall. Anal fin hyalin, the anterior soft rays with scattered melanophores. Pelvic fins completely hyalin. Pectoral fin hyalin, first rays dusky black. Stations: 56(1). 1 Habitat: Uncommon at Raroia. Taken in open water in Garue Pass. Raroian name: TUATAU. FAMILY POLYNEMIDAE Thread Fins During the period at Raroia only one specimen of. this family was seen. The natives use the family for food but were vague about its distribution and abundance. Polydactylus sexfilis ere waPoctere : fins with six free ‘filaments ventrally. Stations: 45(1). at me Habitat: Rare at Raroia. Speared on outer reef near Oneroa. Raroian name: MOI.’ FAMILY SERRANIDAE Sea Basses This group is second only to the Scaridae as food for the Raroians. ‘Twelve species were taken at Raroia and such intensive collections were made of this family that I doubt that more than one or two species were missed. Despite the fact that several species of Epinephelus are hard to distinguish, the natives are keenly aware of the differences and unerringly name all specimens—-young or adult. Almost all species have at least one name. Ex- cept for Grammistes, the unidentified genus and Paracenthistius, all species are commonly used for food. The last named fish is poisonous at Raroia and not eaten. The family is most abundant in holes in the coral on the coralliferous outer bench, in Garve Pass and around lagoon coral heads. The young are abundant on shore reef flats, especially around the outside of.the atoll. Cephalopholis argus is by far the most predominant of this family, particu- larly on shore reefs. Of the eight main marine fish zones (see ecology) the Serranidae are prevalent in all but the surge channels. rug The natives spear sea basses with single and four-pronged spears, es- pecially fishing: for them along the shore edge and in the inter-islet chan- nels. Serranids readily bite on a hook baited with all kinds of meat. This cee ilies method is most successful along the coralliferous outer bench, in Garue Pass and around lagoon coral heads. ; Grammistes sexlineatus (Thunberg).--Dorsal fins separate. Distinctive mark- ings yellow and brow longitudinal stripes. Stations: 11(2), 22(5), 45(2). Habitat: Moderately common in west shallow inter-islet channels; rare in neighboring regions (tide pools near shore on outer reef flat and lagoon shore reefs); not noted elsewhere. Raroian name: none. Variola louti (Forsk&l).--Pectoral fin with yellow edge. Small red dots over head and body. All fins but pectorals prolonged. Life colors of speci- men from station 62: Ground color bright red over head and body, with brilliant vermillion to bluish-red spots. Head covered with spots above horizontal level from lower edge of mouth. Red spots present on upper lip, none on lower. Lower half of head yellow; pale blue behind jaws. Iris red, blackish above, and a narrow black streak below. Body bright red, reddish-black above to pure red below. Body covered with red spots that are paler than the background color; these spots extend ventrally to a horizontal from lower edge of pectoral fin. Belly and breast pure red without markings. Spinous dorsal ground color dusky black with numerous red, purple and blue spots on the rays and membranes. Soft dorsal reddish-black, covered with numerous bright red spots which fade out on posterior edge to pale yellow. Edge of soft dorsal black. Caudal fin reddish-black, middle rays shading to pale yellow at tip. Anal fin reddish-black, three to seven red spots on basal half of each membrane; distal edge of fin shades to black, posterior edge shades to hyalin. Pelvic fins edged in black anteriorly shading to red on second soft ray. Remainder of fin pale red. Pectoral fin broadly edged by a pure yellow band. Remainder of ‘fin shades from black tinged with red to dusky red at base; a few red spots on basal part of fin. Stations: 33(1), 62(1). Habitat: Noted in Garue Pass and on coralliferous outer bench; well known to the natives but apparently not abundant. Raroian name: HOKAKOKA AND MAERE. Cephalopholis urodelus (Schneider).--Fourteen soft anal rays. Distinct V present on caudal fin that is bordered with black above. Two black spots on top of caudal peduncle. Red spots over side of body. Life colors of 150 mm. specimen from station 61: Head brown, paler on cheeks and opercles. Iris vermillion with a broad black ring in the center which shades out on the anteroventral margin. Lips brow. Upper half of preopercle with orange spots. Occiput and nape brow, with a few dark M7 Ee brow spots on the side. Body anteriorly .brom above shading to red- orange posteriorly and ventrally. Sides of body covered with bright red to orangish-red spots that fade out above anteriorly; breast yellow, orange above pelvic fins, Darsal fin brightly colored; spinous dorsal with tips of membranes bright ‘red; remainder of fin brom with suffuse orange spots. Soft dorsal .with.a broad brilliant red horizontal band; remainder of fin brown with two longitudinal series of red to reddish- orange spots. Caudal peduncle superiorly with two black spots. Caudal fin bright red, margins of upper and lower rays hyalin; middle of fin _ with bright red. membranes and paler rays. Anal fin same as. Soft dorsal except last rays fainter red. Pelvic fins orange-red, shading to pale orange posteriorly and basally. Pectoral fins orange shading to’ yellow- orange posteriorly. ep ba fas Stations: Bikey 66(1). Hebi tet: Dalen on coral head on east side of lagoon and on shore ;:... - net of west side of lagoon. Rare at Raroia, but much more common at Takume. a: Raroian name: none. Cephalopholis argus (Bloch) .--Caudal fin rounded. Anal fin with nine ‘soft. rays. Body brownish-black with black-ringed blue spots. Life colors of adult from station 62: Head and body dark brown with indistinct narrow vertical brom stripes, particularly evident over anal fin. | Head, body and fins covered with close-set blue spots. Head darker brown than body.’ -- Iris dusky gold, narrowly pale blue next to iris; a dusky black crescent «-~ present above lens. Body solid brow overall. Blue spots evenly scattered over body except sparse below pectoral fin. Spinous dorsal same as body... - Soft dorsal solid black shading into brown basally. Dorsal fin covered with blue spots as on body, except’ more concentrated distally. Caudal °° and anal fins same as dorsal. Pelvic fins same as soft dorsal except na lacking hyalin edge and spots are not more numerous than on body. Pec-— toral fin basally black shading distally to brom; edge of fin hyalin; «°° basal half of fin covered with numerous blue pak a more Ce el than on body; outer half tinged with ‘red. arian Stations: 8(7), 9(3), 10(2), ene 14(12), sD 20(17), 23(4) 5 28(5), 33(2) , STKE) 5 60(1), 61(18), 62(3), 65(10). Habitat: Very abundant throughout the lagoon around coral heads and shore reefs, and in Garve Pass. . Young abundant in tide pools near . shore, on outer reef flat and in inter-islet channels. Largest adults on coral heads in center of lagoon. ; eee Raroian name: HAPUTU. Paracanthistius maculatus (Bloch).--General color red without bands. Small ~ black ringed blue spots over head, body and vertical fins. Life colors of specimens from station 52: General color solid brick red over head, ore body and caudal fin; reddish-black above. Head with numerous tiny black ringed blue spots. Jaws without markings. Iris gold, narrowly edged in reddish-gold, black anteriorly. Spots: present over all of head except jaws. Body with similar spots:as on head. Ventrally spots become solid black. Spinous dorsal reddish=black with numerous obscure blue spots. Caudal fin same as body, narrowly edged in white. Pelvic fins reddish- black, first rays black, edge of fin dusky black. Pectoral fins similar to pelvic. Anal fin same as body except for numerous ‘Spots narrowly edged in white. Stations: 19(1), 2602), 33(13), 42(1), 52(2); 54(2), 60(1).. Habitat: Very abundant in Garve Pass and neighboring regions in deeper water. Moderately common on coralliferous outer bench. Poison- ous. Not eaten by natives. Raroian name: KEKEREVAE (ena TONU e(aeeibeye Epinephelus merra Bloch.--General color white with numerous ena black spots on head and | body. All fins white edged. Stations: 11(6), 14(16), 20(6), 22(12), 35(1), 47(4), 55(33), 65(1). Habitat: Almost entirely restricted to inter-islet channels among coral and ro and rocks. Raroian name: GAREA. Epinephelus hexagonatus (Cuvier and Valenciennes) .--Life colors of specimen from station 44: Head and body brownish white ventrally, green dorsally, covered with large brom spots that do not tend to form lines or sad- dles. Brom spots present in maxillary groove. First dorsal edged black; basally fin brow with reddish-browm spots. Soft dorsal edged white with a submarginal black line; remainder of fin covered with close-set brown spots. Caudal fin same as body, edged with a white line. Soft anal edged in white; submarginally with a broad black stripe; remainder of fin similar to hody; membrane between first two spines hyalin, other spinous membranes dusky bleck. Pectoral fin pale brown with small blackish spots on rays; basal half of fin with light blue lines around spots. Pelvic fins covered with brown spots; edged white. Stations: 7(4), 8(3), 9(5), 10(1), 11(8), 12(1), 14(21), 23(5), 28(7), 33(1), 42(2), Aé(1), 45(6), 49(7), 50(3), 61(8), 65(4). Habitat: Abundant at Raroias; occurs where there are hiding places under rock end coral. Raroian name: VEVE. Epinephelus maculatus (Bloch).--Coloration mottled over head and body. Black saddle present on caudal peduncle. Lower half of head and body covered with tiny dark spots. -Th- Stations: 7(4), 8(3), 9(2), 22(2), 23(3), 28(5), 31(1), 33(2), SoU, A2(1):5. 4301), 16(4) , AN), SSCL) ar Habitat: Very abundant at Raroia where there are hiding places under rocks and coral. This species and Cephalopholis argus are the dominant sea basses around lagoon coral heads. It is one of the main food fishes of the natives. Raroian name: TITIRIRI (small), KUFARUFARU (medium), KITO (big), KINAO (fat). Epinephelus tauvina (Forskal) .--Head and body covered with reddish-brom spots. No stripes on dorsal fin. Life colors of specimen from station 43: Ground color overall greenish-white. Head, body and fins covered with reddish-brown spots. First dorsal edged with a dusky bromish- black line. Second dorsal similar except with a very narrow white edge. Caudal fin narrowly edged in white; submarginally bordered by a black band. Anal fin similar to soft dorsal. Pelvic fin edged in brownish- black; first pelvic ray same color as other rays, not white. Pectoral fins similar to caudal fin. Stations: 11(1), 28(1), 33(2), 43(1), 47(3), 62(1). Habitat: Moderately common at Raroia. It is most abundant on top of lagoon coral heads among dense algal growth and coral. Scattered individuals are found in inter-—islet channels and Garue Pass. Raroian name: FAKETA, Epinephalus socialis Gunther.--Four black spots present on back. Head, body and fins covered with suffuse small light brown spots. Light area be- tween spots a dull light blue. da Sistine: 125) 0 VACLA: pane M73), 20(1)’, fetal 15(6), 58(9), Sc Habitat: gh in erotic channels, Garue Pass, and on outer- reef flat. Raroian name: KOKOTIKA. Epinephelus sp. one.--Solid red overall, with oblique white lines in tail. General color a uniform brilliant red. No markings on head and body except for numerous small darker red spots. Iris red, distally reddish- black. Dorsal fin same as body except soft dorsal with numerous red spots. Caudal fin with central region reddish-black; oblique lines above and below through fin pure white; upper and lower rays edged hyalin. Anal fin same as soft dorsal except for dusky black band down middle of fin and pale spots basally on posterior rays. Pelvic fins red, the first soft ray edged in black. Pectoral fins solid red. =15— Stations: 62(2), 66(1). Habitat: Rare at Raroia. Taken by hook and line off bottom in Garue Pass and on coraliiferous outer bench of west side of atoll. Raroian name: TUKOROKORO. Genus one species one.-~-Characterized by long flowing fins. General color é black with numerous blue spots. Coloration of specimen 110 mm. in length from station 13 two months after preservation: Blackish-brown to solid black overall with bright blue spots covering head and body. Iris iridescent dark blue. Body and head with each scale having a blue spot in the center. Branchiostegal membranes brown with blue spots. All fins but pectorals black, covered with closely spaced blue spots. Pectoral fins hyalin. Stations: 13(1). Habitat: Only one noted at Raroia on lagoon coral head. Raroian name: none. FAMILY PSEUDOCHROMIDAE Two species were recorded at Raroia and perhaps others will be discov— ered when the unworked material is examined. The family lives in cavernous surge channels and in hiding places in coral and under rocks in the lagoon. The natives believe they are the young of sea basses and have no name for them. Aporops bilinearis Schultz.--Two lateral-lines on side of body. No black spot on operculum. Light spots on brown body. Life colors of speci- mens from station 58, 45 mm. in total length: Head and body uniform brown, shading to pale brown below. Head greenish before eye. Body with suffuse darker brow spots above pectoral fin. Dorsal fins browm at base. Outer unscaled portion pale black, shading to hyalin edge on posterior soft rays. Caudal and anal fins pale black with hyalin edge. Pectoral fins dusky black. Pelvic fins hyalin. In 60 mm. specimens head and body have darker brow spots evenly scattered but without any definite arrangements. Vertical fins dark browmish-black, very narrowly edged in hyalin. Stations: 17(8), 22(1), 28(1), 57(1), 58(10). Habitat: Most abundant in cavernous surge channels, hiding deep in cracks and under rocks. Occasional individuals observed in lagoon deep under rocks and coral. Raroian name: none. Pseudogramma polyacanthus (Bleeker).--Black spot on operculum. General color mottled brow, Life colors of specimens from station 57: Overall dark brow, irregularly mottled with a darker brow reticulated pattern over all of body and upper half of head. Large dark brown spot on up- per edge of operculum. Suffuse brow lines radiate from posterior border of eye. Iris dark brown, lens ringed with gold. Jaws pale eéx- cept for dusky brown tips. Lower surface of head pale. Membranes of spinous dorsal brow; membranes of soft dorsal bromm, fading to pale. _browm at tips. Caudal fin solid brown. Anal fin brom, fadiie o pale brown at tips. Pelvic fins dusky brown at base, membranes hyalin. Pec— — toral fins dusky brom. =~ : : Stations: 7(2), 57(2). Habitat: Observed only in lagoon in same habitat as Aporops. - Raroian name: none. FAMILY LETHRINIDAE Mu & Snappers While only three species of lethrinids appear to occur at Raroia, they are very abundant around the islets inside and outside the lagoon, and pro- vide an important food for the natives. The natives know the lethrinid names without hesitation and are well aware of where they occur around the atoll. Monotaxis grandoculis (Forskal).—-Characterized by three broad solid black saddles on back in life that immediately fade after death. Life colors of specimens from station 61: Head olive grey. Interorbital greyish- yellow. Tip of upper jaw black, tip of lower jaw dusky black. Iris Silver, bordered by black above and posteriorly, and by two dusky black spots below. Body with three broad solid black saddles on back that shade out to yeliow on mid-lateral region. Between biack bands pal blue. Pectoral fin base solid black. Breast and belly dusky olive, blackish between pelvic and pectoral fins. Caudal peduncle dusky yel- low. Spinous dorsal dusky black, shading into a paie red margin. Soft dorsal anteriorly like spinous dorsal, but posteriorly pale red to hy- alin. Caudal fin basally pale yellow, tips of outer rays pale orange- red, middle rays faintly streaked with black longitudinal lines at middle; otherwise fin hyalin. Anal fin pale red anteriorly, shading to pale yellow posteriorly. Pelvic fins hyalin. Pectoral fins rays pale orange, membranes hyalin. . Stations: 27(1), 31(1), 61(3), 66(2). Habitat: Very restricted in distribution. The young are almost entirely confined to around lagoon coral heads near the eastern side La of the atoll. The adults are rarely seen in the daytime, but come into the surge channeis and onto the outer reef flat at night to feed. Raroian name: MU. Lethrinus mahsena (Forskal).-—Cheracterized by a red mouth, by uniform body coloration, and by yellow pectoral fins. [Life colors of specimen no. 2306 from station 33: Ground color of body olive brown to grey; edge of each scale (particularly anteriorly and ventrally) darker brom. Intire body colored evenly, back slightly darker. Head brown, squama- tion on operculum same color as body. Lips reddish in the folds. Inside of mouth bright red. Iris golden brown, variously mottled brom, blue, and silver. Upper edge of eyeball shaded brownish black. Spin- ous dorsal dusky brown, edged in pale red. Soft dorsal fin yellowish- brow basally, pale red at edge, last rays pale red almost to base. Caudal fin brown, a pale yellow vertical band across middle, edge of fin pale red. Anal fin yellowish—brow, edged in pale red. Pelvic fins with yellow rays; anterior soft rays dusky black; membranes hyalin. Pectoral fin with rays yellow and membranes hyalin. Stations: 24(2), 33(1), 45(2). Habitat: Northern, eastern and southern inter-islet channels; abundant in Garuve Pass. Prefers channels with swift moving water; occurs in schools. This species is one of the more important food fishes. It is speared under rocks and also caught with hook and line. Raroian name: TAMURE. Lethrinus rostratus Cuvier and Valenciennes.--Characterized by a long snout, bright red throat inside mouth, and red dorsal fin. Life colors of specimen from station 60: General color olive white, somewhat greenish, without markings on head or body. Iris silver. Spinous dorsal red- dish, shading to brighter red at edge. Each spine with two suffuse brow spots. Soft dorsal pale red, with two suffuse brown spots on each ray, membranes with approximately three large white spots that extend onto the rays. Caudal fin generally pale, same color as body, with reddish posterior edge. Anal fin pale, outer edge of membranes red. Pelvic fins white, outer half of membranes red, tip blackish. Pectoral fins white, membranes hyalin. Stations: 31(1), 60(1). Habitat: The natives say this species is common at Raroia al- though only a few individuals were observed during our study. It is said to be most abundant in deeper water (20-40!) in Garue Pass and along the outer reef edge. Raroian name: MEKO. 73 FAMILY LUTIANIDAR Snappers Seven species of this family were taken at Raroia, approximately half of the species that occur there. More were not taken because of inadequate gear, although considerable effort was expended trying to obtain additional forms. The family is most abundant in deeper water around the outside of the atoll, and along the eastern shore in the lagoon. Seldom are more than scattered individuals observed in the rest of the lagoon. Large schools occur in the closed off inter-islet channels on the east side of the lagoon...::.: The natives have a different name for each species and use practically all the snappers for food. The large adults of Lutisnus bohar are poisonous and natives know within a few inches of length when an individual is poi- . sonous or not. No other lutianids are poisonous, according to the natives. Lutianus monostigma (Cuvier).--Characterized by a black spot behind pectoral fin on lateral line. Stations: 11(2), 24(1), 27(1), 30(1), 35(4), 60(1). Habitat: This is an important food fish at Raroia, widely distri- buted in deeper water -around coral heads and shore reefs. It was only once observed in schools, but individuals were often noted hiding in coral heads and: in schools of Lutianus kasmira and Lutianus marginatus. - Raroian name: PUAKI (small) , .TERO (medium), PARAI (large). Lutianus kasmira (Forsk&1).-—Characterized. by a yellow body with three longitudinal bright blue stripes... phataous: 35(1)',. S7C1) 38 (3) . Habitat: Entirely confined to the east and northeast side of the atoll in the lagoon. Large schools were noted in incomplete inter- islet channels around coral heads. Only a few adults were observed on eastern lagoon coral heads. Seldom used for food by the natives since they rarely bite on a hook and are difficult to spear. Raroian name: HUTIHUTI and. TOHARE. Lutienus marginatus (Cuvier).--Characterized by a yellow body with numerous narrow yellow stripes on lower half of sides; caudal and dorsal fins edged black. Life colors of specimens from station 47: Head yellow on gill cover; anterior to squamation grey. Lower scales on preopercle and opercle with light orange spot in the center. Body yellow, dusky on back shading to pure yellow on belly. Caudal peduncle yellow. Spinous and soft dorsal fins with a submarginal black band that becomes darker on the soft dorsal. Fins narrowly edged hyalin. Caudal fin _79- black, narrowly edged hyalin. Anal fin pure yellow as are paired fins. Lower half of body with yellow-orange on each scale, most in- tense below and before pectoral fin, forming into longitudinal lines posteriorly to caudal peduncle, Stations: 24(1), 35(2), 37(1), 38(4), 47(8), 60(12), 65(1). Habitat: Scattered individuals throughout lagoon around coral heads; abundant in large schools in closed off inter-—islet channels around dead coral and beach rock caves. This is an important food fish for the natives and is taken by spears. Raroian name: MAGU MAGU. Lutianus gibbus (Forskal) .--Life color notes of specimen from station 44: No spots. Red present below on head. Dorsal fin orange as other fins. Head and body dark above, olive brown, shading to reddish below, par- ticularly on throat and head. Indistinct yellow longitudinal lines present on mid-side of body on the middle of each scale row. Fins without markings, all. orange. Dorsal browmmish except for orange tips. Vertical fins scaled far out on fins. Stations: 44(1), 47(2). Habitat: Common off outer reefs of atoll. A few adults noted in lagoon around coral heads and on southem sand flats. Raroian name: PUAKI (small), TERO (medium), PARAI (lerge). Lutianus bohar (Forsk21).--Characterized by bright red head and body and blackish fins and back. Black mark behind maxillary. Black axillary spot on pectoral fin. Life colors of specimens from station 33: Head brilliant red tinged with orange, particularly brilliant on under- Side. Upper half of preoperculum and operculum tinged with green. Top of head from snout to dorsal fin dark reddish—black tinged with green. Maxillary dusky reddish—black where normally hidden by groove. Upper lip dusky reddish-black. Streak behind maxillary dusky reddish- black. Groove before eye bright red. Iris brilliant golden reddish- orange. Tip of lower jaw dusky black. Body reddish-black above shading to brilliant reddish-orange below that in life is often pur- ple, pink or tinged with blue. Belly purplish-red. Spinous dorsal solid blackish-red. Soft dorsal reddish-black, with black edge. Caudal fin bright red, narrowly edged in black. Anal fin red at base, anterior edge black, outer wunscaled portion reddish—-black. Pelvic fins solid red, outer region of fin anteriorly black, narrowly edged in black posteriorly. Pectoral fins with black spot at upper base; upper half of fin reddish black to solid black at tip; lower haif and base red. Stations: 33(8), 62(1), 63(1). -80- Habitat: Abundant around outside of atoll on coralliferous outer bench and in Garue Pass. The natives catch them for food with hook and line and occasionally by Spear. ‘the large adults called MERO MERO are poisonous. ; aezolan nane name: TAGAU (very small), TATATATA (niedium), MERO MERO large Lutianus sp. one.--Characterized by broad black region across caudal peduncle and caudal fin and a red stripe on dorsal fin. Life colors of speci- mens from station 61: General color of head and body greyish-white, somewhat olive, with a faint greenish tinge. Head without markings, light. Iris gold, orange-gold above. Body tinged with red on breast, belly, and region atove anal fin. Caudal peduncle posteriorly covered with a broad black region; caudal fin basally black; posterior edge pale yellow-green to pale black. Dorsal fin edge white, submarginally black; black bordered by a suffuse red longitudinal band that extends through spinous dorsal and ends on last soft rays. Basal portion of spinous dorsal white anteriorly, black posteriorly on each membrane. Soft dorsal basally black. Anal fin red; anterior half edged in white, submarginally with a black and red band. Pelvic fins anteriorly edged white, outer region of first two soft rays black; remainder of fin pale red, most concentrated on anterior rays. Pectoral fins pale-yellow on rays, with upper basal region pale red; membranes hyalin. Stations: 61(5). Habitat: Seldom observed around coral heads in lagoon. A small school was hiding in the coral at euene 61, a coral head in the lagoon of Teremu Islet. Raroian name: TAEFA. Aphareus furcatus (Lacépéde) .--Characterized by browmmish—-blue-grey ground color, orange pectoral fins, and elongate last dorsal and anal rays. Life colors of specimens from station 62: Ground color brownish, iri- descent pale blue and grey mixed together on scaled areas. Head solid brown on vnscaled portions. Lower jaw brownish with iridescent light blue. Iris silver distally to gold internally. Dorsal fin yellow-brom, soft rays with faint reddish edge. Caudal fin edged red, remainder pale black. Anal fin bright yellow tinged with orange; posterior rays and membranes pale red, all of fin edged pale red. Pelvic fins pale yellow on membranes; eniemion rays pale red, tips of first rays white. Pec= toral fins pale eee ‘ ee 62(10). Habitat: Schools in deep water over edge of outer reefs. The natives occasionally take them by trolling. Raroian name: TAHAKARI. -Sil— FAMILY KUHLIIDAE Ahole Only one species recorded at Raroia and the natives say there is only one kind. It is occasionally used for food, but usually discarded. Kuhlia sandvicensis (Steindachner).--Fifty-five scale rows between opercu- lar margin and base of caudal fin. Caudal fin plain with posterior margin black. Life colors of specimens from station 45: Head and body Silver, blackish above. Dorsal fin membranes hyelin basaily, dusky black at tip. Caudal fin white with posterior margin black. Anal fin hyalin, except for scattered melanophores medially. Pelvic and pec- toral fins hyalin. Iris silver, blackish at posterior margin of lens. Stations: 16(1), 27(7), 45(7), 58(1), 65(10). Habitat: Reef flats on west and southwest side of island. Feeds on reefs at low or high tide day and night. Raroian name: AHORE. FAMILY APOGONIDAE Cardinal Fishes Approximately 12 species of cardinal fishes were taken at Raroia, and probably only a few additional species occur there. All species observed were taken. This group is found wherever there is coral, and is most abun- dant on the lagoon coral heads and west shore reefs. The cardinal fishes are inconspicuous, usually remaining in the holes in the coral, and are completely ignored by the natives, who have no name for them. All speci- mens were taken by rotenone, this being the only effective means for col- lecting them. Very little restrictions of distribution to definite sections of the atoll were observed, nor were they particulerly more abundant in one area than another. They prefer the quieter waters of the lagoon and are much less in evidence on the outer reefs or in the inter-islet channels. The following data for the cardinal fishes have been prepared from only a small part of the total collections, and do not include all the species taken or all the material for each species. While the fishes of this family are eaten in Tahiti and are sold in the central market at Papeete, they are not eaten at Raroia. In Tahiti they are taken by common beach seines on the coral flats next to shore and on the outer reef flat, but seines are not now used by Rarcians. Apogon ceramensis (Bleeker).--Color notes prepared at station 9: Ground color light with a single black streak from tip of snout through eye -82— to caudal fin. A black spot present. above lateral-line on base of caudal fin. | Anterior three dorsal spines edged in black. A fine narrow black line, near base of soft dorsal rays. Stations: 9(5), 13(2), 28(1). Habitat: This species is especially abundant on the tops of lagoon coral heads in shallow water, living in holes in the coral rubble and holes in the sand. Apogon doryssa (Jordan and Seale).--Distinctive colors orangish red, particu- larly at the fin bases. No markings except for scattered melanophores on mid-dorsal line, mid-lateral region of caudal fin base, and on head. Stations: 7(3), 9(4), 12(1), 58(1), 61(1). Habitat: One of the most abundant species at Raroia, occurring _practically wherever the family is found. Apogon aroubiensis Hombron and Jacquinot.-—-Distinctive colors dark brown with light blue stripes. Stripes becoming reddish on head. Fins reddish brow with suffuse blackish browm stripes. Color streaks do not continue on caudal fin. Life colors prepared from specimens of station 57: Head and body with blackish brown longitudinal stripes separated by pale blue iines. Lines on head become narrow and pure white. Brown stripes become reddish. Iris pure silver with greenish tinge. Spinous dorsal with rays dusky black, reddish on posterior margins of each spine. Soft dorsal blackish brow at base, distally pale red. Caudal fin with median rays dusky, remainder of fin pale yellow. Anal fin with brownish black base, fading out anteriorly. Remainder of fin pale red. Pelvic fins red basally shading out to hyalin distally. Pectoral fin pale red at base shading out to hyalin distally. , Stations: 7(6), 28(3), 57(2). Habitat: Abundant at Raroia, particularly on the lagoon coral heads. sd thts Apogon isostigma (Jordan and Seale).--Distinctive colors consist of black spot on operculum; approximately nine longitudinal rows of black spots. Life colors of specimens from station 49: Head and body gen- eral color dark browmm. Head with two black lines radiating from posterior edge of eye. Iris pale yellow. Black spot on operculum narrowly edged in white. A longitudinal dark brow line above and below the opercular spot. Hach scale on body tends to have. a. black spot at its base, forming longitydinal lines. First dorsal solid brovm. Second dorsal brownish at base and orangish for most of fin. Membranes hyalin distally. Caudal fin very dark, brovm to blackish, rays orangish. Upper middle and lower middie rays pale orange at edge. apa Anel fin rays orangish-browm, membranes browmish, darkest medially. Pelvic fin with orange-browm rays and pale browm membranes. Pectoral fin with pale reddish brow rays and hyalin membranes. Stations: 28(4), 49(4). Habitat: Observed only in the lagoon where it is moderately common. Apogon bandanensis Bleeker.~-Distinctive colors silver with dusky vertical stripes hardly evident. Conspicuous band across caudal’ peduncle; another from lower edge of eye. Tip of first dorsal dark. Stations: 7(16), 9(3), 13(1), sly Gly a 23(16) , 49(3). Habitat: Very abundant on lagoon coral heads. Apogon frenatus Valenciennes.—-One of the largest cardinal fishes. Third dorsal spine longest. Black streak across soft dorsal and anal. Black blotch on caudal peduncle. Stations: 7(1), 9(4), 27(1), 28(6), 42(1), 65(2). Habitat: Solitary individuals around coral kexws Apogon auritus Valenciennes.--Black spot on operculum bordered in whice. Body head, and fins reddish-orange with about eight sufluse ve tical bands on body. Stations: 7(5), 13(1). Habitat: Moderately abundant around lagoon coral heads. Apogon sp. one.~-Apparently this is a very small species. Ground color chocolate brown overall. Head with two oblique dark brow stripes. Spot on operculum. Fins blackish. First dorsal with a black spot on posterior membranes edged in white. Edge of soft dorsal, soft anal, and caudal edged in white. Pelvic fins black. Pectoral fin with a brownish-—black vertical stripe basally. Each scale row on body darker browm centrally, tending to form longitudinal stripes. Stations: 17(3). Habitat: This species lives in the sea urchin holes on the rein 18 line ridge, outer reef west side of atoll. Paramia guinguelineata (Cuvier).--Canines in upper jaw. Six bands on each side, including mid-dorsals and mid—ventrals. Flack spot at caudal fin base. Colors of specimens from station 23 one month after preser- vation: Stripes browmish-black. Background color yellowish-white, tinged with silver. Iris browmish—black above on border, remainder —84- sSilvery-white. Fins gate except for dusky black on first spines of dorsal fin and outer caudal rays. Yellow particularly concentrated around caudal spot. FAMILY KYPHOSIDAE Rudder Fishes Only one specimen was seen the entire period at Raroia, although the natives claim this group is rather common around the atoli. Kyphosus sp. one.--Characterized by two longitudinal black bands above and below eye. Body descaled but apparently upper half of body solid black, bordered by a golden longitudinal band. Breast and belly pure white. Dorsal and anal fins solid brown. Pectoral fins dusky brow. Pelvic fins brownish red anteriorly, pale yellow on last rays. Caudal fin brom. Stations: 45(1). Habitat: Not know. A native speared the one from ehation 45 along the outer reef in a surge channel, but the ra. ve: were vague as to where it occurred. Raroian name: TAHAKART, FAMILY SIGANIDAE. ° One species was taken at Raroia which the natives say is uncommon. None were observed in the field. Their paucity at Raroia is strange, since the group is so abundant at Tahiti, forming an important native food there. The natives eat them at Raroia, but seldom find then. Siganus sp. one.-~Body bluish brown. Head and body covered with yellow spots. Fins pale, marbled brown. Colors of specimen from station 64: (speci- men described before preservation but not seen alive). It appears that the ground color is bluish-browm. Head and body covered with closely spaced yellow spots. Iris ringed with gold, dusky black, and pale blue rings. Head bright blue, as are breast and belly. Spinous dorsal and spinous anal with brow rays; membranes marbled brown and hyalin. Soft dorsal and anal with rays brom and membranes hyalin. Caudal fin browm with a broad white posterior edge. Pelvic fin rays brow; membranes blue with pale brown Sell hea fins:with pale yellow rays and hyalin membranes iy ean ei li ea -85~ Station: 64(1). Habitat: Uncommon at Raroia. Speared along shore on west side of lagoon. Raroian name: MARAVA. FAMILY PEMPHERIDAE Catalufas The one species taken at Raroia was found only in deep cavernous surge channels next to the coralline ridge. While only a few would be in each channel they were always found in this type of habitat. The natives have no name for this group. Pempheris otaitensis Lesson.--Pectoral fin base and first rays of dorsal fin black. General head and body color brassy. Life colors of specimens from station 45: General color brassy shading into black dorsally and along anal fin origin. Iris silver. Dorsal fin with anterior edge black and upper third of anterior rays black; remainder of basal por- tion of fin white shading into hyalin at tip. Caudal fin with upper and lower rays black, tips of all rays black. Base of caudal rays pale black, middle third of fin white. Anal fin black at base, dusky white on first rays and middle third of fin, tips of all but first rays black. Pelvic and pectoral fins pale; scattered black pigment spots on pelvics but none on pectorals. Stations: 45(5). Habitat: Typical of deeper cavernous surge channels 20-30 feet deep in region of Oneroa. FAMILY CIRRHITIDAE Hawk Fishes This family is typical of the surge channel zone and rarely was found outside this area at Raroia; and then only at the fringes of this habitat. Eight species were taken, and it is doubted that other species occur at Raroie, except the genus Cheilodactylus which was observed around lagoon coral heads, Cirrhitus pinnulatus was far more abundant than all other cirrhitids combined and occurred in all surge channels. At least this species is nocturnal, coming onto the coralline ridge and outer edge of the reef flat to feed at night, even at low tide. . Great numbers can be seen on the reef edge either at low or high tide at night, and are almost as numerous as -86- the squirrel fishes. The natives will eat Cirrhitus pinnulatus, but usually do not take them. They know little about the other cirrhitids and rarely see them. The natives have a general name for all cirrhitids--PARATUKI, but the "real PARATUKI" is Cirrhitus pinnulatus. Cirrhitids live in cracks in the coral and surge channel walls and un- der rocks on the bottom of the troughs. Except when affected by poison, not a single individual will be seen when observing in the daytime by skin diving or look box. At night when lights are directed down into the channels hundreds will often be seen moving about on the bottoms of the troughs. Cirrhitus pinnulatus (Schneider).--Life colors of specimens from station 58: Head and body with reddish-brown spots. Head with a light blue back- ground color from snout posteriorly to opercular edge below eye. This region has longitudinal series of orange-brom lines, irregularly ar- ranged and wavy. _Upper half of nead colored and marked the same as body. Iris black, golden red anteriorly and posteriorly. Lips and lower jaw crossed by black mottled lines; otherwise lower section of head pale yellow to greenish-yellow. Body with a spot extending ventrally to a parallel from lower edge of pectoral fin. The region above is mottled with large brown blotches, which vaguely tend to form bands. Breast. and belly white. Region before pectoral fin spotted like upper half of body. Spinouvs dorsal edged hyalin. Tufts at tips of fins white. Base of dorsal fin dusky red. Middle of membranes with a longitudinal series of dusky red spots, one in each membrane. Remainder of fin pale yellowish-brom. Soft dorsal colored same as bedy. Rays with oblique series of three or four reddish-brown spots that basally continue on the membrane. Caudal fin rays with four or five reddish-brown spots that extend slightly on the membranes. Upper and lower rays yellowish. Anal fin with 3-5 reddish~browm spots at base; base of fin yellow to dusky red anteriorly. Pelvic rays yellow to pale orange; hyalin at tip; membranes hyalin. Pectoral fins mottled like body at base; one or two series of red spots on middle of rays. Membranes hyalin. Stations: 17(30), 22(1), 26(56), 45(13), 58(42). Habitat: This species is the most abundant cirrhitid in the surge channels, each trough usually having approximately 100 young to adults. At night it comes up on the coralline ridge to feed; it even does this at low tide when it flops around from pool to pool between waves. The natives will eat this species but do not prefer it over other fishes. Paracirrhites forsteri (Schneider) .--Side of body with pale longitudinal stripe. Large black spots on upper side of body. Head reddish. Dif- fers from typical forsteri by lacking black band above white stripe. Stations: 17(1), 22(1), 26(1), 45(4). Habitat: There are usually 2-6 individuals in each surge channel on the west side (most specimens were seen in deeper water from 10-25', and none were collected in less than 12' of water). 87 Paracirrhites cinctus (Gunther).--Characterized by vertical brow bands on body. Life colors of specimens from station 58: General color light brown with brow markings. Head brow above fading to pale brown be- low. Head orangish-browm on snout with an orange-brown band from eye to upper jaw. Lips dusky browm, upper lip black at symphysis. Iris golden red, dusky black around anterior upper and lower sides. Body with five slightly oblique broad vertical. brow bands that are darkest dorsally. Breast dusky brown. Belly wnite. Spinous dorsal evenly pale brownish-black, tip hyalin. Soft dorsal brownish-black at base, rays dusky black; outer portion of membrane hyalin. Caudal fin rays dusky black. Membranes hyalin. Anal fin with dusky reddish-black rays, membranes hyalin at tip. Pelvic fin rays pale orange-brown across middle, membranes hyalin, Pectoral fin rays dusky brown, tinged with red; lower rays pale black. Membranes hyalin. Stations: 26(4), 45(7), 58(6). Habitat: This species is found only in surge channels that are more protected. Rather rare at Raroia, being found in only certain surge channel area. Paracirrhites arcatus (Cuvier).--Supraocular loop reddish-orange. Yellowish- orange to reddish-orarge stripe extending down side of back. Stations: 17(5), 26(1), 58(1). Habitat: Moderately rare. ly taken at the deeper open end of the surge channels. All of the field observations note that they were seen hiding in cracks on the side of the channels, blending closely with the encrusting algae. Paracirrhites sp. one.--Characterized by a general dark purplish-brown color and a pure white spot on body behind pectoral fin. Life colors of specimens from station 45: Ground color over head and body a dark pur- plish-brown. Head color uniform except darker behind eye, on inter- orbital and at preopercular margin. Lips reddish. Gill membranes black medially at throat. Body purplish-browm with dark brown spots-—- irregular above, smaller and arranged in lofigitudinal rows below. A pure white ;spot covers parts of 9 scales and all of 2 lateral line scales at 16th and 17th lateral line scales, behind and above pectoral fin. Body coloration extends half way out the caudal fin. Spinous dorsal fin brownish-black with reddish-brown spots extending up to membranes. Soft dorsal with body coloration and spots on scales at base; remainder of membrane hyalin; soft rays brownish-black. Caudal fin with posterior third of membranes hyalin. Rays brownish to brownish-black at tips. Anal fin with spines and rays brownish-black. Membranes of soft dorsal black anteriorly, hyalin for posterior mem- branes. Pelvic fins with rays browmish black, membranes hyalin. Pec- toral fin with upper branched rays reds; lower simple rays light brow; membranes hyalin. —-88— Stations: 45(3), 58(1). Habitats ubarn at. Raroia. . ALL individuals were taken in narrow surge channels at 8-10! depth. The natives have never seen ‘this spe- cies before and apparently it lives deep in the cracks in the coral. parererrnites sp. two.—This species is similar to arcatus but lacks the broad white body band and the outer caudal rays are prolonged. Life - Colors of specimens from station 66; overall color solid brom. Head ‘not lighter. Head with reddish-orange nostrils tipped with brom, 3 orange-red bands at lower opercular margin that may be divided into two spots; supraocular loop orange-red, narrowly edged with black and blue lines. Inside of loop same brow as remainder of head. [Iris narrowly gold to metallic red next to lens. Outer portion dusky purple to black. Spinous dorsal fin brown, shading to narrow black edge. Middle of membrane brownish~-red to reddish-orange. Soft dorsal reddish- brown on basal half, outer half of membranes hyalin. Rays reddish- brown to orange-red distally. Caudal fin edged in black dorsally and ventrally. Upper and lower prolonged rays orange-red, shading to more brommish-red and reddish-orange on middle rays. Membranes uniform dusky black. Anal and pelvic fins brom. Pectoral fin brown, upper rays tipped with red. Stations: 66(2) Habitat: Only a few specimens were seen in the bottom of surge channeis. Cirrhitoides bimacula Jenkins.--This is a small hawkfish with a pointed snout, two distinctive black spots and-reddish-brown vertical stripes. Stations: 17(1), ‘58(1)ee% Habitat: Uncommon at Raroia, ‘preferring the narrower, deeper surge channels. Apparently it lives under rocks on the bottom of the channels. Genus (undet.). sp. one.--Nothing like it in Fowler (1928), Schultz (1943)> or Tinker. Overall a brilliant solid red, including eye. Interorbital ., and postorbital region blackish. Upper half of body blackish, as are .... posterior spinous dorsal rays. Tips of last soft dorsal rays blackish. Tips of caudal rays blackish. Outer portion of soft anal rays black-. ish. Tips of pelvic rays blackish. Pectoral fins solid red. Life colors of specimen from station 66, 75 mm. in standard length: Bril- _Lliant solid red in life. Head blackish on symphysis of lips, over top of head across snout to back; eye broadly ringed with reddish black; median ventral region dusky reddish-black. Remainder. of head red. Iris golden red. Body solid black at base of dorsal fin shading to red in region of lateral line. Remainder of body red. Spinous dorsal black with the outer half of the membranes tinged with red; soft dor- sal with outer half solid red and basal half solid black. Caudal fin -89- bright red, same as body, with tips of lower rays dusky black. Anal fin with the outer posterior membranes shading to dusky black. Pel- vic fins solid red; membranes shading to dusky black distally. Pec- | toral fin solid red. Stations: 17(1), 66(1). Habitat: Only two specimens were seen at Raroia; one in a surge channel and the other in a scoured out trough lined by dense coral in Garue Pass. Both specimens were in about 20' depth of water and ap- parently hide in cracks in the coral. Their brilliant red color stood out above all else and it is probably a rare species at Raroia. FAMILY CHAETODONTIDAE Butterfly Fishes Sixteen species of this family are know to occur at Raroia, and I doubt that more than one or two other species exist there. Fifteen species were taken and are reported upon; an additional unidentified form was seen in deep outer reef surge channels, but was not obtained. Only a few butterfly fishes are actually common at this atoll; most forms are rare, either being markedly restricted to a certain part of the atoll, or widely scattered with only one or two existing at any particular locality. Three species, Chaetodon setifer, C. trifasciatus, C. lunula, are common at Raroia in the lagoon. The remainder of the species are much less abundant, often only a few occurring at a par- ticular locality. Butterfly fishes are abundant on the outer reef flat (es- pecially C. setifer) but do not go into the surge channels and deeper water. Only ©. guadrimaculatus, Centropyge flavissimus and an unidentified Chaetodon somewhat like C. reticulatus, were observed in the surge channels; never more than one or two were seen at any locality. The greatest concentration of butterfly fishes is on the shore reefs in Garve Pass to a depth of about 15'. Only a few chaetodontids were observed below this depth in the Pass. In the lagoon butterfly fishes will school around the coral heads to a depth of 25-30 feet, but rarely are seen below this depth. The natives call all members of this family KORAI, but have additional names for a few of the species. None are of use to the natives, since they do not eat them. Just why only certain species (some relatively rare) should have particular names, and other equally distinctive species not, I could never find out. I would. guess that originally each species of chaetodontid had its own native name, and now the natives know only a few of these original names. Heniochus permutatus Cuvier.--Three vertical black bands on body. Fourth and fifth dorsal spines elongate. Life colors of specimens from station 61: Head with a longitudinal black band on top of lips and snout, fading out on interorbital. Snout and interorbital region yellow. Underside of head pure white in life, faintly yellow in preservative. Iris gold. Vertical bands on head and body brownish-black to black. First band through eye lighter above than below, solid black around eye. Second vertical band solid black distally, shading to brownish-black on middle of side. Outer half of spinous dorsal hyalin with tips of rays white. Second band extending from meybranes of third to fifth dorsal spines; ventrally covering all soft anal rays. Third vertical band mostly on base of soft dorsal, covering basal half of fin and extend- ing suffusely forward to sixth spine. Caudal peduncle crossed above bya brownish-black saddle. Anal fin edged hyalin on anterior and posterior rays. Longest rays black at tip. Pelvic fins solid black. Pectoral fins hyalin. Area on body not covered by dark bands, pure white. Stations: 23(1), 28(1), 60(1), 61(2), 66(5). Habitat: Abundant around the sides of lagoon coral heads and on the inner face of lagoon shore reefs. It has the striking habit of ‘Swimming upside down on the ceilings of caves or overhanging coral Shelves. Most abundant on western shore reefs. This species is always seen in schools, seldom swimming individually. Practically all speci- mens had to be speared since they avoid rotenone. Raroian name: KORATI. Megaprotopon strigangulus (Gmelin).--Characterized by V-shaped lines on sides of body. Life colors of specimen from station 61: Head and body pale white to greyish-white above. Head with solid black band through eye that is narrowly continuous across interorbital, and broadly continuous across isthmus and breast; black band bordered i a white line which in turn is faintly bordered by yellow. Top of snout with a yellow saddle behind upper lip. Lower jaw yellow on each side. Otherwise upper part of head slate grey; nape between black band and dorsal fin orange. Iris silver to yellow, dusky black above, and suffuse black posteriorly and below. Body with approximately 22 V-shaped lavender lines that be- come smaller and closer together posteriorly, as the scales become smaller and closer together. The lines curve abruptly posteriorly above and below. Dorsal fin orange, fading out on scaled basal portion. Edge of last three or four spines and edge of soft dorsal narrowly hyalin with a submarginal black line. Caudal fin black, shading to orange on ‘caudal peduncle; abruptly yellow posteriorly and yellow-orange on upper and lower edges; yellow bordered by a vertical black line; edge of fin hyalin. Anal fin yellow-orange, edge same as soft dorsal. Pelvic fins pale yellow. Pectoral fins hyalin, upper basal area ae distally rays narrowly edged black. Stations: 13(1), 23(2), eat 49(1), 61(8). Habitat: In lagoon on coral heads and shore reefs, but rather un— a common. Usually only scattered individuals seen, except on a few eastern coral heads (such as station 61) where they are moderately abundant. Rarely taken by rotenone; most specimens speared. Raroian name: KORAI Chaetodon ornatissimus Cuvier.--Black band through eye extending onto breast. Oblique orange bands present on body. Life colors of specimens from station 33: Solid black vertical bands present on head. Grey patch anteriorly across interorbital. Lower jaw grayish-black. Iris yellow, with a vertical black band. The black band through the eye is bordered by bright yellow. Body pure yellow on breast and belly, cusky yellow above on nape; otherwise body and caudal peduncle white to grey ven= trally and on vertical fins. Sides of body with approximately six oblique bright orange bands bordered by suffuse black. Dorsal fin | basally with body coloration. Unscaled edge basally black. Spinous dorsal yeliow. Soft dorsal yellow with black edge. Caudal fin with vertical black bands near base and submarginal to edge of fin; region between black bands yellow or with a vertical yellow line. Anal fin similar to soft dorsal except for a broad dark browm band bordering basal black band next to body. Posterior edge of fin hyalin. Dark brom band shades to orange at posterior edge. Pelvic fins pure yellow. Pec- toral fin rays pale yellow, membranes hyalin. Stations: 66(4). _ Habitat: This species is abundant only on the shore reefs in Garve Pass. ‘The only other individuals were seen on "Makoto" coral head (see station 43). They live at a depth of 5-10' in dense actively growing coral. Never taken by rotenone despite repeated attempts. All speci- mens speared. Raroian name: KORAI. Chaetodon setifer Bloch.--—Distinctive markings consist.of a black spot on soft dorsal, a broad black band before eye and oblique lines on kLody. Life colors of specimens from station 61: general color white anteriorly and yellow-orange posteriorly. Head white, except for broad black band below eye; suffuse pale band extending above eye and yellow lines ex- tending across inter-orbital. Band below eye bordered by white line - anteriorly, continuous below, but not extending on breast. Gill men- brane medially with a longitudinal dusky black line. Lips dusky black. Body with five complete bands extending obliquely downwards and for- ward; eight complete bands extending obliquely upwards; two or three separate V-shaped bands present. Below pectorel fin are two incomplete lines. All of these lines on body are lavender, becoming broader and yellow-orange on dorsal fin. Posterior part of body yellow-orange. Dorsal fin with membranes and spines dusky yellow, edged in darker brown, shading posteriorly to a black edge. Dorsal fin filament yellow- orange. Posterior edge of fin black. Large solid black spot present Boo on soft rays behind filament. Caudal fin yellow-orange. Middle of fin with a vertical yellow band bordered by narrow suffuse browish—-black lines; posteriorly there is a narrow brownish-orange band that narrows to a line above and below; edge of fin hyalin. Anal fin with a broad yellow-orange band following contour of fin. Submarginally there is a narrow black line bordered posteriorly by a narrow white line. Pel- vic and pectoral fins hyalin. Stations: 9(2), 16(1), 22(3), 23(6), 27(1), 28(6), 37(2), 38(3), 45(1), 49(2), 50(1), 52(1), 60(3), 61(4), 65(3). Habitat: This is one of the most abundant species at Raroia, oc- curring in great numbers on the lagoon coral heads, shore: reefs, and outer reef flat. Specimens easily taken by a ad Raroian name: KORAI. Chaetodon trifasciatus Mungo Park.--Life colors of specimens from station 57: Head with a narrow black stripe through eye from before dorsal origin to mid-ventral line. This stripe bordered by a bright zyeliow band. Below eye the yellow bands are almost as wide as eye. Above eye these bands are narrow, the anterior yellow stripe shading into orange. Posterior band bordered by yellow-green.. Both above and below eye, yellow band posteriorly bordered by narrow light blue line extending entire length, bordered by a browmish-black vertical line. Front of head yellow-grey to black, margin of upper lip black. Operculum yellowish-brom. Mid- line of nape before dorsal fin orange-browm. Iris black above and below; solid yellow otherwise. Body shading from a pure yellow an- teriorly on belly to a purplish-orange posteriorly. Longitudinal stripes on body purple. Spinovs dorsal white distally to pale yellow basally. Soft dorsal with an elongate tear-—drop shaped black band extending from caudal peduncle onto base of soft dorsal fin; soft dorsal pale at edge; middle of fin crossed by purplish—-black lines; fin dusky yellow between lines. Caudal fin crossed in middle by a vertical black line bordered by yellow-orange anteriorly and yellow posteriorly; remainder of fin pale; base of fin white. Anal fin basally with a black club-shaped line bordered by pure yellow that extends forward anterior of anus. Edge of soft fin yellow, bordered by dusky black that shades into red- dish-orange. Pelvic fins pure yellow. Pectoral fins hyalin, with a bright yellow base. Stations: 7(1), 9(2), 13(1), 22(1), 23(16), 28(9), 37(1), 42(10), 49(4), 57(1), 61(10), 66(2). Habitat: This species is the most abundant iottoubier eek at Raroia, occurring in great numbers in the lagoon on coral heads, shore reefs and also on the outer reef flat. Specimens easily taken by rotenone. Raroian name: KORAI. -93- Chaetodon lunula (Lacépede) .--Most distinctive marking is a V-shaped black band on forehead. Life colors of specimens taken at station 47: Head with a solid black band through eye, bordered dorsally and nosteriorly by equally large broad white saddle. Remainder of head yellow-orange. Iris black, narrowly ringed at lens by dusky gold. Black band through eye narrowly bordered anteriorly by pale blue. Body above with a V- shaped black band anteriorly, which is interrupted dorsally in prac- tically all specimens. This V is a browmmish black wedge, bordering white and orange above eye. Area between 'wedge and coloration around eye is orange. Middle of body with approximately 11 red-orange bands curved obliquely upward. These bands extend dorsally to :body edge and slightly onto soft dorsal fin. Below pectoral fin are approximately 11 red to reddish-orange spots thich have no connections to oblique bands. Lower part of body shading to yellow on belly, superiorly shading to browmmish-yellow. Caudal peduncle crossed by black band that extends narrowly upward and forward onto dorsal fin. Black band bordered an- teriorly on body by bright yellow stripe. Spinous and soft dorsal bordered by broad black band extending to upper angle of soft dorsal. Remainder of spinous dorsal yellow. Soft dorsal with yellow band bor- dering black that ends on posterior edge of fin. Caudal fin yellow with vertical orange band across middle and'a submarginal black band. Edge of fin broadly hyalin. Anal fin edged’ in broad black line and an orange submarginal line. Pelvic fins yeliow. Pectoral fins basally yellow, hyalin for outer 4 of fin. 65( fe 22(2)')23(2), 28(2)4" 440) joao), 270599 GOts)s 6201), 5(1). ) Habitat: One of the most abundant butterfly fishes at Raroia, occurring in the lagoon on coral heads and lagoon reefs; scattered in- dividuals on outer reef flat. Specimens taken by spears and rotenone. Raroian name: KORAI MU or KORAI TIKAMU. Chaetodon citrinellus Cuvier.--Distinctive colors consist of a black stripe through eye, numerous blue spots on body, and no marks on caudal region. Life colors of specimens from station 55: Ground color solid yellow. Head dusky brown on snout shacing into light blue below horizontal level from mouth. Vertical stripe through eye solid brown to blackish- brow. Iris gold to greenish-gold, browm above and below. Remainder of head brownish-yellov. Stripe below eye not bordered by distinct color bands. Above eye stripe bordered by narrow suffuse band of light blue. Body overall yellow, slightly greenish, though in life bright yellow. Spots on middle of side metallic blue, shading ventrally, dorsally and posteriorly into solid orange spots. Spots on caudal peduncle solid orange. Dorsal fin yellow-orange, posterior soft rays with numerous orange spots. Anterior dorsal spine dusky black, gradually becomine confined posteriorly to a submarginal black line that continues to upper angle of soft dorsal fin. Last dorsal rays yellow-orange at tip. Caudal fin bright yellow-orange, basal 2/3 of rays with numerous One orange spots. Anal fin edged in black. Black bordered anteriorly by a pure yellow band. Remainder of fin orangish-yellow with orange spots posteriorly. Pelvic fins white, tinged yellow-green. Pectoral fins pale yellow-orange. Stations: 55(2), 57(1), 61(1), 66(1). Habitat: Rare at Raroia. Pairs were occasionally seen on the western shore reefs in the lagoon (practically always at the mouth of inter-islet channels) and extremely rarely on lagoon coral heads. Never observed outside the lagoon. All specimens were taken with spears. Raroian name: KORAL, her By. c reticulatus Cuvier.--Characterized by 12 dorsal spines, black pel- vic fins, and the presence of white spots on a black body. Life colors of specimens from station 47: Head with a broad black band through eye bordered by yellow. Snout black except for dark grey immediately before upper part of eye. Posteriorly head grayish-white. Iris grayish-white, a vertical black line at anterior border of eye. Body anteriorly grayish-white; belly black before pelvic fins bordered dorsally by yellow streak. Posterior area of body black; edge of: each scale solid black, middle of scale grayish-white bordered by gray—black. White spots on scale larger anteriorly, tiniest near caudal peduncle; caudal peduncle solid black. Spinous dorsal white to gray. Soft dorsal gray anteriorly to black posteriorly. Edge of soft dorsal hyalin, bordered ‘by black line at posterior angle. Caudal fin basally black, its out- line lunate. Edge of fin hyalin, bordered by two black lines vertically on middle of fin. Anal fin basally with tiny light gray spots. Most of fin solid bleck. Soft anal edge hyalin. Pelvic fins solid black. Pectoral fins hyalin. b y Btetdiens: Vea). A7( 2), seen Habitat: Very rare at Raroia except on reefs in vicinity of Garuve Pass, west side of atoll. Found in association with C. ornatissimus. All specimens taken by spears; it always escaped rotenone. Raroian name: KORAIL. Chaetodon facula (Bloch).--Distinctive markings are two vertical black bands on body and a black spot on caudal peduncle. Life colors of specimens from station 58: Head brownish-orange. Band passing through eye black; band narrow above eye. Iris yellowish-black, black above and below. Body with two broad black bands superiorly, shading out on level with pectoral fin. Ground color before hind edge of second black band bluish-white to pure white. Bleck bands extend on base of dorsal fin. 16 oblique pale black lines present on middle of side. Black band on head not edged in another color band. Operculum bluish—-white, same as ~95- body. Ground color behind. vertical from last body band abruptly solid yellow-orange. A black saddle is present at caudal fin base across lower side of caudal peduncle. Spinous dorsal yellow. Soft dorsal fin solid yellow-orange, brownish at tips above upper angle. Caudal fin edged hyalin with submarginal brown vertical lines; remainder of fin yellow-orange. Near base of fin are six reddish-orange spots in a ver- tical row. Soft anal yellow-orange, narrowly edged in orange before outer angle. Two faint submarginal black lines on middle of fin. Pel- vie fins and pectoral fins white. Stations: 11(2), 23(8), 28(5), 58(1), 61(2), 65(1). Habitat: Scattered over entire atoll lagoon wherever there is coral growth but not common at any locality. Rarely individuals were observed on outer reef flat, but never seen in surge channels. Very seldom taken by rotenone; most specimens speared. Raroian name: KORAI KEKE. Chaetodon unimaculetus Bloch.--Distinctive markings a black spot on back and a black ring on caudal peduncle. Life colors of specimen from station 66: Head grayish-white, dusky on top of snout. Broad black line extend- ing through eye; this line not bordered by other colored bands, con- tinuous across breast. Iris blue, broadly black above and below. Body with a large round black spot in region of lateral line, suffusely bor- dered by a bread white ring. Upper section of body yellow, lower part white tinged with blue. 6 oblique yellow lines present above pectoral fins that are edged with brow. Dorsal fin solid yellow; soft dorsal edged white with a vertical black band above caudal stripe. Black band on dorsal preceded by a gray stripe. Caudal fin white at base; caudal rays hyalin. Anal fin solid yellow, marked like soft dorsal except for gray band. Pelvic fin solid yellow. Pectoral fin hyalin. Stations: 66(1). Habitat: Very rare. Two specimens seen in lagoon: one on coral reef next to shore in Garuve Pass; one on "Makoto" coral head among dense coral growth (see station 43). Raroian name: KORAI. Chaetodon bennetti Cuvier.-——Distinctive markings consist of a black spot bordered in blue on middle of side of body and of two blue lines on lower part of side. Life colors of specimens from station 61: Overall solid yellow tinged with orange. Head with browmish-black band through eye, that is narrowly separated across nape and ends ventrally at opercular margin. Black line yellowish in center and narrowly edged solid black on each side. Hye band bordered by a blue line anteriorly and posteriorly. Front of head with a dusky brown line on top of snout from upper lip to interorbital. Operculum with continuation of vertical -96- blue line that is mainly on body. Body with solid black spot on upper part of side that is ringed with a blue band and a narrow suffuse black line. Lower side of body with two oblique blue lines; the lower line is suffusely edged in black. The upper line is suffusely edged in black at the upper end only. Another faint blue line extends vertically from caudal peduncle to base of anal fin. Caudal peduncle with a dusky brown saddle near end of dorsal fin. Dorsal. and anal fins same yellow color as body. Caudal fin mostly hyalin, basally yellow. Pelvic fins yellow. Pectoral fins yellowish at.base, otherwise hyalin except for fine black edge to rays distally. Stations: 61(1), 65(1). Habitat: Very rare. Specimens taken were the only ones observed, except for an adult seen at station 42. Raroian name: KORA. Chaetodon quadrimaculatus Gray.--Characteristic markings a broad black saddle on back with a brilliant white spot in the center. Life colors of speci- mens from station 47: Head before eye yellow. Vertical band through eye yellow-orange bordered by brownish-black lines anteriorly and posteriorly. Posteriorly head pale blue to dusky yellow. Body solid black above shading into longitudinal brown spots below. Middle of back with a brilliant white spot that is the most distinctive marking of this fish as it swims in the water. Another similar white spot is present in a finger of yellow-orange that extends dorsally into the black coloration beneath anterior part of soft dorsal. Below pectoral fin body pure yel- low. Spots do not continue ventrally on body. Band across caudal peduncle brownish-black. Spinous dorsal yellow-orange for outer part of fin, brownish at base. Posterior spines black for most of their length; then blue bordered narrowly by black; tips of membranes yellow- orange. Soft dorsal basally black to brownish-black. Anteriorly a pale blue line narrowly edges the black. Caudal fin yellow-orange with a narrow hyalin'edge. Anal fin yellow-orange with a blue Tine’ through spines and rays starting from base of first spine to outer ¢ of soft anal. fin. Blue line narrowly edged in black. Pelvic fins solid yellow- orange. Pectoral rays pale yellow, longest rays dusky black ‘ae tip; membranes hyalin. Beene: A2AL.) yh) porBBC 1). Habitat: This striking species is rare at Raroia. Approximately Six specimens were seen during the entire summer, both in the lagoon and surge channels of the outer reef. All three specimens were taken by spears. They completely avoided rotenone. Never more than two were seen at cee same locality. Chaetodon eect Cuvier.--Characterized by black ovate aan on ave and a long filamentous soft dorsal ray. Life colors of specimens from arc station 61: General ground color pale greyish-white tinged with laven- der. Underpart of head solid orange. Eye with a short narrow black line above and below. Iris silvery white. Upper border of eyeball with a black line that continues dorsally; eyeball orange anteriorly and posteriorly. Body with prominent black ovate spot dorsally that covers most of dorsal fin. Black bordered by a broad white band ex- cept for upper and posterior margins. Approximately seven longitudinal lavender stripes present on lower side behind pectoral fin. Breast solid orange, same color as head. Membranes of spinous dorsal with a longitudinal biack line that is bordered by hyalin shading to yellow tips. Soft dorsal above black area hyalin, shading into yellow. Caudal - fin hyalin, the upper and lower rays orange. Caudal peduncle crossed by an orange ring that extends anteriorly and ventrally. Anal fin same as body for most of fin; edge narrowly black. Middle of fin with an orange-red crescent which extends anteriorly as an orange line onto mid- dle of last spine; posteriorly the crescent is more red and bordered by dusky black lines. Pelvic fins yellow-orange. Pectoral fins hyalin, a faint orange line across base. Stations: 7(Lyet2eGL) Si6RCA)s Habitat: A moderately rare species of rather confusing distribu- tion in the lagoon. In some areas it is moderately common, and yet on neighboring coral heads and shore reefs it is absent. It normally oc- curs in small schools; very seldom just indivicually. Pomacanthus imperator (Bloch).--Life colors of specimen from station 45: Ground color solid black. Lines yellowish-white. mtire dorsal and soft anal and outer one-half of caudal fin hyalin. Black on caudal fin bordered posteriorly by vertical yellowish-white line. Life colors of specimen from station 60 (adult 200 mm. in standard length): Head and body black to brownish—-black. . Head dark brow, lips darker. Broad black band the width of eye crosses interorbital. This black band bor- dered by dusky blue line. Posterior edge of operculum and anterior region of body with a broad vertical band that extends ventrally onto breast and belly; anteriorly this band is bordered by dark blue stripe. Body with approximately 17 horizontal orange lines that extend obliquely upward; 11 extending into dorsal fin. -Dorsal fin white at edge. Spin- ous dorsal same color as body basally, with orange lines extending on posterior spines and continuing on soft rays. Soft dorsal posteriorly with orange filament at angle; last rays before angle narrowly edged in white, ventrally in black. Posterior to filament the black band is broader, and narrowly bordered by orange. Caudal fin orange; caudal peduncle gently fading from stripes into orange of caudal. Anal fin solid black, basally with narrow brom lines curving the same direction as those on body. Pelvic iins anteriorly narrowly edged in black; soft rays orange, shading to black posteriorly; membrane between first two soft rays orange. Pectoral fins with black rays and hyalin membranes. Stations: 45(1), 60(1), 65(1). aoe | Habitat: Entirely confined to outer reef flat and channels of atoll, although the natives said the adults also occurred around the lagoon coral heads. Centropyge AES (Cuvier).—Distinctive colors a solid yellow-orange overall with the eye ringed in blue. Life colors of specimens from sta- _ tion 61: Head with dusky grey on lower jaw. Blue line around eye broadest above and below. Blue line on bony opercular edge extending from edge of gill opening ebove to near where opercular spine crosses below. Spine yellow, with black tip. Last six dorsal spines and-eall of soft rays narrowly edged in hyalin with a submarginel black line. Posterior edge of caudel fin hyalin and with black line exactly as in soft dorsal. Anal fin same as soft dorsal; spines edged with hyalin and black line. Pelvic and pectoral fins the same color as body -- yellow-orange. , LAT: Stations: 23(1), 26(2), 61(16), 65(3) 66(1). Habitat: Scattered individuais Waenevon there is active coral growth at a depth of 3-15'. This species prefers the lagoon coral heads, usually being found on the west face. At no locality was this species abundant. os bey FAMILY POMACENTRIDAE Damsel Fishes Seventeen species are recorded from Raroia, and I doubt that other spe- cies occur there. All species observed were collected. The Pomacentridae form one of the most important groups inhabiting the coral reefs at Raroia. Most of the species are. restricted in habitat. Dascyllus arvanus occurs the deepest to a depth of 80' on the coralliferous outer bench. Abudefduf phoenixensis and Abudefduf sp. one are confined, to the surge channels. Abudefduf imparipennis is the dominant fish on the sea urchin zone behind the coralline ridge and in the shallowest inter-islet channels. Pomacentrus nigricans dominates the tops of the. reefs, in-the lagoon. Amphiprion bicinctus and Dascylius trimaculatus live among the tentacles of large sea anemones in Garue Pass. The natives have names for some of the species, but most of the pomacentrids have no names. MAMO is a general name referred to some species, but the natives are not sure for just how many damsel fishes this term is ~ correct. Probably the name originaily referred -to all pomacentrids and is being lost, since the natives pay little atterjtion to this group, and do not eat any of them. Amphiprion bicinctus Ruppe?l.--Characterized by two vertical blue bands on side of head and body. No white markings on head. Head and breast orange. Life colors of specimens from station 66: Head, breast and is\o Mh lower half of body anterior to anus bright orange. Vertical. light blue band on head and hody bordered by a black line. Iris orange, with a dusky blue crescent around posterior margin. Body shading to brighter | orange on caudal peduncie and to solid orange on anterior lower part of body. Remainder of body orengish-browmm. Spinous dorsal fin orange- brown, edged in black. Soft dorsal similar but without the black edge; longest rays dusky black at tip. Caudal fin orange, tins of rays dusky black. Anal fin like spinous dorsal, orange-brow with a black edge. Anus and surrounding area dusky black. Pelvic fins orange, the anter- ior and outer rays black. Pectoral fins orange, the outer region pale black. Stations: 66(4). ! Habitat: This species lives only among the tentacles of large sea anemones, which occur at Raroia only in a small patch in Garue Pass. These sea anemones are at a depth of 25', among abundant free-growing algae. Abudefduf lives in association with Dascyllus trimaculatus. Raroian name: none. Chromis caeruleus (Cuvier).--Life colors of specimens from station 61: Gen- eral color solid dark blue shading to light blue below. In water the | overall coloration is often an iridescent green or blue over head and body. Top of snovt and interorbital pale black. Iris blue, narrowly gold around lens. Black band present at upper origin of pectoral fin. Dorsal fin pale black, with a black edge to spinous rays. Caudal fin bluish-black on scaled portion; outer rays black shading to pale black on middle rays. Anal fin dusky black. Pelvic fins with scattered melanophores on rays, membranes hyalin. Pectoral fins same as pelvics, but most of the pigment is on the upper first ray. Stations: 9(9), 13(2), 37(1), 57(1), 61(60)._ Habitat: This species prefers the calmer areas of the lagoon and the greatest numbers occur on the eastern coral heads. While it lives from the surface down to a depth of approximately 35' it prefers to remain around the upper edges of coral heads. It often is in associa- tion with Dascyllus aruanus. Raroian name: MAMO . Abudefduf septemfasciatus (Cuvier).--No spot present behind dorsal fin. Six black vertical bars on body. Preorbitals and preopercular edge naked. Stetions: 11(2), 45(1), 58(3). Habitat: Uncommon at Raroia. Small schools observed along shore in lagoon and scattered individuals on outer reef flat. Raroian name: KOTIMU. ~100- Abudefduf sordidus (Forskal) .--Black spot present at pectoral base. Another larger black spot present on upper half of caudal peduncle. Head and body with brom vertical stripes. 7 Stations: ge 10(2), 11(2)e' 16(1)) 170), 20(1), 22(3), 26(3), 35(2), 45(4). Habitat: Scattered individuals along shore, but primarily lives in inter-islet channels and on outer reef flat. Raroian name: KOTIMU. Abudefduf sexfasciatus (Lacépéde).--Life colors of adults from station 60: Head dark blue above, blackish in mid-dorsal region. Upper lip blue. Lower lip dusky blue. Head light blue below. Iris gold, ringed with black. No bands on head, except first body band torders operculum. Body anteriorly dark blue akove and light blue below. Five vertical black bands present that do not extend onto belly. Spinous dorsal black; soft dorsal black anteriorly, blue posteriorly; upper edge of fin dusky black out onto filament. Caudal firi with middle of each lobe solid black; middle of fin bluish-black. Anal fin edged black, basally blue. Pelvic and pectoral fin rays dusky black, membranes hyalin. Stations: 12(1), 22(11), 37(2), 38(5), 60(7), 62(2). Habitat: Scattered individuals throughout lagoon around coral but abundant only in eastern and northern inter-islet channels and coral heads. Also lives in slightly brackish enclosed ponds on northeastern islets. Often occurring in association with Pomacentrus pavo. Raroian name: TUKUPAHI. Abudefduf phoenixensis Schultz.--Head and body browm. Vertical stripes on head and body yellow to orangish-yellow. Black spot present on pec-— toral fin base. Last body stripe followed by a vertical black bar, darkest dorsally. Entire caudal fin light yellow, almost completely white behind black bar. Anal fin blackish-brow, yellow-browm below black bar. Soft dorsal posteriorly edged orange-brown below black spot. Eye light blue. Yellowish-brown spots on cheeks and suborbi- tals. Raroia specimens differ from Schultz's types in having a black bar on pectoral fin base. Stations: 47 (6) 5 27(2).,) S82). Habitat: This is the only pomacentrid besides Abudefduf sp. one found in surge channels and is entirely confined to this region. It is moderately common in the surge channels, and apparently lives in holes in the coral buttresses and under rocks. Raroian name: TUKUPAHI. ~-101- Abudefduf imperipennis (Vaillant and Sauvage).—Keys out to A. imparipennis but differs somewhat in coloration. Upper part of head, body and most _ of caudal peduncle greenish-yellow. Lower half of head and body ANS bluish-white. Ventral and posterior scale rows with an indistinct pale blue center, that tends to form faint blue longitudinal? lines, particularly on the caudal peduncle. Spinous dorsal same color as back. Soft dorsal light yellow to hyalin distally. Caudal fin yellow to hyalin posteriorly. Anal fin slightly yellow. Pectoral and pelvic fins hyalin. Stations: 11(1), 17(36), 22(9), 58(5). Habitat: This species is rare in the lagoon, but one of the domin- ant species in the outer reef entrances to shallow inter-islet channels and is the dominant fish in the sea urchin zone behind the coralline ridge. It is quite rare on the remainder of the reef flat and absent from the surge channels. It lives in sea urchin holes and under rocks and coral. Raroian name: MAMO. Abudefduf dickii (Liénard).--Characterized by a vertical black bar on poster- _ ior part of body extending onto dorsal and anal fing. Life colors of specimens from station 57: Head uniform brom without markings. Darker brown on top of head. Eye narrowly rimmed in black. Iris light blue, Garker blue above and below. Body crossed by a broad black band that shades out abruptly on dorsal and anal fins. Anterior to this vertical bar the body is solid brow; each scale with a blackish vertical line at posterior tip, forming vertical lines of olack. ' Kegion below pec- toral fin and belly solid brow without vertical blac bars. Dorsal fin same color as body where covered with scales. ‘f£pinovs dorsal dusky black, with pale centers on the membranes. Soft dorsal also dusky black with longest rays solid black; posterior reys hyalin at tip. Caudal peduncle rosy—white shading to a peach color on caudal fin. Anal fin with spines and membranes black. Pelvic fins brown, © first spine and soft ray black. Pectoral fins orange. ‘Stations: 57(3), 58(1), 61(1), 65(1), 66(1). Habitat: Uncommon at Raroia and very restricted in distribution to definite scattered localities. It lives in holes in coral heads and is rarely found on the outside. Raroian name: none. Abudefduf giaucus (Cuvier).--Body light blue with darker vertical fins. Anus black. Young with a brilliant blue line from tip of snout past upper edge of eye. Caudal lobes mounded. Stations: 7(4), 8(130), 9(6), 10(73).)\ RY, oe 20(5), 22(66), 37(3), 38(1Z),,. 57(2),. ~102- Habitat: Probably the most abundant pomacentrid at Raroia, oc- _ curing almost entirely along the shore on the outer reef flats, inter- islet channels and shore reefs. It is the dominant fish in the shallow _inter-islet channels continuous between lagoon and outer reef flat. Raroian name: MAMO. Abudefduf sp. one.--Caudal fin emarginate and outer rays dark. Black spot present on upper pectoral fin base. Life colors of specimens from station 59: General color silvery blue on top of head and back abruptly dark blue. Lower half of head and body light silvery blue. No mark- “ings on body or head. Upper edge of eye black. Iris iridescent silver, appearing light blue or green in life. Dorsal fin narrowly edged in white, submarginally bordered by a narrow black line. Spinous dorsal dark blue, soft. dorsal also dark blue, except for basal portion of last soft rays which are hyalin. Caudal fins with outer rays bluish-black, inner rays pale ‘blue to hyalin distally. Anal fin with entire edge : blue. Pelvic fins white. . Pectoral fins hyalin, a black spot at upper base. Leh Stations: 59(3). Habitat: Apparently definitely restricted to ngeep surge channels with gravel or cobblestone bottoms. Raroian name: None. Abudefduf sp. two.--Keys out to Abudefduf filamentosus in Schultz (1943) but coloration entirely different. Anus pale white. Caudal.fin lobes pro- duced. Anterior part of body bromish-black; posteriorly: abruptly white. Life colors of specimens from station 66: In life almost solid black posteriorly to a vertical from anterior soft dorsal rays. Pos- teriorly rays and body bluish-white shading to white. Outer portion of caudal fin dusky black and upper edge of dorsal and lower edge of anal black. Pelvic fins black. | Pectoral fins hyalin with solid .black base. ' if Stations: 66(4)._ Habitat: Scattered individuals observed only on the shore reefs in Garue Pass in the vicinity of large sea anemones, but not wes aueee the tentacles. Raroian names: none. Dascyllus aruanus (Linnaeus). --Characterized is black and white or black and light blue v blue vertical stripes. Stations: 7(16), 9(19), 13(1), 22(9), 37(1)., 42(9), 61(21). Habitat: The most abundant pomacentrid in the lagoon around all coral growth. Uncommon in inter-islet channels and rare on outer reef -103- << flat. It was observed to depths of approximately 80' on the corallif- erous outer bench. It especially prefers staghorn coral and pieces could be lifted out of the water without Dascyllus aruvanus leaving them. Raroian name: TUKUPAHT. Dascyllus trimaculatus (RUppell).--Solid black. Adults with a white spot on lateral-line under spinous dorsal, which is lost in preservative. Life colors of specimens from station 66: Head black above to dusky black . below. No markings. Iris blue. Body black, each scale dusky blue at base and black at outer margin. A white spot of. varying size on lateral- line and adjacent scales under last spinous dorsal rays. Most young lack this spot. Spinous dorsal solid black at edge shading to dusky black at squamation. Each membrane above scaled portion has a dusky green suffuse spot. Soft dorsal black at base shading to dusky hyalin distally. Caudal fin jet black. Anal fin jet black in region not scaled. Pelvic fins black with whitish base. Pectoral fins black. | Stations: 66(31). Habitat: Seen only in Garuve Pass, hovering over the same large sea anemones as are used by Amphiprion. They will also dive into the neigh- boring holes, but most dive into the sea anemone tentacles to escape danger. dt :: e Raroian name: none. Dascyllus sp. one.--Solid dusky black with a vertical solid black band through pectoral fin. Posterior edges of caudal and soft dorsal fins yellow. No white spot on. body. Life colors of specimens from station 23 one Month after preservation: Body dark brown with dusky blue base to each scale. A suffuse vertical black band extends to pectoral fin base dor- sally to nape and ventrally to pelvic fin base. Spinous dorsal solid black. Soft dorsal basally black, remainder of fin hyalin to pale yel- low. Caudal fin basally dusky brow; most of fin pale yellow. Anal fin black, last. ray dusky black at base and hyalin at tip. Pelvic fin jet black. Pectoral fins browmish-black at base, pale yellow to hyalin distally. No trace of white spot on hody but I have a recollection that there might have been one in life, though the field notes do not discuss this coloration. Largest specimens without trace of vertical black band. Stations: 23(2). Habitat: Taken on shore reefs in lagoon near Garumaoa Village but otherwise not observed in field. Raroian name: none. Pomacentrus pavo (Bloch).--Body and head blue with brilliant metallic blue spots. Black spot on upper opercular border. Life colors of specimens —LOA- from station 37: Head and body solid blue. Head dorsally dusky black. Each scale on head and body with a brilliant-metallic blue spot, these spots small and indistinct on top of head. Large scale of upper bor- der of operculum solid black with a light edge. Upper lip with a median blue streak and a lateral blue spot. Lower jaw covered with a horizontal streak. Body dark blue above (black in' formalin) to lighter blue below, each scale with a brilliant metallic blue spot. Caudal peduncle blue anteriorly shading to yellow posteriorly. Spinous dorsal bluish-black, distally with three rows of longitudinal pale blue streaks on each membrane. Soft dorsal also bluish~black, with irregular con- tinuations of pale streaks. Longest rays pale yellow distally. Caudal fin bright yellow shading to dusky blue at the base of each lobe. Anal fin blue-black, not as dark as dorsal fin, cba mece tT at tip. Paired fins hyalin. Stations: 37(2), 38(3), 61{6). Habitat: It is moderately abundant in the lagoon around coral in the calmest localities; this species is a typical inhabitant of coral heads close to shore on the east and north sides of the atoll, especially in protected bays and closed off intér-islet channels. Raroian name: MAMO. Pomacentrus albofasciatus Schlegel. and Muller.-—General color black or dark brown with a black spot at end of dorsal base preceded by a white spot. Stations: 9(3), 65(2). « Habitat: Same distribution as Pomacentrus nigricans but consider- ably less abundant. Raroian name: MAMO. Pomacentrus nigricans (Lacépede).--Solid dark brow fading into yellow-brom on edges of dorsal and caudal fins. Black come at upper pectoral base border and at end of dorsal fin base. _. Stations:. 7(25) , 8(3),. 9(87), 11(2), L216) ote, 22(1), 23(48), 28(95), 49(2), 57(3), 61(83), 65(12). Habitat: Almost entirely confined to yeeben where it is very abundant on top of coral shelfs and coral heads. It is rare in depths below a few feet. It is also found in inter-islet channels and oc- casionally on the outer reef flat. Raroian name: MAMO. -105- FAMILY LABRIDAE Wrasses Apparently the third largest fish family and certainly one of the most conspicuous groups at Raroia is the Labridae. Twenty-five species are re- corded in field notes, although less than half the material collected was examined. Perhaps an additional 10 species will be added after further study. All species noted at Raroia except three were collected and exten- Sive notes prepared of many species. According to the natives a few addi- tional species @f "PAKOU") were missed. Wrasses are abundant at Raroia wherever there is coral, forming an especially important part of the lagoon fauna on coral heads and shore reefs. All the species are partial to definite habitats, although only part of this data is presented in the fol- lowing notes; the remainder of the notes will have to be checked with the unrecorded collections. Of particular interest is the large number of species (9) of the genus Thalassoma occurring at Raroia. In fact this is the dominant labrid genus at this atoll and occurs in great numbers in the lagoon. The next largest genera in number of individuals at Raroia are Pseudocheilinus, Halichoeres and Stethojulis. The natives refer to most wrasses as PAKOU and have specific names for only a few species. The natives commonly use most of the larger wrasses for food, preferring most of all the genus Cheilinus. Epibulus insidiator (Pallas) .--Characterized by a protractile mouth. Life colors: Solid black to solid yellow and all stages in between. Stations: 7(1), 23(5), 28(3), 49(3), 61(1), 66(1). Habitat: Scattered individuals around lagoon coral heads and shore reefs, although never particularly abundant at any locality. Raroian name: TUPIROPIRO. Labrichthys sp. one.--Head and body fully scaled with remarkably large scales. Life colors of specimen from station 57: Head and body olive brown. Head with light upper lip; reddish lower lip. Interorbital dusky black. Bordering posterior edge of eye is a broad yellow band edged with brown stripes meeting above on occiput and ending at opercular border below. FEyeball green. Iris light blue, narrowly edged with red around lens. Body uniform olive brow, except for vertical yellow band edged with brom stripes at origin of caudal peduncle; yellow continues anterior to black spots on anterior part of soft dorsal and anal fins. Spinous dorsal uniform olive brow. Soft dorsal with a black spot at anterior edge ringed narrowly with light blve and red lines. Rays behind spot red, membranes hyalin. Caudal fin white basally, reddish-brom medially and out each lobe. Tip of each lobe solid white. Anal fin with spot and rings similar ~106- to dorsal. Rays behind red ring with hyalin membranes. Pelvic fins light blue at base and tip; middle with a broad browa band. On belly behind pelvic fin is a large round brow spot that matches the brow band on the pelvic fins. Pectoral fins red. Stations: 57(1). Habitat: Two specimens were taken at the base of lagoon coral heads at 6! and 25' depths. They appear to be the young of a pre- sumably different colored adult. Raroian name: none. Stethojulis phekadopleura (Bleecker).--Bright orange spot above pectoral fin base. Pale longitudinal series of spots on lower half of body. . Suf- fuse red streak along body at level of pectoral base. Two black spots ringed in bive on caudal peduncle. A suffuse band extends horizontaily across head beneath eye.. Stations: 9(6), 22(2); 23(1), 28(2), 491), 50(4). Hebitat: Uncommon;.small schools occasionally observed in shallow water in inter-islet channels and around lagoon coral heads. Most specimens were noted in less than 10' depths. This. species definitely prefers shallow. sina. areas with dead coral and rubble. eroien name: “PAKOD. Stethojulis sp. one.~--Cheracterized by 9 dorsal spines. Projecting canine in upj,er jaw at recius. No canines at symphysis of lower jaw. Three red lines horizontally. across. side of head, two of which continue along midlateral region of body. Mid-caudal rays black. Colors of specimens from station 14 one month after preservation: General ground color yellow-green, brownish on back. Head with 3. longitudinal red bands. Upper bands start on snout at level from top of eye, continu- ous across top of snout. These bands. extend posteriorly from middle of eye to above pectoral fin,base. A faint orange line extends obliquely backwards from eye, fading out on nape; not. continuing across top of head. Mid-Lateral band on head starting from posterior edge of upper jaw, extending to lower border of eye, then across operculum to above pectoral fin and fading out over middle anal rays. Lower mid- lateral band continuous across throat, but not reaching opercular bor- der. This line continues on body bordering operculum to near pectoral _ fin base; then there is 2 spot;at pectoral base and the line continues on the body below the mid-lateral line. Head with a broad suffuse black spot between eye and upper opercular border. Iris dark blue. Fins pale yellow. Caudal fin with.a black streak on basal half of mid- dle rays. Life colors of specimens from station 28: Same as previously described except for effects of .preservation.. Mid-lateral bands pink. A large yellow saddle edged in black present over pectoral fin base. Iris gold. Lower half of side with five series of bromish spots near -LO7- the center of each scale. Preopercle and chin pale blue. Outline of back edged with a blue line, Stations: 14(2), 28(5). Habitat: Recorded from inter-—islet channels and lagoon coral heads. Raroian names PAKOU. Leptojulis sp. one.~--Canine present at corner of upper jaw. Large black spot on first soft dorsal rays. Life colors of specimens from station 57: General ground color dark blue to bluish-black. Head with numerous pale lines, largest below eye. Lines on top of head faint, ending on snout. Iris blue-black with a paler longitudinal streak below lens and a pale line from the upper posterior margin of the eye. Lips pale. The pale coloration mentioned previously in this description is a slate grey mixed with blue; all pale markings on the body are this color. Bedy with many closely spaced pale longitudinal lines on the middie of scale rows, extending out to caudal base. Dorsal and anai fins intricately marked. Spinous dorsal with 3 longitudinal pale wavy lines. Soft dor- sai anteriorly with a solid black oval spot narrowly edged with a oure white and a narrow bluish-black line. Remainder of soft fin with a large irregular longitudinal streak down the middle of the fin. Tips of soft rays hyalin. Caudal fin uniform pale black. Anal fin ex- actly the same as soft dorsal except lacking spot and rings. Pelvic fins paie black. Pectoral fins yellcw-green basally shading to hyalin on outer half of fin. Stations: 57(2). Habitat: Recorded from lagoon reefs north of Garue Pass. Rare at Raroia. ; Raroian name: PAKOU. Thalassoma marnae Schultz.--Characterized by a suffuse broad longitudinal line on head and body; emarginate caudal; black pectoral base. Pre~ viously know only from young type. Life colors of specimens from station 61: General color above pale greenish-olive drab, white below. Head dusky olive drab above, a suffuse broad black band extending from snout through eye to operculum. Lower edge of line at lower margin of eye. Region: below line abruptly dusky white. Iris greenish- Silver. Lips dusky green, blackish medially. Body with upper half blackish to olive drab, abruptly white below. A suffuse broad median lateral blackish band continues from head posteriorly and fades out on caudal peduncle. Medially on side the black is composed of suffuse vertical stripes, usually two to a scale. Pectoral fin base with a black spot above. Spinovs dorsal fin black. Soft dorsal dusky black with a hyalin edge. Caudal fin pale black on tips of lobes. Anal and pelvic fins hyalin. Pectoral fin pale yellow at base, hyalin on outer half. ~108~ Stations: 61(3). Habitat: Observed only on large eastern ‘lagoon coral heads among dead coral r es See name: PAKOU. inkuevittata along ace base. Black spots in ep ara of five spinous and first soft rays. Life colors of specimens from station 57: General color green with white below. Head with black spot on snout above symphysis of upper jaw. Occiput orangish-brom. A blue band extends from upper posterior edge of eye across operculum; blue edged with blackish line. Another bend extends from lower edge of eye to upper edge of pectoral fin. Two red streaks present on upper part of oper-— culum. Head shades to white below. Iris green, edged in gold around lens. Body with 2 oblique red stripes under pectoral fin. Three longitudinal orangish-red stripes extend mid-laterally along body, the upper two connected by numerous oblique orangish-red lines. Dorsal fin hyalin with suffuse green on basal half; first two spines and first membrane with a large black spot. Fourth from last spine with a black spot at base; last four membranes with short red lines continuing ob- liquely forwird. Another large black spot between second and third soft rays. Fourth from last soft rays with a black basal spot; fourth to sixth soft rays with red lines continuing forward into fin. Caudal fin hyalin except middle ray green at base and yellow-orange in the middle of the fin, hyalin at the tips. Anal, pelvic and pectoral fins hyalin. Stations: 57(2), 58(25). Habitat: Recorded from both lagoon and outer reefs. _Raroian name: PAKOU. Thalassoma schwanenfeldii (Bleeker).--—Anal fin without black spot. Dorsal fin with 3 black spots. Body with 6 wide vertical black bands. Life colors of specimens from station 45: General color greenish, some specimens shading to pale blue below. Head blackish above, 2 large orange spots below eye. A black or bluish-—black or solid blue line posteriorly from eye to opercle border. Iris green, narrowly gold next to lens. Underside of head pale blue to white. Body bright green to dusky green above. Six vertical black saddles on back; the first extends down to a horizontal from pectoral fin, the others do not. No longitudinal pale bands on body. First 3 black bands usually contin- ued below with suffuse yellow-green coloration. Belly and breast pale blue to white. Last band on top of caudal peduncle very small in some specimens. Dorsal fin pale green to hyalin, with 2 or 3 spots near base; first spot on first spinous membrane, second between 2nd and 3rd soft rays, last over 5th black band, and is continuous with -109- the band on the body. Caudal fin pale yellowish-green, outer rays may be pale orange. Anal fin pale green to hyalin. Pelvic fin hyalin. Pectoral fins pale yellow, base more intense yellow. Stations: 7(16), 9(12), 13(1), 23(3), 28(9), 38(1), 45(2), 49(1), 50(1), 61(11). Habitat: This species is one of the most abundant wrasses at Raroia in the lagoon on shore reefs and coral heads. It is particu- larly abundant on the west side of the atoll in the lagoon wherever there is coral. It also was found in the enclosed brackish ponds on the northeast islets. Raroian name: PAKOU. Thalassoma hardwicke (Bennett) .~-Characterized by 6 vertical black bands and a black spot in the anal fin. Life colors of specimen from station 61: Head and body yellow-green above and blue below. Head with 5 broad orange bands radiating from the eye, the uppermost meeting across the interorbital. Lower jaw and lower part of preoperculum blue. Upper helf of head yellow-green, dark on top of head. Two large orange spots on nape, the uppermost continuous across top of head. Iris red around lens, outer half green. Body with 6 black vertical bands on side of body, the last a spot near top of caudal peduncle. Next to 7 opercular flap on body is an irregular black spot that is not continu- ous across back. Other black bands are continuous over top of body. First black band long, extending below a vertical from pectoral fin. Remaining bands not extending below mid-lateral line. Along mid- lateral region from second band to middle caudal rays is a pink stripe. Pectoral bese black. Dorsal fin with a median black longitudinal band that extends unbroken to above fourth band and ends at fifth band. Remainder of fin dusky hyalin. Caudal fin dusky yellow on base and outer lobes, median rays dusky black. Anal fin anteriorly with a black spot at middle of second ray. Middle of fin pale dusky yellow. Pel- vic fins hyalin. Pectoral fins hyalin except for ies black outer half and brownish-black upper edge. Stations: 23(6), 61(1). Habitat: Rare at Raroia. Field notes record small schools in the lagoon on scattered shore reefs and coral heads. Raroian name: PAKOU. Thalassoma umbrostigma (Ruppell) .--Black spots on head. Short vertical black lines on scales. Green longitudinal band on middle of soft dor- sal and soft anal fins. Stations: 10(16), 11(3), 14(5), 17(1), 20(2), 22(10), 23(1), 26(13), 212) 9: 4505) SSCL) 5, 65(5)) ~110- Habitat: Abundant wherever there is coral growth around the- ‘atoll. Adults are speared for food by the natives on the outer reef flat and coralline ridge. Mi, Raroian name: PAKOU. Thalassoma trilobata (Lacepede).--Double red line extends down side of body. Vertical narrow green blotches present between vertical red lines. Stations: 9(2), 22(4), 23(1), 26(2), 27(1), 28(2), 33(1),/ 58(10), 61(6), 65(1). ‘Habitats Same as Thalassoma umbrostizma although not as abundant, and tends to inhabit deeper “water. Raroian name: PAKOU. Thalassoma purpureum (Forskal).—-Two longitudinal atines along bases of anal and dorsal fins. Stations: 7(2) 5 17(14)),.45(6); 58(2), 60{1), 65(4). EFabitat: This species occurs throughout the reefs but is most abundent on the outer reef flat and surge channels. It is a deeper water snecies, coming into the shallow areas to feed at high tide. Raroian name: PAKOU. Thalassoma sp. one.--Differs from all described in the literature available Ty naving head pattern of purpureum; three longitudinal stripes on body without cross lines. Life colors of specimens from station 52: General color bright green above and bright blue below. Head with broad solid orange stripes on snout, occiput, nape and operculum. Iris green with a reddish gold ring around lens. Eyeball orange below, green above. Edge of opercular flap orange. Body with 3 longitudinal orange stripes, the uppermost brownish-orange, all stripes extending onto caudal fin. An oblique orange stripe ‘present below pectoral fin and 2 orange spots on pectoral base. This region around pectoral base green. Entire underside of head and belly bright blue. Dorsal and anal fins basally with a longitudinal orange stripe. Edge of fins blue-green. Caudal fin blue-green with an orange longitudinal stripe inside of outer prolonged rays in each lobe. Median rays edged in faint orange. Pelvic fins solid blue. Pectoral fins blue, dusky black on first 2 rays. Stations: 52(2). Habitat: Recorded from outer reef surge channels north of Garue Pass. Raroian name: PAKOU. 7 Ie Thealassoma sp. two.—-General color yellowish-orange to orangish-green. Head with two longitudinal orange-red lines. Dorsal fin with a black spot and blue line along its iength. Colors of specimens from station 23 one month after preservation: Head dusky above to yellow green below. Lips yellow-green, dusky at tip. Iris black posteriorly, golden an- teriorly. Below eye a longitudinal orange-red line extends from behind rictus to preopercle border. Orange spot present on operculum slightly . above a horizontal from eye. Another orange line present below eye on a horizontal from lower lip. Two redcdish-orange spots behind eye. Body dusky brovmish-yellow green above to paler yellow-green below. Breast and belly reddish. Side of body with two vertical dusky blue lines on each scale that form vertical lines on side of body as in umbrostigma. Base of vectoral fin with a black spot. Dorsal fin with a black spot on first two membranes - continued the length of the fin by a pale longi- tudinal blue band that is edged by a darker blue line. Remainder of spinous dorsal bright yellow to yellow-green. Soft dorsal otherwise hyalin with a bright yellow edge. Caudal fin pale yellow-green with a bright yellow posterior edge on median rays. Anal fin basally with a narrow longitudinal red band bordered distally by a bright blue line. Remainder of fin bright yellow tinged with green. Pelvic fins hyalin tinged with yellow. Pectoral fin basally bright yellow shading svt to hyalin. Head of larger specimen at station 23: two cark blve rFands ex- tend posteriorly from eye. Two broad orange-red bandas extend longi- tudinally below eye. Snout with an orange band between eye and pos— terior half of upper lip. Lower side of head bright yellow-green. Stations: 23(2) Habitat: Recorded from western shore reefs in Bape Rare at Raroia. Raroian name: PAKOU. Halichoeres trimaculatus (Quoy and Gaimard).--Black spot on caudal peduncle. Blue and orange spots on head. Canine at rictus of mouth in upper jaw. Black spot on pectoral axis. General color green with orange-red fins. Stations: 8(18), 9(22), 10(1), 11(4), 12(1)5 1407), 23(4), 28(9), 37(1), -38(7), 42(1), 49(1), 50(6), 65(5).- Habitat: Abundant in shallow water wherever there is coral; most prominent in the lagoon on shore reefs and coral heads. Raroian name: PAKOU. Pseudocheilinus hexataenia (Bleeker).--Characterized by red and green longi- tudinal stripes and by pointed snout. Stations: 7(7), 9(1), 57(1). -112- Habitat: One of the most abundant, yet least conspicuous, wrasses at Raroia. Particularly common in lagoon in 5-30! depths at the base of coral and sides of coral heads. Raroian name: MAMO. Gomphosus tricolor Quoy and Gaimard.--Characterized by solid indigo blue fins. Colors of specimen from station 23 one month after preservation: Overall color indigo blue; each body scale with a reddish-brown vertical band basally. Head mottled. green around eye. Iris narrowly gold around lens, green for remainder. Dorsal and anal fins blue-green, considerably lighter than body. Caudal fin blue-green, the upper and lower edges very dark blue-green. Pelvic fins pale blue-green, first spine and soft ray black. Pectoral fin black at upper base, middle of fin crossed by a vertical black band. Outer half of fin blackish blue-green, basal half brighter blue-green. Stations: 23(1). Habitat: Rare at Raroia, only a few individuals observed on lagoon coral neads and shore reefs. Raroien sane: KUTU (small), PAPARARI (large). Gomphosus varius Lacépéde.~—-Characterized by a black dorsal fin with white margin, and black anal fin with white spot on each ray. Dark spots on body. ; Stations: 7(2), 9(1), 13(3), 23(1), 28(3), 42(1), 57(2), 61(2). Hititat: Common in lagoon around shore reefs end coral heads. Often odserved in association with Labroides. Raroian name: KUTU (small), PAPARARI (large). Cheilinus undulatus Ruppell .--Characterized by black stripes extending from eye ob. “iquely . forward and posteriorly. Life colors of specimen from station 52: General color olive brown. Head uniform brown shading into blackish~prowa on occiput and nape. Most of head except above eye with reticulate pattern of narrow orange lines extending over lips, opercles, snout, and all of gill membranes; these reticuliations shade into red on operculum proper. Lips and anterior lower half of head shading to green. Body blackish brom in a broad band along back. On side of body each scale with a broad vertical olive brown line; these tend to become larger on dorsal and posterior scales. Scales in pectoral fin region, breast and belly, with fine reticulated red lines. Vertical fins with narrow vertical brownish—black stripes. Interspaces green. Edge of caudal fin yellow-orange. Pelvic fins light green, with red reticula- tions on bases of first two rays. Pectoral fins with basal two-thirds of fin with fine red EET ED oOo posterior third of fin light brown to hyalin. ~113- Stations: 33(1), 52(1). Habitat: Common in Garue Pass and in deeper water beyond surge channels around outside of atoll; occasionally coming into surge chan- nels to feed. Raroian neme: MARAIA (extremely lerge) , TATIKA (very large), TAPIRO (large to small). Cheilinus trilobatus Lacépede.--Young characterized by pale blue stripe. edged in black; general color brown; snout prolonged. Nine dorsal spines present. Life colors of adults from station 49: Ground color green to brommish-green. Head bright green with red wavy streaks, 4 few radiating from eye anteriorly. Suffuse darker areas at opercular margin, jaws, interorbital and nape. Body darker than head, with four distinct broad verticai brown bands which extend onto the dorsal and anal fins. Each scale has a reddish vertical streak across its base and onto the tip of the preceding scale and the anterior portion of the upper and lower scale, forming oblique red streaks. Dorsal fin with dark brow vertical bands through the first, fourth to fifth, and seventh to ninth spines; ground color of spinous dorsal green; there are also two iongitudinal wavy orange lines that extend onto the first soft rays. Soft dorsal green basally, pale orangish-red over middle of fin and hyalin distally. Caudal fin marked like body at base; medially there is a broad greenish-brow blotch bordered posteriorly by a narrow irregular green streak; this streak is bordered by an ir- reguiar red line; edge of caudal fin pale green. Anal fin similar to spinovs dorsal; two reddish-brown wavy streaks near edge of fin; an- teriorly spines and rays narrowly edged pale green; posteriorly fin green basally, mottled reddish-brown and green across middle, shading to pale red and hyalin distally. Pelvic fins green on the rays and’ bright red on the membranes. Pectoral fin pale orange, light green at base of rays. Stations: 7(2), 23(2), 49(4). Habitat: Lives in the ulva-like Microdictyon algae on coral heads and shore reefs, particularly at the southwestern protected side of the island. In fairly shallow water, seldom free swimming, usually hiding in algae. » Raroian name: TUPIROPIRO. Cheilinus chiorouvrus (Bloch).--Characterized by 10 dorsal spines and by black spots on body anteriorly. Life colors of specimens from station 49: Ground color green to brownish-green. Head mottled green to brownish- green, covered by red streaks and spots which extend onto lips; also numerous blue spots on postorbital region. Body with four suffuse vertical bands which are greenish-browm, same pattern as in trilobatus, except lighter. Many anterior scales have black spots; red spots on ~li4- occiput. Each scale with a-blue spot at base--no vertical red lines at base of scales; blue spots are not obscured by a vertical band, most distinct on caudal peduncle. Dorsal fin mottled green and greenish- brown; also with suffuse scattered black spots. Tips of spines pale, tips of membranes pale red. ‘Soft dorsal anteriorly same color as body and first dorsal with blue spots and black spots and mottled green coloration; posterior rays green at base; medially suffuse pale red with pale blue spots, hyalin distally. Caudal fin light green at base, ir- regularly reticulated vertically with suffuse red, green and brow lines. Anal fin same as dorsal; membrane between first two spines black. Generali color mottled green with blue spots overall. Felvic fins with green rays and pale bromish-red membranes. Pectoral fins hyalin, rays pale orange. | Stations: 13(1), 23(3), 42(2), 49(3). Habitat: Same as Cheilinus trilobatus. Raroian name: TUPIROPIRO. Labroides cimidietus (Valenciennes).--Life colors of specimens from station 61: Head blue with a median black line on top of head from tip of upper jaw posteriorly to dorsal fin. Side of head with a solid black stripe through sides of lips and eye, becoming larger posteriorly across oper- culum. Iris gold with a blue-green circle in middle. Body with the median black stripe on back becoming suffuse posteriorly and abruptly smaller near end of spinous. dorsal, ending on body shortly thereafter. Black band on middle-of side becomes larger posteriorly, always with a precise outline, extending to end of tail...-Area between these two black bands is solid. blue. Region -below mid=lateral band on body light blue, becoming: darker: bright blue next to anal fin. Middle of side red shading ventrally to pale red. Belly pale red. Red areas tend to have narrow. red longitudinal streaks on each scale row. Dor- sal fin solid black anteriorly, narrowing posteriorly into a median longitudinal black line that extends the length of the soft dorsal. Basal portion of the soft dorsal intense dark blue; edge pale blue. Caudal fin black in middle, intense dark blue above and below. Anal fin anteriorly solid dark blue shading to pale blue distally; poster- ior rays with a median longitudinal black band. Peivic fins hyalin. . Pectoral fins hyalin except rays edged in black. Stations: 13(1), 28(1), 61(5). Habitat: Common in. lagoon in coral on shore reefs and holes on coral heads. Rarely observed on outer reef. This genus is very shy and extremely difficult to obtain. Three species of Labroides were noted at Raroia but only a few specimens of one species were taken. Raroian name: none. . Coris angulata Lacépeéde.--Colors three weeks. after preservation of speci- mens from station 45: In life most conspicuous marking is a pure white -115- vertical band on body. In formalin head dusky pale green above to white below. Spots on head pure black, tending to be ringed by a pale blue line on occiput and nape. Iris dusky gold, with three spots near outer margin. Body anterioriy pale dusky white with a pale yellow shading. White band extends to mid-ventral region. Posterior half of body black above shading ventrally and posteriorly into dusky pale brown, somewhat yellow tinged with green. Dorsal fin anteriorly (be- fore posterior margin of vertical white band) dusky black with white edge; numerous black spots basally. Remainder of dorsal solid black with white edge. Caudal fin edged in pure white. White bordered by solid black that shades anteriorly into black spots. Anal fin edged white, bordered by solid black that shades basally to dusky black. Black area of fin covered by numerous intense black spots. Pelvic fins with anterior rays dusky black shading to hyalin. Pectoral fins with a black spot on middie of base; fin pale blue, upper edge tinged with orange. Stations: 45(2). Habitat: Rare at Raroia. Individuals noted on outer reef flat at high tide hiding in holes. Two were observed on iegoon coral heads among thick vlva-like algae. Raroian name: MARARI. Novaculichthys taeniourus (Lacépéde).--Life colors of specimen from station 49: Gereral color yellowish-brom. Ground color of head yellon-green to greenish-brom; lines radiating from eye brown, shading to a lighter greenish-brow, edged in light blue. Each scale on body (ex- cept those on belly) greenish-brom, with a basal light blue spot. Belly scales bright red, edged in white and then hyaiin. Large scale behind opercular flap with a long solid black base. Scales posterior to lower pectoral rays forming a verti¢al solid black stripe. Dorsal fin same as body except membranes between first three spines with two large black spots. Remainder of fin with oblique rows of greenish- brown spots with hyalin interspaces, becoming more brown posteriorly; edge of soft dorsal hyalin. Caudal fin crossed by a broad white vertical band; edge of fin narrowly hyalin; remainder of fin greenish brown with vertical irregular orangish-—browm lines. Anal fin darker than body, with oblique rows of brown spots, and greenish-brom inter- spaces. The tip of each soft ray is hyalin. Distally the membranes tend to be blackish, particularly anteriorly. Pelvic fins with reddish-black rays, and solid black membranes; tips of last three soft rays white. Pectoral fin browish-black basally, upper margin reddish, median region yellow, distally pale red and hyalin. Stations: 47(1), 49{1). Habitat: Rare at Raroia. A few individuals noted on large coral heads in southwest region of lagoon and one observed and collected on southern sand flats among coral rubble. Raroian name: MARARI. me FAMILY SCARIDAE Parrot Fishes The dominant group in the open water around the coral heads and shore reefs is the family Scaridae. The natives eat more parrot fishes than any other group. None are poisonous. The Raroians have names for each species and know their habits and distribution well. They say that there are two main groups--the majority of the scarids grow large, over a foot in length, while others remain small, remaining about 6" in length. Fourteen species were recorded in the field but at least six species were observed that were not taken and probably approximately twelve species occur at Raroia that were not collected. They occur in all marine habitats and great numbers come over the reef flats at high tide to feed. Most of the larger species are speared with four—pronged barbed spears by the natives along the coralline ridge and inter-islet channels. Accord— ing to the natives a few parrot fishes are taken by hook and line. Most of the species are wide ranging in habitat at Raroia, but a few are def- initely restricted to certain coral heads and habitats in the lagoon. Since the family Scaridae is poorly known in the tropical Pacific, the greatest amount of color notes and kodachrome photographs were taken of this group. Scarus forrosus (Valenciennes) .--Characterized by a blue-green rectangle on tail. Life colors of specimen from station 52 immediately after capture: General color of head purplish red, greenish-yellow on nape. Two lines radiating from posterior border of eye: upper blue-green, lower green. Lines on lips extending to eye solid blue, bordered and shading into green. Second stripe below mandible medially blue and distally green. Longitudinal stripe on subopercle margin green. Edge of operculum blue-green immediately before pectoral fin. Dental plates white. Body yellow-green above shading to purplish-red below and to green on caudal peduncle. Above horizontal from lower border of pectoral fin each scale has a vertical green line basally and an orange line at tip; these lines form oblique steps. Posteriorly the green markings become large around spots and the orange stripes are faint. Body at anal base bright green shading to blue posteriorly. Caudal peduncle covered with large green spots with suffuse orange in between. Dorsal fin edge solid blue ending at second to last ray; dorsal base blue-green edged ventrally by a yellow-green band shad- ing into coloration on body. Area between the blue markings {oranges membranes between first two spines purplish-orange. Last 13 rays blue. Caudal fin edged in blue-green above and below. Medially a blue-green rectangle borders the edge of the fin. Remainder of fin reddish-orange, basally shading into caudal peduncle coloration. Anal fin same as dorsal except blue at edge also covers first spine. Pel- vic fins blue except for a red; streak extending from base to tip of ~-117- second and third soft rays. Pectoral fin rays blue-green, reddish at tips of longest rays. Stations: 28(1), 52(1). Habitat: Schools otserved in lagoon around corel keads and along reefs. Raroian neme: TOKATT. Scarus puicnellus (Ruppel.l) .--Characterized by two broad blue~green stripes below lower jaw. No stripes between rictus and eye. Fine green spots over heed and lower half of body. Life colors of specimens from sta- tion 52: General cclor green above shading into orange below. Ven- trally head bright orange. Lips submarginaily bordered by blue-green lines which do not meet at rictus. A short broad blue-green stripe extends behind lower jaw band. Gill membranes medially with a blue- green spot anteriorly and an elongate blue-green line posteriorly. Upper helf of head covered by short green lines and spots in a reticu- lateé pattern. Body dark green above shading into irregvlar green Spots and stripes below. Antericrly each bocy scale has severel dull orange spots and stripes except breast scales whict. are solid crange. Caudal peduncle solid green except for a reticulated orange pattern ventraily. Dorsal fin dark greenish-orange for most of fin; edged in blue-green, besally dark green. Caudal fin narrowly edged in blue- green; longest upper and Lower rays orange. Middle of fin blue-green with orange spots. Anal fin blue-green, with orange reticulations and spots basally extending out to end of fin on last rays. Pectoral fins greenish-black, lower rays with hyalin edge. Spines and first rays of pelvic fins blue. Remainder of fin orange with blue spots on mem- branes. . Stations: 34(1), 52(2), 54(2), 55(1), 58(2). Habitat: Abundant along outer edges of surge channels, coming into surge channels and over reef flat to feed. Raroian name: GNAVERE. Scarus micrchinos (Bleeker).--Characterized by large gibbous forehead, by a suffuse line from rictus across side of head, and by the lack of mark- ings before eye. Life-.colors of specimen from stetion 52: Upper haif of head and body green to blackish-green, lower half yellow-green. Head with broad brignt green stripe bordering lower jaw and extending posteriorly across heed to opercular flap. This line expands poster- iorly, becoming suffuse and dark dorsally. Upper lip narrowly edged with a green line that does not quite meet lower jaw streak. Eye surrounded narrowly by green that extends in broad suffuse band onto nape and posterior region of head. Iris gold. Eyeball metallic blue above, bive-green below. Teeth narrowly edged in white, remainder blue-green, more greenish towards tips. Gill membranes with a median ~118- green streak that is interrupted once. Body dark green above shading into bright green near mid-lateral’ line. Lower half of body abruptly yellow-green. Caudal peduncle all green except for anterior lower portion which is colored like lower half of body. Dorsal fin edged blue-green; base of fin with a narrow irregular green line that extends slightly up on the membrane at the middle of each membrane. Caudal fin green, an orange line borders outer most long ray to tip of pro- longed filaments. Irregular spots and lines vertically in middle third of fin. Middle rays blackish near tip. Anal fins similar to dorsal fin except anterior membranes green in center instead of orange, shading to greenish-orange posteriorly. Pelvic fins with spine blue; membranes of soft rays white in center, blue-green distally; soft rays blue-green. Pectoral fin dark; first ray blue; remainder of fin green to blue-green; upper tip shading to solid black; lower rays edged hyalin; yellow-green across base. Stations: 19(1), 33(2), 47(1),-52(1). Habitat: One of the most abundant parrot fishes at Raroia; common both inside and outside the lagoon in large schools. Paroian names: KOPAHOPAHO (small), TEGATEGA (medium), TONAE (very large). Scarus forsteri (Valenciennes) .--Characterized by dark green teeth, by a broad stripe from behind jaws to opercular border. Life colors of specimens from station 33: General color of head green to bluish- black above. Two green stripes extending posteriorly from eye. Isthmus with a short median green line. Head ventrally brown to greenish posteriorly. Body bluish-green dorsally, green laterally, and browm ventrally. Each scale dark brown basally, forming oblique longitudinal streaks. Anal fin base blue-green on body. Caudal peduncle green. Dorsal fin edged with a bright blue stripe, basally with a narrow bright green stripe; last 8 soft rays with a median longitudinal green . stripe; last 2 rays green. Caudal fin dark; edged in bluish-green °* dorsally, ventrally and posteriorly; medially with a wavy broad brown vertical line extending into each lobe. Anal fin same as dorsal; last 9 rays with a median longitudinal streak. Pelvic fins with first and ‘last rays blue; middle of,fin brow. Pectoral fins dark; first 3 rays blue-green; next 3 rays brown; remainder of fin blue-green, fading to hyalin at edge of fin. fst Stations: 332), 2542), 47d). S401), GCA). Habitat: Abundant both inside and outside lagoon. Raroian name: KOPUMERI (large to small), NOGA (very lerge). Searus sp. one.--Characterized by a black spot on caudal peduncle surrounded a. a light area. -119- Stations: 7(6), Lat), 23(12), 28(4), 4eC3) 4 45(1), 49{2), 61(35), 65(6). Habitat: Abundant both insice and outside lagoon. | Raroian name: PITIKA KONIHO. Scarus sp. two.--General color pure whitish-yellow without any markings on head, body or fins. Teeth white. Lips cover teeth. Stations: 45(9). Habitat: Abundant at southwestern end of Raroia on coral heads and reef flats and in surge channels. It would move onto the outer reef flats in great numbers at low tide to feed. Raroian name: KUKINA (small to large), HAUEKE (very large). Scarus sp. three.--Lips not edged with dark coloration nor are there any dark markings on the head. Lignt orange spots and streaks on head. Life colors of specimen no. 2302 from station 33: Solid green overall. Head uniform green except mottled with lighter yellow-green markings on lower opercular region. Lips same color as remainder of head. A submarginal light orange wavy line present along lower jaw and is not continuous across chin. Numerous light orarige spots and streaks behind mouth and around eye, some radiating from eye for a short distance. Iris gold. Eyeball green. Body solid dark green without spots or streaks as on head. Base of each scale on side of body with a pale orange vertical stripe, becoming spots'‘on caudal peduncle. Breast and body solid green. Scales not dark along caudal base. Dorsal fin green, with a blue-green edge to the fin that extencs to last rays. Last rays not darker than other rays. Base of dorsal fin with an ir- regular line that extends slightly up in each membrane. Blue-green spots and streaks over entire fin. Caudal fin dark blue-green shading to green on posterior margin. Middle of fin with a reticulated network of orange spots and streaks, one streak extending out to the tip of each lobe. Anal fin with a blue-green edge and irregular blue-green basal line similar to dorsal; two irregular longitudinal series of blue-green oblique zigzag lines along middle of fin. Pelvic fins with spine and first ray blue-green; remainder of fin green. Pectoral fin derk; median rays blackish; remainder of fin blue-green; tip of fin hyalin. : Stations: 39{1), 34(1), 35{2), A5(1) 5) 54(1). Habitat: Common in lagoon and along outer-reef, but most abundant next to coralline ridge. Raroian neme: NOGA. Scarus sp. four.--Head and body dark reddish—-browm. Fins reddish—black. Region around jaws red. Teeth pale. Life colors of specimens from -120- Station 57: Head uniformly dark. Opercular flap black. Iris browmish- gold. Region around jaws red, fading to reddish-brow distally. Top of head and back of body almost black. Each body scale brownish—black at base, forming oblique darker lines. Breast and belly same color as remainder of body. Dorsal, caudal, anal and pelvic fins reddish- black. Pectoral fin rays dusky reddish-black, membranes hyalin. Pel- vic fins ene submarginal black line, hyalin Ae Stations: 45(2), 57(2), Habitats: Observed on outer reef flat feeding at high tide and along the shore poets nora of Garue Pass. Raroian name: The natives have a name for this species but could not remember what it is. Scarus sp. five.--Dusky black band behind pectoral on back. Believed by Raroians to be the same as Scarus two but that does not have dusky mark- ings. Otherwise this species is light colored. The life colors of the specimen from station 47 show a narrow black edge to the spinous dorsal and outer long caudal rays. There is a narrow submarginal dusky ver- tical black line at middle caudal rays. Anal fin with dusky black sub- Marginal line. Pelvic fins pale. Pectoral fins pale except for dusky black upper long ray. Upon re-examination of this specimen one week after preservation in formalin it was noted that there was a faint reddish area behind the black body aa this was not. particularly noticed in life. Stations: 21(1), 33(1), 47(2). Habitat: Observed in lagoon along shore. Raroian name: KUKINA. - Scarus sp. six six.--Characterized by a red line from rictus across body to anal eo pa origin. Caudal fin lunate. Small spots present over head above mouth. Life colors of tagged specimen no. 2304 from station 33: Body and head background color bright solid green which is. not darker above, Head with reddish-orange markings which ventrally shade into orange and pink. Upper lip bordered by orangish-red line that extends across operculum in horizontal line to middle of pectoral fin base. Lower border of ‘head solid pink to orange. Symphysis of lower lip bordered with ‘reddish orange, remainder of lower lip bordered by green. Iris orange to gold, yellowish anteriorly and posteriorly. Eyeball greenish- blue. Upper half of head covered with very closely spaced reddish— orange spots that form into irregular.short lines on top of head. Teeth pure white. Body anteriorly covered with very closely spaced reddish-orange spots, approximately 10 to 15 on each scale. Breast and belly orange to pink, with large green spots, abruptly green above. Body solid green below pectoral fin. Dorsal fin green, edged with a blue line, each membrane with a long orange streak, becoming elongate By eae and passing over the posterior membrane on last 3 membranes of fin. Caudal fin edged dorsally and ventrally blue-green. Center of fin with a vertical green crescent edged in blue. Anal fin with basal 2/3 of fin solid orange-red; outer edge of fin narrowly bordered with blue that shades into a green edge. Pelvic fin edged in bluish-green on first spine and ray; middle of fin yellow-orange tinged with red. Pectoral fin blue, yellow across base; first ray biue~green, extreme tip reddish. Stations: 33(1) Habitat: Observed only once at Raroia but natives say it is fairly common around the reefs. Raroian names: TOTOKE. Scarus sp. seven.--Head with blue-green lines on orange backgrounds; upper lip bordered by orange; two blue-green lines under lower lip. No spots “on head or body. Projecting canine at rictus of upper jaw. Life colors of tagged specimen no. 2307 from station 60: Ground color of’ head orange, shading to green dorsally and ventrally. Teeth light orange. Upper lip orange. Lower lip bordered by a blue-green line that extends posteriorly behind rictus a short cistance; another blue- green line is present immediately below. Isthmus with suffuse blue- green, longitudinal mid-line. Eye with five suffuse blue-green lines radiating from it. Those above broad, continuous across interorbital; interrupted above and behind eye into distinct spots. Lower posterior edge of operculum shades into light green. Iris gold, dark blue- green above. Nape and occiput blue-green, particularly intense be- fore dorsal origin. Cheeks and upper posterior half of head orange. Body green, tinged with blue, shading into light green below. Edge of each scale orange to orange-—brown, particularly distinct on back, fading out to solid green on breast; brightest green on caudal peduncle. Dorsal fin edged with a blue-green line extending to penultimate ray. Base of fin with a very narrow blue-green line; posterior region of spinous dorsal and all of soft dorsal with blue-green spots on the mid- die of each membrane. These spots continue on all soft rays, becoming larger and vertically elongate posteriorly. Remainder of fin solid orange. Caudal fin green, tinged with blue; center of outer filaments pale orange, that is connected vertically across middle of fin. Edge of middle ray pale black. Anal fin solid blue-green except for wavy longitudinal orange band near base, that curves out to near tip of fin on last rays. Pelvic fins blue-green ‘on spine and first soft ray; re- mainder of fin pale orange, pale green at tip. Pectoral fins solid blue-green on upper rays, bordered posteriorly by reddish-brown; base of posterior rays blue-green; outer region of lower rays pale yvellow- orange. Stations: 60(2). Habitat: Observed on shore reefs, east side of lagoon. Raroian naine: HOMOHOMO. -122- Scarus sp. eight.--Teeth solid green. Body solid brow tinged with red. Young with white spots on side of body. Life’ colors of specimen from station 63: overall dark brow, tinged with red. Anterior edge of pelvic fin green, remainder more reddish than body. Stations: 33(1), 35(1), 58(1), 60(1), 61(2), 63(1). Habitat: Common in lagoon, but not abundant at any particular locality. Raroian name: KUTU. Scarus sp.’ nin ‘nine.—~AL1 green. ‘Scales on body with white edges, forming ver- an ever oblique lines. Green lines on head consisting of two broad bands extending from’ rictus to eye. This species does not grow over. 6? in ‘Length according to. natives. Life colors of specimen from station 28 one month after preservation: General color of head greenish-vhite. ‘Markings on head solid green. Lips bordered with green and a broad © distinct green band passes from rictus to eye. Another green band is present below lower jaw that is continuous below, ending before a’ ver- ticai from cye. Mid-ventral region of head with a green line from anus to gecen marking below lower jaw. Two green lines extend from poster- ior border’of eye towards operculer: margin, becoming suffuse on body, No markings on top of head. Body green, anteriorly pale greenish- white. Breast and belly with a white band next to mid-ventral green line. Region bélow pectoral fin solid green. Brightest green on cau- dal peduncle. ' Dorsal. fin with a’ green edge that is submarginally bor- dered by a black line. Base of fin with bright green to bluish-green spots. Middle éf. fin pale white, with a pale green spot.in center. Anal fin the Same. Caudal fin with upper and lower rays solid green. Middle’ of fin abruptly pale white except for two irregular vertical green “bands. * ‘Outer region of. middle rays pale white. Pelvic fins an- teriorly and posteriorly green. ‘Middle of fin pale green. Pectoral fin hyalin. Base’ of pectoral fin with a vertical black ‘band. Stations: 13(2), 28(1), 66(2). Habitat: Observed in lagoon and in Garve Pass. Fairly common at Raroia. Raroian name: MAKERE. + Scarus sp. ten.--Dusky white to pure white.. Moderately deep bodied. Natives say this species does not grow more than '4" in length. Life colors of specimens from station’ 23 one month after preservation: Overall dusky white. Head dusky’ black at margins of preoperculum and operculum. Tris gold shading to black above. ° Fins pale white.’ Dorsal and anal Shading to pale dusky black at. “edge, margin narrowly hyalin. Caudal same as dorsal and’ ng in one speécinien without shading in other tWO. Stations: 23(3);, he (aye See Habitat: Observed on coral heads and shore reefs in lagoon. Moderately common at Raroia. Raroian name: PITIKA PUAGARI. FAMILY MULLIDAE Goat Fishes Six species of goatfishes are recorded from Raroia and probably twice as many occur there. Adequate collections could not be obtained and as a result the field data is brief. Most of the species seem to be wide-ranging in various habitats around the atoll from enclosed brackish pools to over the coralliferous outer bench. Over half the goatfish population at Raroia appears to be concentrated in inter-islet channels, particularly on the ‘lagoon side. According to the natives, large adults are most abundant in Garue Pass. Each species has a definite Tuamotuen name. The natives util-~ ize all larger goatfishes for food. Parupeneus bifasciatus (Lacépéde).--Two vertical bands on side of body and blackish saddle on caudal pedumcle. Second dorsal spine strong. Life colors: Head blackish above. Upper jaw, opercles and snout tinged with red. Underside of head white. Nape reddish-black. Vertical black stripes on body moderately suffuse. Blackish saddle on caudal peduncle with considerable red. General body color yellow on edges of scales and orangish-red on bases of scales. First dorsal fin orangish-red, each membrane yellowish anteriorly. Second dorsal black basally and at edge; orangish-red and blue irregular lines alternating in between. Caudal fin black dorsally and ventrally; general color orangish-red with blue spots. Anal fin similar to dorsal with a blackish edge, remainder of fin with alternating wavy blue and orangish- red lines. Pelvic fins anteriorly edged in white, then blackish to tip; remainder of fin reddish-black. Pectoral fin pale yellow. Stations: 33(1). Habitat: Abundant in Garve Pass. ~ Berozen Name: MOAGA. “ab iehaus trifasciatus (Lacépéde).--Black saddle on caudal peduncle pre- ceded by a broad brilliant white area. Life colors of specimens from Station 45: Head and body rose to reddish-orange, shading to pale yellow below. Lips yellow. Barbels pale red basally for two-third the length, tips pale yellow. Operculum reddish-orange. Iris gold, narrowly ringed with red around lens. Body with three suffuse reddish- brown vertical bands that extend ventrally to a level from top of pec- toral fin. Saddle across cavdal peduncle solid black in life, reddish- black after several weeks in preservative, White area between last -124~ pale band and black spot the most conspicuous marking when alive. In preservative rosy with considerable pure white above. Lower part of body pale yellow to white, a pale red streak above anal fin. Base of caudal fin orange-red behind black spot. Spinous dorsal with pale red rays and yellow membranes. Soft dorsal with yellow on outer half and red on basal half. Last ray black. Caudal fin yellow orange. Anal, pelvic and pectoral fins yellow. Stations: 23(1), 45(2), 61(7). Habitat: Scattered individuals noted in lagoon on shore reefs and around coral heads. Raroian name: KAVETI. ‘Parupeneus barberinus (Lacépéde).--First dorsal spines elongate. Dark streak present dow dom back. Caudal peduncle with a black spot. Life colors of specimen from station 33: General ground color pink to reddish; pur- plish above. Head dusky black to purple above, shading to pink below. Only markings on head are bright blue lines around eye. Lips dusky black above to reddish on lower lip. Iris narrowly ringed orange-red around iens. Remainder of iris mottled blue and green and yellow. Blve markings around eye consisting of 7 lines radiating toward snout, across interorbital, and posteriorly across operculum. Blue lines pre- sent on pre-opercular border. Body dusky above, with a pale black longitudinal line extending posteriorly from a level from eye, shading out below soft dorsai. Black spot on caudal peduncle covering lateral line gcale and half of neighboring scales above and below. Breast and belly white. Body below longitudinal line pale red, particulariy near edge of each scale. Spinous dorsal black on first membrane, tips of rays blackish; posteriorly membranes dusky. Soft dorsal membranes pale green and red, rays dusky. Caudal fin dusky pale black. Anal fin hyalin with oblique blue wavy line. Pelvic fin edged with black, first rays pale red, posterior rays hyalin. Pectoral fin dusky pale brown on rays; membranes hyalin. Stations: 33(1), 60(1). Habitat: One young seen on small coral head in lagoon near Teremu Islet. Large adults common in Garve Pass and in 20-40' depths around the bases of lagoon coral heads. Raroian name: TAKIRE. Mulloidichthys auriflamma (Forskal).--Teeth of lower jaw villiform. No dark markings on head or body. A longitudinal yellow stripe present on hody. Life colors of specimens from station 22: Head with suffuse red on snout, occiput, nape and opercles; yellowish behind eye and on lower half of head. Body reddish, with a median lateral yellow band passing through the lateral-line posteriorly. Above and immediately below yellow band there tends to be a pale longitudinal line on each scale -125- row. Vertical fins yellowish-orange except for caudal fin which is pure yellow. Pelvic and pectoral fins yellow-orange with a tinge of red. , Stations: 22(2), 35(27). Habitat: Small schools observed on lagoon shore reefs, and on outer reef flat. Large schools observed in inter-islet channels. Raroian name: KUO. Mulloidichthys samoensis (GUnther) .--General color light yellow to yellowish- orange on head end body. A mid-lateral longitudinal stripe is broken by an irregular black spot above pectoral fin tip. Head with yellow lines radiating from eye, on cheeks and operculum. Dorsal fins light yellow. Caudal fin orange. Anal fin tinged with red, as are pelvic fins. Pectoral fins light yellow. Life colors of specimen from station 58: Lips tinged reddish. Opercular border yellow-orange, operculum with a large pale red area ventrally. Iris silvery yellow. Under- side of head, breast and belly pure white, except for a pale red spot before each pelvic fin. Top of head and back of body dusky, tinged with pale greenish-yellow. Spinous dorsal anteriorly yellow-orange at base shading into red at tip; posterior region of fin hyalin. Soft dorsal rays pale yellow, membrane brighter yellow. Caudal fin uniform orange. Anal fin tinged with yellow. Pelvic and pectoral fins hyalin, rays tinged reddish. Stations: 9(4), 11(1), 16(3), 22(2), 23(12), 28(4), 44(1), 45(2), 49(1), 58(1), 60(2), 61(5), 65(1). | Habitat: Widely scattered in small schools around entire atoll in lagoon and on outer reefs. Raroian name: VETE (large), KOUMA (small). Genus one, sp. one.-—-Probably young of adults already discussed. A free- swimming goatfish blue above to silvery below. Stations: 22(2). Habitat: Small schools observed at night on outer reef flat on west side of atoll. Raroian name: KOUMA (general name for small goatfish). FAMILY CARANGIDAE Jacks Eight species of carangids were taken at Raroia, approximately one- half the number occurring at this atoll. The expedition did not have the es | proper gear to obtain a more complete collection. C09 spear guns and ade- quate hook and line fishing gear would have helped considerably. Carangids are- particularly abundent in Garue Pass and along the outer. reef edge, but also are common airing the iagoon in open water. The natives eee names for each species of Saranela, and know them well. There was no hesitation in naming specimens. The natives have superstitions about spearing carangids and. Bengt Danielsson will include them in his re- port. All the carangids are used for food by the natives. and are avidly speared along the coralline ridge and in the deeper inter-islet channels. Natives also troll for them and occasionally take them by hand line fishing near the bottom. All specimens were obtained by spear and trolling. Decapterus sanctae-helenae (Cuvier).--Six rays in first dorsal plus two be- hind. One finiet behind dorsal and anal fin, operculum with a black spot. Life colors of specimens from station 54: General color sil- ver. Top of head black. A suffuse vertical black spot on posterior edge of operculum. Upper side of body blue. A suffuse iridescent stripe dom mid-lateral region of body, onto posterior keels. Body iridescent silver below. Fins hyalin except outer caudal rays and upper pectoral rays dusky black. First two soft pelvic fin rays with scattered melanophcres. Stations: 54(20). , Habitat: This fish moves into the lagoon at Raroia during certain seasons of the year. It is a food fish prized by the natives, who catch them in stone traps and by herding schools onto shore with woven palm branches. During the period we were at Raroia this species was particularly abundant near Garue Pass in the lagoon on the west side of the atoll. Raroian name: KOPERU. Scomberoides sancti-petri (Cuvier).--Life colors of specimens from station 56 immediately after capture: Iridescent silver overall. Lower part of head, breast and belly silver white. Dorsal fin black, including last soft rays. Caudal fin black except for silver scales which are iridescent blue. Anal fin with a large’ black blotch in middle of anterior soft rays, covering first four rays. Remainder of fin hyalin, except for dusky: black pigment on anterior edge of each posterior soft ray. Pelvic fin with bromish-black streaks on first and second mem- branes. Anterior three membranes and rays otherwise yellow with white edge. Remainder of fin hyalin with white tip. Pectoral fin dusky black, hyalin at posterior edge. Stations: 33(2), 56(1). Habitat: This species is hodeth tel common in the vicinity of Garue Pass. Individuals are often seen in open water around coral heads in the lagoon. Natives take it with hook and line and value it for food. e127 Raroian name: RAI. Caranx adscensionis (Osbeck).--Color almost solid black. Approximately 25 scutes along mid-lateral line. Life colors of specimen: no. 2305 from station 33: Head, body and fins black, without markings. Head and body iridescent. Iris iridescent silver; orbit edged with a black line around the anterior border. Lips black as head and body. Body scutes jet black. Spinous dorsal dusky black at base shading to black at tip. Soft dorsal with jet black rays; membranes pale black at base shading to solid black at tip. Caudal fin pale black at base and aiong outer edge of lobes, shading to solid black postericrly; fin narrowly edged in white. Anal and pelvic fins same color as dorsal. Pectoral fin black... Stations: 33(4), 62(2). Habitat: One of the most abundant carangids at Raroia; widely distributed in the open water inside and outside the lagoon. Particu- larly abundent in Garue Pass. Especially prized by the natives for food. They take this species with spears along the outer reef edge and by hook and line in Garue Pass. Raroian name: RUHI (large), RUHI PEPE (small). Caranx armatus (Forskal).--Twenty-eight soft rays in dorsal fin. First dor- sal and anal soft rays long. Lateral-line gently arched. Body with yellowish-brow spots. Life colors of specimen from station 58: Gen- eral color iridescent silver, bluish above. Body with scattered brow (yellow edged) spots on mid-lateral region of body. Vertical fins black, white edged. Pelvic fins pale black, edged hyalin. Pec- toral fin with upper rays dusky black, lower membranes hyalin. Iris golden, irregularly black at lens border. Stations: 29(1), 58(1), 59(1). Habitat: Fairly abundant in open water outside and inside lagoon. Natives often take this species with spears and valve it for food. Raroian name: NAENAE, Caranx speciosus (Forskal).--Life colors of specimen from station 59: ' Solid iridescent golden yellow over head, body and all fins. Head with black band cbliquely forward through eye. Body with numerous vertical black bands, alternating narrow and broad. Stations: 59(1). Habitat: Rare at Raroia; one specimen seen and taken in surge channels on the west side of the atoll. Raroian name: RERIIOGA. -128- = melampygus Cuvier.--40 scales on straight section of lateral line. ~~ Bateral—Line with a high arch over pectoral fin. 23 soft dorsal rays. Black spots on back. Life colors of specimen from station 35 one month after preservation: Black spots present on tack down to a level from pectoral fin. General color brassy yéllow to black on back. Iris with reddish-orange spot anteriorly and a vertical reddish-orange line behind lens. Stations: 35(2), 58(1). Habitat: Abundant in open water outside and inside lagoon. Natives - often sy SEES species with spears and value it for food. on names: TATAHAUTA (very large), PARUKU (large), PAATHERE (medium, taken from Tahitian), RUPO (onal) Caranx ferdu (Forskal).--Sceutes and shoulder girdle normal. Soft dorsal with 32 rays. Stations: 7{1). Eabitat: Cbservyed and taken at coral heads in lagoon. Natives value this species fcr food and say that it otcurs both inside and outsiae of the lagoon. Raroian name: PAKEVA. Caranx sp. one.--Lateral line gently arched above pectoral fin. Dorsal with 23 soft rays. Scattered black spots over top of head and back of bocy.: Life colors of spécimen from station'63: Head iridescent blue atove, shading to silver below. A few black scattered spots over top and side of head. Iris gold. ‘ips pale black. Body iri- descent blue above shading to silver below. Upper two-third of body covered with irregular black spots. Lateral line scutes yellow an- teriorly. Spinous dorsal hyalin, tinged with blue. Soft dorsal i blue, anterior edge black. Caudal fin yellow on outer rays, bluish to bluish-black on posterior edge. Anal fin blue, with white’ edge. Pelvic fin hyalin, anteriorly dusky. Pectoral fin pale, dusky black above, yellow distally, and hyalin ventrally. Stations: 19(1), 60(1), 62(1), 63(1). Habitat: Abundant in open water in lagoon and around outside of atoll. Valued for food by natives. Raroien name: PARUKU. -129- FAMILY THUNNIDAE Tuna & Wahoo The Raroians said that several kinds of tuna are abundant around the atoll but never caught one during the period we were there, despite daily attempts to do so. Acanthocybium is common about the atoll and approxi- mately five were taken from June to August by trolling. One of the adults was preserved. The natives also say that Coryphaena occurs about the atoll but none were obtained. The natives make their om metal hooks and mother-of-pearl and feather lures for trolling for thunnids. Bengt Danielsson has prepared descriptions of the hooks and fishing methods and will include them in his report. Acanthocybivum solandri (Cuvier) .-~Head and body blue above shading to white on belly. Body with numerous vertical blue bars. Stations: One taken (about 4' long) near Garue Pass on the inside of the lagoon July 3, 1952. A few others were taken by the natives, but were not seen. ; Raroian name: ROROA. FAMILY PARAPERCIDAE Only one species was noted at Raroia, but there may be other smaller forms in the unrecorded material. The natives use the same name for it as for lizard fishes (Synodontidae) but do not use it for food. Parapercis tetracanthus (Bleeker).--Dorsal fin continuous. Black spots present on caudal fin. Life colors of specimen from station 11: ten vertical bands on side of body that are dark grey or brownish-grey. Fins hyalin except caudal fin which has black spots over membranes and rays. General ground color of head and body mottled grey and brown. Stations: 11(1), 14(3). Habitat: Confined to inter-islet channels under rocks in very shallow water. Observed only on west side of atoll. Raroian name: KARAEA. -130- FAMILY BLENNIIDAE Blennies There are more species and individuals of Blenniidae at Raroia than any other fish families, with the possible exception of Gobiidae. Only a small percentage of the material was recorded in the field, totalling 26 species. Probably 10 or more species will be added after more detailed study. With- out a doubt the main center of blenny distribution, both in number of spe- cies and specimens, is on the outer reefs of the atoll on the reef flat and in the surge channels, particularly the latter. Many of the species are strikingly restricted in habitat, and occur only in definite locations. The genus Cirripectes is completely confined to the outer reef’ surge channels, dovm to a depth of about 25'. A few scattered individuals were found in inter-islet channels, but compared to the thousands of Cirripectes seen in the surge channels it is obvious that they had strayed out of their normal habitat. This genus was noted to come up on the coralline ridge at night, presumably to feed. At rotenone stations in the surge channels, the number of Cirripectes killed invariably exceeded that of all other fishes combined, indicating that this genus is undoubtedly the most characteristic form of the surge channels. A few species of Salarias are also restricted to the surge channel zone, but this genus is actually more typical of quieter water. Many spe- cies are extremely abundant in the tide pools next to shore on the outer reefs and in the inter-islet channels. To a lesser extent this genus is abundant in the lagoon around coral heads and shore reefs. I doubt that all species of Salarias occurring at Raroia were taken and the problem of recognizing sexual variation versus species variation is a troublesome one. he blennies are called KOKOROHUE by the natives; none are used for 3 food. Cirripectes brevis (Kner). ~Characterized by 50-60 tentacles in fringe across nape, Black spots separate, not arranged in groups. Life colors _of specimen from station 45: Ground color brownish-green. Spots over head and body. | Iris gold around lens, shading to dusky orange. Spots on head tend to join in short lines. Front of snout with two broad vertical black bands . that ‘cross upper lip--do. not extend on lower jaw or underside of head. Front of snout and lower side of head pale blue. Body darker than head, spots become much smaller than on head. Belly yellowish. Posteriorly ground color becomes brom. Side of body with five indistinct broad vertical darker bands that shade out below lateral-line. Spinous dorsal with a row of black spots directly on base. Remainder of membranes pale black except | for broad oblique an- terior margin. Rays reddish-black, more red on tips of anterior rays. Soft dorsal dusky black overall, a row of black spots on lower part of membranes next to back, tips of first three rays red. Caudal fin men- branes dusky black, the rays black except tips of upper rays red. dnal fin dusky black, outer one-fourth solid black. Pelvic fin dusky Oe ay black. Pectoral fin colored like head on basal half;browish-green with large black spots. Posterior half of rays brown, membranes hyalin. Stations: 45(1). Habitat: Outer reef surge channels on west side of atoll. Only one specimen recorded. Cirripectes leopardus (Day).--Characterized by approximately 32 tentacles in nape fringe. Spots on body often arranged in groups of 4 and 5. Life colors of specimen fram station 5%. Ground esicr lieht biue to pale brom. Head with dark brom spots surrounded by a blue network. Tips covered with small brow spots. Iris gold; dorsal, ventral and posterior margin with a black spot; a suffuse one tnedinal gray outer _Margin anteriorly. Fringe tentacles brownish-black to black. Bottom of head with brown spots similar to side of head. Supra-ocular fringe black with white tip. Body with light browm spots arranged in groups of 4-8 over entire body, including belly. Spinovs dorsal with 4-5 light brown spots on each membrane, the edge dusky black. Edge of first ray black; otherwise membranes hyalin pale brown. Caudal fin orangish—browi, edged in black. Indistinct spots on rays. Anal fin dusky black, with suffuse dark spots. Pelvic fins dusky black. Pec- toral fin with brow spots out middle rays to end of rays. Upper and lower rays dusky black. Stations: 26(1), 27(1), 58(1). Habitat: Outer reef surge channels on west side of atoll. Only three specimens seen. Cirripectes sp. one.—-Closest to Cirripectes jenningsi Schultz. Life colors of specimen from station 17: General color reddish-bromm. Head and body evenly scattered metallic blue spots, those on head tending to be faintly outlined in black. Body with faint vertical bands (approxi- mately 6). Fins without distinct markings. Dorsal pale basally, black- ish on outer half. First two rays reddish-brom. Metallic blue spots on bases of most soft dorsal rays. Lower caudal lobe black. Upper posterior portion of caudal fin pale. Anal fin uniformly pale black. Tips of lower pectoral rays pale black. Tips of pelvic rays pale black. Belly and undersurface of head greenish-white. First two anal rays in males swollen. Stations: 17(55).. Habitat: The most dominant and largest Cirripectes in the surge channels on the west side of the atoll. Cirripectes sp. two.--Perhaps Cirripectes sebae. Life colors prepared at station 17: Ground color brom to reddish-brown. Head brom with light red spots and streaks. Body brow above, reddish below (particularly ~132- over anal fin) with a median longitudinal pale band disappearing on caudal peduncle. Fins reddish, particularly dorsal fin. Spinous dorsal black basally. Soft dorsal reddish--brown basally, brown dis- tally. Anal fin red at base, black for remainder of fin except tip of each ray and membrane abruptly lighter. Caudal fin outer rays red, black in center of fin. Pectoral fin with outer rays red, upper and lower rays browmm. Pelvic rays reddish-brown. Belly and underside of head tan. Stations: 17(10), 58(30). Habitat: Outer reef surge channels on west side of atoll. Cirripectes sp. three.--Life colors of specimens from station 17: General color brownish black. Head with a prominent black spot on preopercle, edged in pale blue. Metallic blue spots on head body. Spots on body arranged in irregular vertical rows. All fins except caudal uniformly black. Caudal with outer rays red and central rays black. Belly and. . ventral surface of head light brow. Stations: 17(71), 58(53). Habitat: The most abundant Cirripectes in the outer reef surge channels on west side of atoll. Cirripectes sp. four.--Life color notes of material from station 17: Gen- eral color brom. Large brow spot on operculum edged in pale light blue. Three pairs of metallic light blue spots on top of head. . Another light blue metallic spot on operculum. Sides of. body anter- iorly with narrow bluish—brown vertical stripes. A large brilliant red patch on posterior part of body and caudal peduncle, preceded by a black blotch dorsally. Black blotch grades into brown stripes ven- trally. Fins without markings. Caudal fin reddish above and below, grading into browmmish-black centrally. Dorsal and anal dark brow. Lower pectoral ray tips dark brow, remainder of fin light. Tips of pelvic rays brow. First two dorsal rays lighter brow, not prolonged. Iris gold, upper and posterior rim black. Nasal flaps red. Stations: 17(2), 58(2). Habitat: Moderately common in the outer reef SEES channels on west side of atoll. Cirripectes sp. five.--Life color notes of material from station 17: Ground color browm overall. Head and body with darker brown vertical stripes which become indistinct ventrally. Between these stripes are metallic light blue spots. Nasal flaps and supraorbital cirri yellowish-orange. First two dorsal rays reddish brom. Remainder of dorsal fin brow to blackish brown on the rays. Membranes light. . Lower part of spinous dorsal membranes dusky black. Caudal fin blackish, particularly middle rays. Outer rays tinged with red. Anal fin black. Tips of rays white. oo First ray glandular. Pectoral fins with tips of lower rays light brown. Remainder of fin hyalin. Tips of pelvic rays dusky brown. Stations: 17(12). Habitat: Outer reef surge channels on west side of atoll. Cirripectes sp. six.--Life colors of specimens from station 17: Ground color light brown with seven darker brom vertical double bars on body. Num- erous metallic light blue streaks and spots over head and body. Young red. Color tends to blend with coralline algae. Supraorbital cirrus branched. Stations: 17(18). Habitat: Outer reef surge channels on west side of atoll. Cirripectes sp. seven.—-Distinctive colors plain brow on head and body. Broad red margin to spinous dorsal. Life colors of specimen from sta- tion 57: Head and body uniform brow without markings. Side of body with a suffuse, irregular, bluish—brown longitudinai band. A light brown band across base of caudal fin on caudai peduncle. Belly blu- ish-brom. Head browm, brownish-white below. Upper lip brown except abruptly light brown for posterior third of length. Interorbital recdish brown. Iris blue. Anterior rim dusky red. Spinous dorsal with red edge; base of fin abruptly black. Soft dorsal black fading to hyalin on anterior rays. Caudal fin dusky black, fading to hyalin at edge on upper rays. Anal fin black, with abruptly hyalin edge. Pelvic fins dusky black. Pectoral fin hyalin above, pale yellowish- black on lower rays. Pectoral fin edged black. Caudal fin rays reddish-black.. Stations: 17(5), 57(1), 58(6). Habitat: Outer reef surge channels on west side of atoll. One specimen found in inter-islet channel above Garue Pass, west side of atoll. Fnchelyurus ater (Gunther).—-Solid black. Dorsal and anal fins joined to caudal. Stations: 7(2). Habitat: Fairly abundant on the sandy tops of lagoon coral heads and to a lesser extent on the lagoon shore reef tops. It lives be- tween depths of 2-25' in holes at the base of coral growing in sandy areas. Salarias zebra Vaillant and Sauvage.-—Body with approximately six sets of double vertical bars. Very dark coloration. Fins black. Cirrus over eye simple. Occipital flap present. -134- ss Stations: 8(31), 10(52), 11(40), 14(5), 22(6), 37(2). ' Habitat: One.of the dominant blennies in the tidepools along the inner edge of the outer reef flat around the atoll and in the inter- islet channels. Rarely seen elsewhere on the atoll. Salarias gibbifrons Quoy and Gaimard.--Distinctive colors consist of black spot on first dorsal membrane. Body with approximately eight vertical .double bars. Life colors of specimens from station 17: Ground color browmmish-white. Head and body reticulated in brownish-red. Vertical bars with light brom between. First dorsal with oblique series of reddish brown spots. Second dorsal with oblique series of black spots. Caudal fin with vertical rows of black spots. Anel fin with two or three suffuse reddish-brown spots, belly white. Underside of head greenish-white. Pelvic fins hyalin. Life colors of specimens from station 58: No cirrus on nape. Numerous red spots on lower side of head and before pectoral fin, but otherwise no red spots on body. Iris green, narrowly edged in gold. Seven reddish-browm streaks from lens. Anal fin blackish distally, each membrane with a narrow pale red streak near posterior edge of each ray. Otherwise basal portion pale. Caudal fin edged in yellow, hyalin except for black spots. No blue spots on body. Tips of first two spinous rays yellow, and submarginally bright red. Pectoral fins with a vertical row of reddish-brown spots near base. Belly white. Dorsal fins hyalin except for described markings. Head greenish above and anteriorly. Upper lip with an irregular longi- tudinal red line and red reticulations, green above at symphysis. Lower jaw uae! in dusky black, red Spens eberall ye Stations: 17(49), 58(12) . Habitat: Outer reef surge channels on west side of atoll. Same distribution and abundance as Salarias sp. 7. Salarias edentulus Bloch.--No canines at corner of lower jaw. General color light brown with 4-6 double vertical bars. A simple cirrus over eye. Mottled red and brillian blue spots on body. Stations: 7(4), 8(84), 9(5), 10(26), 23(1), 28(3), 38(9). Habitat: On eon sand—topped coral heads and sandy shore reefs, occurring between 2-10' depths. Salarias sp. one.~-Supraocular tentacle fringed. Simple tentacle on nape. Black streak behind eye. Sides of body spotted in groups of four without stripes. Vertical stripes down caudal fin. . Dark spots on dorsal fins. sehr : Stations: Dae 22(1). Habitat: Shallow tidepaela, next to ahere on: outer reefs, -west side , of atoll. Not observed elsewhere. th -135- Salarias sp. two.--Caudal joined to dorsal fin. Long nuchal cirrus. Simple ocular cirrus. Caudal fin yellow, with light edge. Dorsal fins spotted, not streaked. Caudal rounded. Life colors of specimen from station 22: Ground color yellowish-green. Head streaked and spotted by numerous reddish lines. Vertical bars on body brownish-orange. Spots on body and dorsal and anal fins orange-red to red. Dorsal fin with green as well as reddish-orange spots. Caudal fin with reddish- orange spots at base, but otherwise lacking markings. Anel fin edged white. Very light black submarginally. Orangish-red spots on basal half of fin. Pectoral fin with orangish-red streaks at base of rays, remainder of fin pale yellow. Belly crossed by six dusky lines. Iris orange, without streaks, three faint orange spots around lens. Stations: 22(1). Habitat: Inter-islet channel north of Garue Pass, west side of atoll. A strikingly beautiful species that is apparently rare at Raroia. Salaries sp. three.--Lower caudal lobe longer; upper ravs truncate. Supra- ocular cirrus simple. Tiny cirrus on nape. Reddish-orange markings over ali. No dark markings. Life colors of specimen from station 22: Ground color yellowish-brown. All stripes and spots orange-red except for a few blue spots. Head with red spots above parallel from lower berder of eye. Lower part of operculum and suborbital region sparsely spotted. Upper lip and tip of snout between lip and nostrils densely spotted. Blue spot behind lower border of eye. Vertical orangish- red bar behind eye. Gular region pale. Iris orange, with an incom- plete (opening at upper posterior corner) orangish-red ring. Body with seven sets of vertical bars; each set with metallic blue spots above and below. Ventrally no spots. First dorsal with short streaks along the rays; membranes hyalin. Soft dorsal lightly specked with orangish-red anteriorly, hyalin posteriorly. Caudal fin without mark~ ings, slightly darker ventrally. Anal fin compietely lacking mark- ings. Pectoral fin with orangish-red:‘streaks basaily. Stations: 22(3). Habitat: Tidepool next to shore on outer reef flat north of Garuve Pass, west side of atoll. Salarias sp. four.—-Distinctive colors black on vertical fins. Paired fins completely without markings. A simple cirrus over eye. No cirrus on nape. Life colors of specimen from station 22: Ground color browmish green. Pale greenish-brown bands on body. Three mid-lateral elongate metallic blue spots on body. No spots anywhere else. Head plain, with a suffuse vertical black band from eye to mouth across upper lip; also a pale black stripe through anteroventral border of eye. Iris yellow-green. Dorsal fins dusky black basally, white distally. Caudal fin white above, black below; the lower rays are longer. Anal fin dusky black distally, black basally, the last rays all black. Pai be ar Bt) SOAS Stations: 22(25). ‘lp ie A br Habitat: A fairly common Species in ane inter-islet channels and in the shallow tide pools: next. to Shore ¢ on the’ outer’ reefs, west side of atoll. Salarias sp. five five.--This species is similar to Sdlarias sp. four, but differs by (1) having a double row of blue spots instead of single spots, (2) having spots ‘in the first dorsal ‘instead of a black base, (3) the pre- sence of a nuchal cirrus in both males and females versus absent. The males of sp. five have ‘the median caudal rays prolonged to a “point. The females have ave the lower' caudal rays longer ‘than the upper. Life colors of specimens from station 22: Ground color greenish—brom, or green. Head faintly mottled’ green. A vertical black streak behind eye. Iris outlined in. ‘black at upper posterior and lower anterior borders. Iris otherwise yellow. A faint dark streak between eye and upper lip. Tip of snout mottled darker. Eight sets of green vertical bars. Two rows of blue spots on side of body, generally five for lower, variable for upper. First dorsal fin light, with two rows of black Spots. Second dorsal brow basally, hyalin for remainder of fin. Caudal fin with a pale black median spot near base of. middle rays, and pale black lower rays. Anal fin as in Salarias Sp. four. Pectoral Pani completely without markings , pale. Stations: 22(10). Habitat: Shallow tide pools next to shore on. the outer reefs, ~-~ ; west side je of atoll. — Sdlarias s sp. Six.--This distinctive species’ is ‘characterized by a “bright -. blue spot ringed in red on the side of the body above the middle of the anal fin. Supraocular girrus simple. Long cirrus present on nape. “Fins pale black at edge. Life colors of the specimen (tag no. 2303) ‘from station 58: Ground color yellow-green, pale to white below. Head bluish-white on throat. Lips anteriorly green shading into pale blue distally. Numerous red spots and short red lines on snout, inter- orbital, around eye, and on operculum. Suffuse. reddish-orange on nape. Iris gold. Eyeball yellow-green. Body with seven brown double ver- tical bands as in Salarias zebra. Region behind pectoral fin and lower Side of body above anal fin covered with red spots. Between each dark vertical band are irregular orange-red lines and spots that fade out anteriorly mid-laterally. Red spots over entire body posteriorly. Dorsal fin with red spots near base of fin. Yellow-green of body ex- tends dorsally onto the fin around the red spots... Both ‘dorsal fins with another series along middle. Remainder of fins pale black with tips of rays almost hyalin. Soft dorsal with red spots over all of fin. Caudal fin uniformly pale black with yellowish tinge; tips of ' rays light. Anal fin same as spinous dorsal. Pelvic fins pale. Pec- toral fins yellow, with orange-red markings on basal third of rays. Stations: 58(1). -137- Habitat: Apparently this species is limited to the outer reef surge channels on the west side of the atoll. Salarias sp. seven.—-Distinctive coloration of marbled red over body, head and dorsal fins. Blue spots along side of body. Supraocular cirrus Simple. Small cirrus on nape. Life colors of specimens from station 58: Head marbled red, except gular region white. Red mottling filled in with pale blue, especially around eye, and upper jaw. Body with eight vertical bands, fading out ventrally; back marbled like head. _ Lower part of sides with approximately three series of longitudinal elongate blue spots. Belly white. Dorsal fins completely mottled red, the circles not arranged in any lines; a black spot near top of first membrane. Soft dorsel narrowly edged in black at tips of rays. Caudal fin pale black on dorsal and ventrai edges, mottled red on middle of fin shading to dusky red at posterior tip. Anal fin pale red for basal half of fin and dusky black for outer half. Pelvic fin white. Pec- toral fin mottled red on basal one-third, remainder of fin hyalin. Stations: 45(1), 58(2). Habitat: A dominant form in the outer reef surge channels living under rocks, in sea urchin holes, and in cracks in the coral. It is fcund from 25' in the bottom of the deeper surge channels to ‘the litho- thamiion ridge at the edge of the reef flat. This species is the same color as the encrusting algae. Salarias sp. eight.--Supraocular cirrus fringed. A large cirrus present on nepe. Body red, with broad vertical bands. Lower side of head with large metallic blue spots. Life colors of specimens from station 58: Head mottled brownish-red. Large metallic silver spots on operculum _ and behind eye. Lips dusky brom. Nasal flaps red. Iris gold, a: black crescent above. Body with seven browmish-red vertical double bars, each bar forked ventrally.. Lower half of side with large metal- lic silver spots. Membranes of dorsal fins hyalin; rays with reddish— brom spots. Caudal fin red. at, base and four vertical rows of brow spots across fin; membranes hyalin. Anal fin membranes hyalin; rays with two longitudinal series of reddishbrowm spots. Pelvic fins hyalin. Pectoral fins reddish. Stations: 58(12). Habitat: Outer reef surge channels on west side of atoll. Salerias sp. nine.--Cirrus above eye fringed. Cirrus present or absent on nape. Pody with longitudinal blue streaks and lines. Adults with vertical fins black. Life colors of specimens from station 58: Sex with cirrus on nape: General color pinkish-red like coralline algae. Head with brow streaks, particularly prominent on bottom of head. Iris gold. Upper lip crossed by seven lines that faintly continue on snout. General head color pale reddish-blue, a pale brownish-white below. Behind eye is a large suffuse blue spot. Body with six sets -138~- ‘of pinkish-red vertical bars which are suffuse-not dominant. Numer- ous longitudinal blue spots and short streaks. on body below a level from pectoral base, comprising the dominant body coloration. Belly pale white. Spinous dorsal pale brownish~black, edge white. Vertical body bands extend onto basal part of membranes. Soft dorsal with six oblique wavy white lines on pale black background. Base of fin same color as body with vertical bands extending up fin. Caudal fin crossed by five vertical black bands which fade out dorsally. Remainder of fin white. Anal fin pale black, becoming darker at tip. Basal part with white spot on each membrane and suffuse white spots out to near tip of fin. Pelvic fins pale, tinged greenish-brown. Pectoral fin pale, tinged greenish at middle, orange-brown at base. Sex without cirrus on nape: ground color bluish to purplish-red. Fins same pat-— tern as other but blacker. Head much darker, including gular region. Vertical streaks on head indistinct. Mid-lateral region along body with a double set of intense purplish-blue spots, corresponding to an intensification of the double bar sets in this region. Stations: 58(6). Habitat: Outer reef surge channels on west side of atoll. Salarias Sp. ten. Dorsal fins separated by a slight notch. Males with a “high crest. Supraocular cirrus fringed. Colors of specimens from station 45 after preservation: Spinous dorsal solid black. Soft dor- sal black with white edge. Caudal fin black shading to a white base. Anal fin dusky to hyalin, a submarginal black band extending length of fin. Body with six pairs of vertical bands. Female with lower fins, no crest, but the characteristic head shape is the same; fins colored the same. No cirri on nape in either sex. Stations: ¥5 (3): Habitat: This species was found out of water on the sides of rocks 3-6! above the outer reef flat of the west side of the atoll. These blennies agilely climb over the vertical rock face and were reluctant to jump back into the water. Enneapterygius minutus (Gunther) .--A vertical black band below eye, extend— ing ventrally. No black marks on operculum. Body with four vertical black bands. 51) aR ECON EP ET i Stations: 7{1). Habitat: Moderately common on sandy bottom’ on tops: and upper Sides of lagoon coral heads. Apparently it lives in holes in the sand at the bases of corals, and to'a lesser extent in holes in the coral. Fnneapterygius sp. one.--Red body with conspicuous bilbek bands across cau- dal peduncle. “Life colors of specimens from station 58: Head brownish- -139- red above, white below. Snout with an irreguiar bromish-red stripe from eye to near tip of upper jaw. A red spot under eye. Operculum and branchiostegal membranes pale black to solid black, not continu- ous below. FPody with three double sets of vertical red bands which do not meet below. Dorsal fins pale red. Cavdal, anal and pelvic fins hyalin. Pectoral fin pale red on rays, membranes hyalin. Iris gold marked with seven red bands radiating from lens. Caudal peduncle © variously solid black to striped with vertical black bands. Stations: 58(19). Habitat: A very abundant inhabitant of the outer reef surge chan- nels on the west side of the atoll. - FAMILY ELEOTRIDAE Gobies The most incomplete field records of ail are for the gobies of the familics Eleotridas and Gobiidae. Since no microscone was evailable, ade- quate -tudy could not be made of these small fishes and most collections of them were not sorted over for field data at all. Thus few habitat notes were made and this category is left off of most species. Six species of eleotrids were recorded and probably twice as many species collected. It is a very abundant family in the outer reef flat tide pool erea next to shore, in the inter-—islet channels, and in the la- goon wherever there is coral, from the shore to depths of at least 40'. The natives pay no attention to them and have no name for then. Gobiomorphus sp. one.--Remarkably distinct color pattern of dark brown, black and white. 3 vertical dark ‘brow stripes on body. Numerous ridges on head. Head, nape and breast not scaled. Stetions: 9(2). Asterropteryx sp. one.~-Second dorsal ray long. Longitudinal black band on back. Spine on preoperculunm. Stations: 7(13), 8(5), 9{16). Trimma sp. one.--General color reddish bromm on body and fins. Head red. Fach scale edged in darker coloration.. Fins relatively. uniform in coloration. Tongue round, not forked. Stations: '7('7). Eviota sp. one.-—-General color bright green with darker vertical stripes. LBA a ee ee Full size one-half inch long. Silver patch behind eye. Pelvic rays © long. ; Stations: 7(10), 9(1), 17(8), 50(1), 58(9). Eviota sp. two.--This species is very small, one-half inch long, transpar- ent. Life colors of specimen from station 53, 11 mm. in total length: General color transparent. No colors seen in fins though rays pale red. Head with a red line from front of eye to tip of snout. Short red line extends from upper posterior margin of eye to upper edge of pectoral fin. This line continues as reddish pigment covering the digestive track. Mandible with red streak along the ventral edge. Body with 13 red stripes on back that extend + the distence dow the side. Ventral series of vertical stripes start behind anus; € stripes counted that extend ¢ the distance up the side. The posterior stripes are somewhat dendritic but meet the upper stripes. Stations: 53{1). Habitat: Taken only by dredges operating at anproximately 40! depths. Genus one, sp. one.--This species has an elongate body with an emarginate caudal fin. Pelvic fins are close together but separate at base. A ack band extends from eye ventrally to the throat. Pectoral fin with a black basal band. Life colors of largest specimen from sta- tion 45: General color of head and body white, markings only on head. Head with a broad black band from eye extending obliquely forward. and meeting below on gill membranes: Iris black above and below, blue anteriorly and posteriorly. All fins white, anal fin distally _ pale orange. Life colors of smaller specimens apparently of opposite sex from station 45: dorsal fin and upper edge of caudal fin edged in black. Pectoral fin with a vertical black band on pectoral base preceded by a white line. Snout dusky black; 2 blue spots on oper- culum behind eye. Otherwise same coloration as other specimens. Stations: 45(2), 61(3). FAMILY GOBLIDAE Gobies The first paragraph under the family Eleotridae applies to the Gobiidae also. Eighteen species of gobies were recorded but this is far from the total number of species taken. Probably about 40 species of gobies occur at Raroia atoll and most likely this is too conservative an estimate. In both numbers of individuals and species it is the largest family of fishes at Raroia and occurs everywhere down to at least 35' depths. Gobies are Vane most abundant in the shallow outer reef tide pool area next to shore and in the inter-islet channels. The greatest variety of species occur on the la- goon coral heads, but not in as great numbers as on the outer reefs. Para- gobiodon completely dominates the fish fauna of the ovter reef flat between but not including the coralline ridge and shore tide fool area. As with the eleotrids, the natives pay no attention to them, and have no names for then. Gnatholepis anjerensis (Bleeker). —~General coloration white to translucent with brow spot. A light brown spot above pectoral fin and another above eye. Station: 7(1). Rhinogobius corallinus Jordan and Seale.--Head broad. Hyes large, meeting on top of head. General coloration light brown, mottled grey. Black spot present on spinous dorsal. Pelvic fins close together and con- nected at base. Station: 8(10). Rhinogobius sp. one.--Third dorsal spine very elongate. Caudal fin ianceo- late. Two series of spots extend down side. Fins hyalin without Markings. Head, nape and breast naked. Stations: 9{1). Bathygobius cotticeps Steindachner.--Upper 8 pectoral fin rays long and silky. Sides of body blotched with dark vertical stripes. Tongue slightly notched. Stations; 8(9), 10(26), 12(1). Habitat: Commonest large goby in shallow outer reef tide pools next to shore and in inter-isiet channels. Its coloration is ex- actly like the gravel and beachrock bottom. Bathygobius fuscus (Ruppell) .--Upper 4 pectoral rays free. Tongue slightly notched. Wumerous longitudinal rows of light spots between suffuse stripes. Stations: 8(4), 11(13), 27(1). ee sp. one.--Upper 8 pectoral rays free and silky. Tongue slightly notched. Wo scales on cheeks or operculum. Life colors of specimens from station 58: General color white and dusky browmmish-black. Head with a black spot behind eye. Lower half of operculum darker than remainder of head and body except for spot behind eye. Snout mottled dusky bromish-black. Underside of head white. Iris gold, shading to pale blue below. Occiput light brow, with a median longitudinal Sues dusky browmish-black streak. Body with 4 saddles on back, extending ~ to mid-lateral region; behind tip of pectoral fin there is a series of 6 spots along the mid~lateral region extending to caudal fin base. Belly and breast white. Edge of dorsal fir dusky black. A broad brownish longitudinal black band extends across widdle of fins. Cau- dal fin dusky, with suffuse brom vertical bands near base. Anal fin with a longitudinal broad pale black band across fin; tips of rays hyalin; base hyalin. Pelvic and pectoral fins hyalin. Base of. pec- toral fin dusky black. Stations: 58(6). Habitat: Outer reef surge channels. Amblygobius sp. one.--Fntirsly white except for black on upper jaw. Peri- toneum shows through belly wall as black. Also a black stripe is present across snout to eye and from eye to operculum edge; a narrow stripe is continuous across front of snouz. Stations: 9(Z). Zonogobius semidoliatus (Valenciennes) .--Numerous vertical blue stripes on head end bocy which extend onto the fins. tations: 8({1). Drombus tutuilae Jordan and Seale.--Tongue round. Distinct ridges on head. General color mottled black and browm. First dorsal with anterior edge dark brown followed by two brown spots. Soft dorsal fin with three longitudinal oblique ciel Stations: 7(1). Paragobiodon echinocephalus (Riippell). --Body and fins black. Head reddish- - brow. Head: covered with hairlike cirri. Stations: 7(1), OCF) Lenses Gobiodon rivulatus (Ruppell).--General colors solid blackish-brom with light belly. No markings. No scales. Gill opening shorter than base of pectoral fin. Some specimens are yellow to white. Stations: 7(2), 17(1). FAMILY GOBIESOCIDAE Cling Fishes A young gobiesocid was taken at station 59, but not noted until the col- lection was sorted at Philadelphia. Probably there are other clingfishes in the Raroian material, but were not noticed in the field. The family Gobiesocidae has not been previously reported from French Oceania. ne FAMILY BROTULIDAE This family is usvally found wherever there are caves in the coral heads and shore reefs. Since there are a great number of caves and long passages in the coral in the outer reef surge channel region, at least one species of brotulid is abundant there. The natives professed to have never seen this group ‘until we showed them examples. Only two species seen at Raroia. Dinematichthys iluotoeteoides Bleeker.--General color usually yelliowish- green to greenish—browm, without markings. Caudai fin separate from dorsal and ana. firs. Stations: 7(3), 17(4), 58(1). Habitat: A dominant species of the sea caves in outer reef surge channel area.on the west side of the atoll between depths of 10-25'. Also occasionaily found around the bases of lagoon coral heads and in caves along the bases of the lagoon shore reefs. Brotula sp. one.--Dorsal and anal fins joined to caudal fin. Head without ‘barbeis. Color pale yellowish-trown. Stations: 23(1), 45(2). Habitat: Taken in outer reef surge channels and on west lagoon shore reefs. Ly Rone 4 5 FAMILY ECHENEIDIDAE . Remoras While several different species of remoras were observed on sharks, only two species were collected. The natives scorn them for food and oc- casionally are bothered by large numbers biting on hooks before food fishes can get to them. The natives have the general term KAKARIURI for remoras. Echeneis naucrates Linneaus.--Characterized by long body and longitudinal dark stripes. Stations: 33(3), 60(1). Habitat: Often schools of free swimming naucrates occur in Garuve Pass, around coral heads and along the shore in the lagoon. The specimens obtained where shot, speared and hooked. Echeneis sp. one.—~Characterized by the presence of.24 plates on the disk, a solid blackish—brown coloration without markings except for longi- dpe tudinal black band through eye. Life colors of specimen from sta- tion 33: head with broad black band from rictus to upper edge of operculum. fins and: sucking disk black. Iris whitish-silver, lens bordered by black above and below on iris. Lips black. Pelvic fins pale black. Body solid blackish brom. genes - Stations: 33(1) Habitat: Natives say this fish often prevents native ‘fishermen: from catching edible fish as it takes a hook so readily. Very abun- dant at pass and north of pass:oan west side. FAMILY ACANTHURIDAE Surgeon Fishes & Unicorn Fishes The fishes of the family Acanthuridae form one of the most abundant families on the reefs, both in the lagoon and along the outer shore. Seven- teen species of the genera Acanthurus, Ctenochaetus, Zebrasoma and Naso were taken, and several additional forms of Naso were seen around lagoon coral heads. The genus Acanthurus is prominent around the entire atoll, but oc- curs in greatest numbers in Garuve Pass and along the outer and inner reefs of the west islets and coral heads. All species are in close association with coral, rarely venturing over sand or gravel bottom;: Acanthurus — triostegus is primarily restricted to the outer reef flats and is one of this area's main "index" species; A. guttatus and the rarer A. achilles are indicative of the surge channels next to the coralline ridge; A. lineatus was seen only in Garue Pass and yet was extremely abundant there; A. nigricans is rather rare at Raroia, and was seen only on the outer reef flat. The natives are particularly aware of the species of Acanthurus and Naso, since they form a moderately important food; the species of the latter genus being considered a delicacy. The natives are rather hazy in naming the less striking acanthurids, and HAMI. is a general name ap- plied to all the less distinctive dark species. Three species of the genus Zebrasoma occur at Raroia. Zebrasoma veliferum is moderately abundant around the lagoon coral heads and lagoon shore reefs of the west side. The natives use this species for food, but do not particularly fish for them. Z. rostratum is common only on the lagoon reefs in the vicinity of Garue Pass. Only two specimens of Z. flavescens were seen the entire summer. The genus Naso is very well represented at Raroia, there being at least _ seven species present; each species, however, is remarkably restricted in distribution, occurring on only definite coral heads within the lagoon and definite areas of the shore reefs. Specimens of Naso are hard to obtain, rarely succumbing to poison, and always very wary of swimmers; two species would bite on a hook, but the remainder had to be speared. The natives were very cooperative in showing:me where each species of Naso occurred, but trying to obtain them was. usually impossible. Naso eoume is taken by . jiggling a dead adult on the end of a twelve foot spear; other adults will CA,5— then be attracted within spearing range, but it is a laborious process. Oniy a few natives at Raroia are skillful enough with this method to spear eoume. The genus Naso is most abundant in Garue Pass. The genus Ctenochaetus is confused with Acanthurus by the natives and the 2 species have no particular names. The natives do not use them for food. Cteno- chaetus strigosus is abundant on lagoon coral heads and lagoon shore reefs of the west side of the atoll. Im order to obtain a good collection of the larger acanthurids, particularly of the genus Naso, a C02 spear gun is indis- pensable. Rubber sling guns do not quite have enough range and power to get the species we lack. Time after time I would spear a unicorn fish, only to see that the spear did not penetrate deep enough to hold him. Many times we would lay a large rotenone cloud over a coral head teaming with unicorn fishes, but they always left at the first whiff, and only a few specimens of Naso lituratus were taken by this means. Acanthurus triostegus (Linneaus).--Life colors based on specimens from sta- tion 58: Ground color dusky greenish-white to grayish-white. Head with a vertical black streak through eye. Above eye brownish black. Below eye pale black. Head whiter than body. Each stripe on body blackish-browm, borcered narrowly by orange~brown. Below pectoral fin are two spots, or a streak, either short, confined to base of pectoral fin, or longer, extending below and behind fin. Iris gold, ringed in dusky black; dark band above, extending almost to lens. Dorsal fin grey, spinous dorsal edged in black. Soft dorsal gray at base to dusky black near edge. Fin edge light. Caudal fin dusky black, its ecge white, to hyalin on central rays. Anal fin dusky black with hyalin edge, darkest near hyalin edge. Pelvic fins white. Pectoral fins hyalin with upper ray edged in biack. tations: 8(5), 9(20), 11(10),14(5), 21(2), 22(12), 23(8), 26(4), 28(2), 42(1), 45(9), 58(30), 65(5), 49(2). Habitat: Very abundant on top of the western shore reefs in the lagoon, but moderately rare elsewhere. This species does not become as large in the Tuamotus as in the Line Islands, seldom surpassing 4" in length. Besides living on the inner western shore reefs, it is also a typical inhabitant of the outer shallow reef flat around the entire atoll, feeding at high tide and hiding along the outer reef face at low tide. Raroian name: AKEGA. Acanthurus guttatus (Schneider).--L fe colors based on specimen from sta- tion 58: Ground color brown with blue spots on posterior half of body, two vertical pale blue lines on middle of side. Head brow anteriorly, a vague orange-brow area from eye to mouth. Operculum bordered in light pale blue that extends narrowly superiorly to nape, joining across top of head. Lips brow. Remainder of juguiar region white. Iris golden, with a ring of browm posteriorly and superiorly. Body orange to orange-brown immediately around pectoral fin. MImmedi- ately above pectoral fin blackish-brown patch. Vertical stripes extend -146- from back to mid-ventral region. Breast shades into white. Belly also white. Region at pelvic base to anal fin yellowish. Posterior half of body behind anterior light bluish line is covered with blue spots narrowly ringed with dark brow; spots present on caudal. peduncle posteriorly to behind a vertical from caudal peduncle spine. Dorsal fins brown basally, covered with blue spots as on body. Spin- ous dorsal narrowly edged in black. Soft dersal with basal one-half ‘solid black penetrated by blue spots. Caudal fin basally pure yellow, outer half of fin blackish browmm. Anal fin covered with blue spots - as on body; basal one-third browm shading into solid black. Pelvic fins bright yellow with narrow black edge; spine and first ray edged in brom. Pectoral fin dusky browish—black; upper ray edged in darker brown. Stations: 10(2), 26(4), 45(1), 55(1), 58(1), 60(1), 65(1). Habitat: Strikingly limited in distribution to the outer reef face surge channels next to the lithothamnion ridge around the en- tire atoll. This species comes over the reef flat at high tide to feed. It is rare in the lagoon. Natives consider this species a great delicacy and eat them raw as soon as they catch them. At low tide this species is often throm up on the dry reef flat by par- ticularly larze waves. Dogs and natives (particularly young children) often go out at low tide looking for them and the natives catch them with their hands as the fish flop about helplessly on the reef flat. Raroian name: KIKITO. Acanthurus achilles (Shaw).--Distinctive colors consist of a red diamond arounc caudal spine and red stripes along bases of dorsal and anal ‘fins. Life colors based on specimen from station 33: Solid black ground color. Head solid black except for a blue edge to opercular opening. Iris solid blue, edged in brownish-yellow around eye ball. Body solid black except for blue edge to operculum and narrow biue line border to red on dorsal fin; triangular patch solid red. Dorsal fin solid black except for narrow red line at base of soft dorsal; soft dorsal narrowly edged in white. Caudal fin white above, below and behind, the red vertical band suffuse with black. Anal fin nar- rowly edged in blue; red streak at base of soft rays. Pelvic fin with first rays bright blue, remainder of fin reddisn black. Pec- torai fin solid black. In the young the body and fins are brown to yellowish brow, instead of black. The red markings are orange to yellowish-orange. Stations; /14(1); \22(1), 33(2),,.45¢7),5, 46(1).,57(2),,.58(6); 65(2), 66(9). : Habitat: This easily recognized species is particularly abundant on the shore reefs in Garve Pass and in the long partially roofed over surge channels on the outer reef flat around the atoll. It is found -147- in small numbers in other habitats wherever there is active coral growth, such as the lagoon coral heads and western lagoon shore reefs. While all the natives know this species as’ "pakurakura" they do not eat it. It is most easily obtained by means of spears. Raroian name: PAKURAKURA. Acanthurus elongatus (Lacépéde) .--Distinctive colors of solid brown with- out pale band through caudal peduncie. Two black spots at caudal and anal base. More than one species may be included here, since there were some slightly different color phases. Life colors based on speci- mens from station 58: Ground color solid brown ‘tinged with orange over head end body. Head with faint vertical orange streaks from eye to mouth and behind to opercular edge. Iris golden-orange behind, pale black for remainder. A complete narrow circle of gold around lens on iris. Region above eye edged in yellow-orenge. Groove before anterior margin of eve pale yellov. Lips black, shading to brown on snout. Jugular region pale brow, with a dark brown patch bordering lower jaw. Body solid color as described above with closely spaced narrow wavy faint blue longitudinal lines. Caudal spine brown, same color as body. Caudal pedurcie posteriorly orangish. Dorsal fin orange-brovw, narrowly edged in tack; with approximately five loagitrvdinal wavy pale iines, Caudel fin blacxish-brown shading to black posteriorly, narrow- ly' edged.white. Anal fin same as dorsal, with two longitudinal pale lines, Base of last rays of both dorsal and anal orange. Pelvic fin edged in black, boin for tip of fin and anterior edze; remainder of fin beownish-orange. Pectoral fin o7vangish, outer edge pale black, elge «f first ray black; base orangish-black. Smaller specimens in same coliection pale across caudal peduncle. Stations: 13(1), 45(20), 57(2), 58(13), 61(3), 65(1). labitat: This species is moderately abundant wherever: there is living coral, particularly being one of the more prominent surgeon : fishes in the outer reef. surge channels beyond the lithothamnion ridge. Large numbers come onto the outer reef flat to feed at high tide, es- pecially on the east and north side of the atoll. . . - Raroian name: OROHEA and HAMI. Acanthurus niericans (Linnaeus).--Distinctive colors consist of yellow band ; on pectoral fin and black band behind ‘eye and before caudal spine. Life colors based on specimens from station 60: Ground color varies : from light gray to dark brovm. All change to dark brow in formalin. Head evenly colored, slightly darker gray or brom on top. Iris \ silver to gold. Eyeball brownish black. Body slightly darker above ; than below. Solid black bar from behind eye to a vertical from mid- Gle of pectoral fin. Another solid black iine ‘extending forward and tapering to a point somewhat behind a vertical from appressed pelvic rays. Caudal spine covered by black skin. Teeth brow, black to greenish-brom distally, narrowly edged in black. Caudal fin white i basally, shading to greyish or brownish black. Middle rays edged in i hyalin, bordered by a black crescent that fades out anteriorly. Caudal £ aie peduncle abruptly white behind spine. Anal fin same as dorsal. Pel- vie fin brownish-black. Pectoral fin brownish-basally shading into black which abruptly is bordered by yellow in a vertical line. Fdge of fin hyalin, the yellow shading ovt to hyalin. Yellow most concen- trated on upper rays. Stations: 23(1), 60(10). Habitat: Very rare at Reroia except on outer reef flats on east side of atoll. The natives seem to know little about this species and only a few natives were certain of its name. It was seen only at the two stations at which it was collected. The natives say it is some- times found around the lagoon coral heads. - Raroian names: MUMU AVAI and AVAI (large adults). Acanthurus lineatus (Linnaeus).--This striking species is characterized by longitudinal blue lines bordered by black. Background color yellow posteriorly, shading to orange anteriorly. Life colors prepared at station 33: Lower side of head and body pale bive to whitish. Upper side with 10 longitudinal blue lines extending onto dorsal fin, bor- Gered by black lines broader than the blue. Head with irregular sets of lines that nave been sketched in field notes. Iris yellow bordered by dusky blue. Dorsal fin with blue lines and black borders extending on base. Remainder of fin biackish-green with longitudinal dusky black lines. Edge of fin narrowly white shading to blue; blue bordered by a black line. Caudal fin dusky black, with a blue U extending out the caudal lobes on the inner margins. Anal fin basally with an orange line that extends out the last anal rays. Remainder of fin same as dorsal except no blue lines, edge of fin the same as dorsal. Pelvic fins edged in black, narrowly bordered by yellow. Remainder of fin bright red-orange. Pectoral fin basaily with three longitudinal blue lines bordered by black; remainder of fin dusky orange. Outer rays dusky black, with hyalin membranes. Stations: 33(6), 66(9). Habitat: Seen only at Garue Pass where it is very abundant on the shallow shore reefs. The natives know very little about this species and could recall no name for it. They do not eat it, although I know this’ species is sold in the central market at Papeete for food. This species can be readily speared; it does not bite on a hook. Acanthurus sp. one.--This species is noteworthy for a white spot below the eye but no pale area on opercular margin. The dorsal and anal bases are yellow. Life colors prepared at station 61: Body and head black. Head with a white longitudinal patch under eye and a narrow white sad- dle behind lips that is continuous below, but not above. Iris green- ish, lens narrowly edged in gold. Dorsal fin basally bright yellow, narrow for spinous and anterior part of soft dorsal, broadening to over half the length of the rays posteriorly. Spinous dorsal otherwise -149- black. Soft dorsal brownish black, black in center; edge of fin very narrowly white bordered by a black line. Caudal fin abruptly lighter than caudal peduncle. Lower and upper rays and outer edge of all rays grey; submarginally a yellow-orange vertical band. The fin is pale brown from band anteriorly. Anal fin same as dorsal. Pelvic fins solid black with dorsal membrane pale grey. Pectorai fin black. Upper part of base white. Caudal spine yellow. A dark blue line across snout, lower edge of operculum and below pectoral fin. Stations: 17(2), 61(1), 66(9). Habitat: This small species is scattered over the atoll, moderately abundant only on the lagoon shore reefs in the vicinity of Garue Pass. It is often in association with Acanthurus achilles. Raroien name: none. It is of no importance to the natives. Acanthurus sp. two.--This species has a broad blue base to dorsal and anal fins which is completely lost in preservation. Color notes prepared one week after Paes eg e4 of specimens from station 47: Solid black on head and body except for yellow-green spot behind eye on large specimen and suffuse yellow-green before and after eye on smaller specimen. Dorsal with narrow blue before base that in life is bril- liant, shading into the black of the remainder of the fin. In larger specimen there is also a blue patch on middle of last soft rays. Spinous dorsal with orange—orowm irregular streaks and spots on men-— branes extending length of spinous fin. Caudal fin solid black. Caudal peduncle spine suffuse black, surrounded by a narrow intense black area. Anal fin largely blue, in life intense blue at base; longitudinal blue streaks near edge surrounded by blue-black and basally blue stripes extend on the membranes; edge of soft fin an- teriorly edged in blue; posteriorly edged in black; submarginal area at posterior angle yellow green; anteriorly fin has orange-brown ir- regular streaks. Pelvic fin basally blue with orange-brown streaks out membranes. Distally orange browm streaks dusky and pelvic fin mostly black. Pectoral fin solid black. ; Stations: 47(2), 60(9). Habitat: Scattered over entire atoll around coral heads and shore reefs, but abundant only in eastern channels between islets. This is the largest Species of Acanthurus at Raroia and could only be taken by spears or hook and line. - Tuamotu name: AVAI. The natives use the adults for food. Ctenochaetus strigosus (Bennett).—-This species is solid dark brom with longitudinal wavy lines. Teeth long and movable, bristle-like. Life colors prepared at station 57: Head without markings except for tiny orange spots over top of head from snout to around dorsal fin base, down to around eye and on operculum. Body with numerous narrow longi- tudinal wavy blue lines extending length of body. Breast solid brow, -150- but lines.start below pectoral, fin; sides of belly with blue lines. Caudal peduncle with blue lines to a vertical from caudal peduncle spine. iris gold, ringed with bive. Dorsal fin brownish—black, nar- rowly edged in black to.vpper angle of soft dorsal. Soft dorsal with numerous straight blue lines as on body extending to posterior edge of fin. Caudal fin: wmiform, dark as body. Middle rays narrowly edged \in hyalin, bordered by a black crescent. Anal fin same-as dorsal. Pelvics same rer, as Body. Pectoral rays DROW, membranes hyalin. ee a eae! 76), 9(1), 23(4), 28(5), 33(1), 45(3), 49(1), BAS) 60(2), 61(24), 65(1), 66(8 de Habitat: Abundant around atoll eid i there is active coral growth, pa particularly on Jagoon coral heads. Tuamotu name: Actually none, although the natives refer to it by the general 1 name ame "HAME." The natives do not eat them. Ctenochaetus sp. o1 Sneneiats odiee taken at station 66: Adult solid black ~ except ot for bright yellow around eye.and pure white tail. Smaller Specimens (9£ mm. £.L.) have reddish-orange on side in region of pectoral fin. Pectoral fin yellow distally. Pelvics tinged with red and basal percvion orange-red. Remarkable for its long movable bristle- - teeth and deciduous scales. I was cut on the hand. by the caudal spine of the largest specimen after it was speared. Immediately there was acute pain in the hand and arm. Swelling and a painful throbbing ‘started within minutes after the cut. Within an hour my hand was so swollen I could bend the finger tips only slightly. In the meantime I had:cut the wound in an attempt to flow the poison out. My arm was - Slightly swollen and ached clear to the shoulder when I moved it. Within two hours after the cut the wound was treated at the village. The hand remained moderately. swollen for a day and completely subsided by the second day. ‘The pain remained at its peak. for a day and did not disappear until after a week had passed. I had been’cut by prac- tically all of the other surgeon fishes at Raroia, but none caused more than an immediate irritant, except for Ctenochaetus strigosus. The effects of the poison from C. strigosus on two occasions lasted for approximately 3-4 hours, but the pain was much less than in C. sp. one. Stations: pot). Habitat: This eet species is characteristic of the large lagoon coral heads with abundant free-growing algae.. It lives on the flat coral tops, feeding at high tide and moving dow the sides of the coral heads at low tide. It never was taken by rotenone, despite re- peated attempts and is very difficult to spear, being wary and extremely agile. Itis moderately abundant on. the ‘shallow reefs oS Garue ees: in gerrics Gee with Zebrasoma rostratum. — Baedicn" name: None, although the natives are. very afraid of their caudal spines and will not spear them. The natives do not use them for food. -151- Zebresoma veliferum (Bloch).--Coloration same as specimens observed in Line Islands. Color notes will have to be taken from kodachrome slide taken at station 42. Stations: 9(1), 23(1), 28(1), 42(1), 66(1). Habitat: A typical species of lagoon coral heads and western lagoon shore reefs, although never more than ten seen at any one locality, or more than 20 on any coral head. This species was never observed where there are strong currents (e. ea. Garue Poe. inter—islet channels) or on the outer reefs. Raroien name: Well known to the natives as PAKERERO; they some- times eat this species, but do not particularly fish for them. They are hard to spear, because they are wary and never stray far from coral growth. Zebrasoma rostratum (Gunther) .--Life colors prepared at station 33: Solid black body and fins. Caudal spine white. Stations: 33(1), 66(10). Habitat: This species is rare at Rharoia except on the lagoon shore reeis immediately south of Garue Pass. A few individuals were also seen on iégoon corel heads. It was found in association with Ctenochaetus strigosus. Reroian name: None. The natives pay no attention to it. zebrasoma fiavescerns (Bennett) .--One specimen definitely observed in the Naso yeliow phase on a lerge coral pooe in the lagoon $ mile from Oteremu Islet. It was approximately 34 ~2,"" long. Its coloration was solid yellow without any black. It was with the only yellow specimen of Insidiator epibulum I saw at Raroia. The life color of the specimen of Z. flavescens taken at station 23 was solid black. One other black specimen was observed on lagoon shore reefs immediately south of Garumaoa village. Stations: 23(1). Habitat: Only three observed during the entire summer so believed to be rare at Raroia. The natives say they seldom see this species. Raroian name: none. tuberosus Lacépede.--Distinctive characters short flat compressed "horn" on forehead; blue spots and iines on body; long caudal fila- ments. Life colors prepared at station 60: General body and head color dark grayish brow. Head with a blue-black suffuse stripe for- ward from eye. Blue spots on head below and above eye. Lips blue- black. Body with numerous vertical blue wavy lines on middle of side. -152- a Sides of breast and belly with numerous small blue spots extending posteriorly over middle of anal fin. No markings around spines other than blue lines and spots. Above blue lines on side larger blue spots same size as on head. Blue lines on body behind middle of anal fin extend to anal fin and curve forward ventrally. Dorsal fin solid brow, blackish brown basally. Caudal fin black basally, to’ blackish brown on outer one-fourth;. edged in white. Anal fin brow, basally with short irregular bive lines, one for each membrane, crossing the poster- ior spine behind. Eye ringed with blue. Iris narrowly ringed blue, most of iris brom. Operculum edged in dark blue.’ Pelvic fins brom, edged in black. Pectoral fin blue-black, a blue line at base on lower rays. At station 33 it was noticed that the blue markings were bril- ' liant blue, especially the splash marking behind the pectoral fin. Naso Dorsal and anal edged in blue. Stations: 33(7), 60(5). Habitat: Very restricted at Raroia, according to the natives and my ovn observations, to a few lagoon coral heeds approximately 2 kilometers northeast of Oneroa Village and to Garuve Pass. it is very abundant in the Pass where it is taken by hook and line.: Raroian name: KARAUA; it is considered an excellent food fish by the natives, who principally take them with spears. eoume (Lesson).--Life colors taken at station 43: Overall greyish black except for white caudal fin. UDorsal fin brownish. Caudal fin black edged; middle of fin white, basally dusky. A vertical pale gray band through middle of white area. Anal fin blackish. Body with numerous vertical rows of tiny spots that oes to form into lines in some specimens. Stations: 43(9), 45(5). Habitat: Very restricted in distribution at Raroia, according to the natives and my own observations to a few lagoon coral heads in the region of "Makoto," 14 miles east of Tetakaga, west side of atoll. These fishes are eee wary, completely avoiding our attempts to kill them by rotenone, or spear them by normal means. They also do not bite on a hook. ‘These unicorn fishes remain about 40-100' away from the sides of the coralheads, remaining motionless in schools ap- proximately 10-20' from the surface over water approximately 80-150! deep. For reasons I cannot guess they primarily gather on the north side of the coral heads. No other fishes seem to associate with them, although occasionally they wovld forage for food at the base of the coral head 40-100' from the surface. Only adults were seen. The natives prize this species as the best eating of the acanthurids and have devised a way of spearing them. A twelve foot single—pointed spear is used, and a dead aduit N. eoume is jiggled on the end of the spear. The ace usually stays sat the surface on the edge of the coral head, thrusting the dead adult at the nearest school, which will ~-153- usually be 8-20' beyond the end of the spear. Some of the school will be attracted by the unusual antics of the dead adult. When a less cautious individual gets close enough, the native will pull the spear out of the decoy with a quick jerk and‘ in one motion lunge at the other, spearing it before it can get away. This method of spearing is very difficult and only a few natives at Raroia can do it. The natives that have mastered this method Mayet hy | can: spear about 4-5 in an hour. Raroian name: TATIHI. Naso annulatus (Quoy and Gaimard).--Life colors prepared at station 33: Solid giay in life with a faint orangish band at pectoral base and a faint orangish area around spine on cavdal peduncle. Dorsal and anal fins with oblique yellowish-orange and blue streaks. Stations: 33(1),..51(1),,,.67(1).. Habitat: Solitary individuals were observed in Garue Pass, and on ne western lagoon coral heads and shore reefs. Not more than three were ever’ seen at any one time. Only large adults seen. ‘Raroian name: UME. The natives spear them for food. Naso lituratus (Bloch).--Distinctive coloration orange around caudal ‘spines, orange. streak from mouth to eye and prolonged caudal filaments. Life colors taken at station 60: General color of head and body gray. Head blackish gray on top of snout and before dorsal'fin. Lips solid orange. A yellow line from front of eye to behind rictus. Yellow across entire interorbital and over post and supraocular region where it is tinged with green. Iris silver, outlined in black. Eyeball lined with a horizontal blue line above iris which is bordered by black dorsally. Anterior nostril surrounded by a blue ring; narrow blue line above posterior nostril. Teeth brow. Body solid dark gray above shading into. yellowish gray on body. Operculum yellowish. An- terior caudal peduncle spine with a solid orange saddle, continuous below but not quite joined on top;. posterior spine with a similar- sized patch which is not continuous around peduncle above or below.. Spines solid orange with blackish tips. Dorsal fin solid black; edge of spines narrowly white and an irregular white spot on each membrane near tip that becomes larger in posterior spines. Soft dorsal narrowly edged in white bordered by a fine black line, then a broad white line. Basal two-thirds of fin black. Caudal fin solid gray; middle rays edged hyalin, then narrowly grayish black. Next to this coloration is a yellow green crescent which ends before the upper and lower fila- ments start. Anal fin narrowly edged in white, bordered by a narrow black line; remainder of fin basally yellow green shading haif way out the fin into brownish orange. Pelvic fin pale yellow-orange, anterior rays bordered in black. Pectoral fin black, white hyaliin tips to the membranes, shaded with yellow, which is particularly dominant at the base of the last rays. ad Stations: 7(1), 23(1), 28(5), 33(2), 42(1), 60(1), 66(1). Habitat: Widely distributed wherever there is active coral growth on the atoll, but particularly abundant on the lagoon coral heads and west shore reefs. This is the only species of Naso that was taken by rotenone, yet not more than one or two were ever killed at a station, despite 10-25 being present in the poisoned area. The natives ac- tively fish for this species with spears, it being a moderately im- portant food fish. Larger young to adults commonly observed. Smaller specimens never seen. Raroian name: -TAREI. FAMILY ZANCLIDAE Moorish Idols _ Zanclus cornutws is found in all eight habitat zones (see Ecology) but is rare in surge cnennels, and absent from inter-islet channels except “or a few in Garue Pass. The natives do not eat them. Zanclus cornutus (Linnaeus).--No spine present at corner of mouth. Life colors of specimen from station 61: Vertical bands black. Body yellow between bands. Head broadly black through eye superiorly to dorsal fin base. Two pale bluish-white bands extend across interorbital from eye. Lower half of snout black. Top of snout with a median black line from snout tip to interorbital. Orange saddie on top of snout bordered by black. Light region on head pure white. Caudal peduncle yellow, biack line bordered by a narrow white line. Dorsal fin pale black at base anteriorly; tip of long filament white; posterior edge of fin narrowly black, submarginally white. Edge of caudal fin hyalin; broad vertical black band on fin bordered anteriorly and posteriorly by a white line. Anal fin pale black anteriorly; posterior edge black bor- dered submarginally by white; last rays yellow at base. Pelvic fins black. Pectoral fins hyalin, upper base orange. “Stations: 45(2), 61(1), 66(2). Habitat: Moderately common at Raroia but extremely difficult to obtain. This species is widely scattered throughout lagoon on shore reefs and around: coral heads. It is most abundant on western shore reefs. Fewer individuals live in pot holes and roofed over surge troughs on the outer reef flat and in Garue Pass. In the lagoon it is often in association with Heniochus. Raroian name: PANAPANA. ~155- .| FAMILY SCORPAENIDAE ~ eee Beshee ” At least ten species - the fasidany Seorpactiiiee | were tuand” ‘at Raroia, but identification of the smaller specimens was*so confusing that only part of the collections are recorded. The family is most abundant on the outer reef flat, particularly in sea urchin holes behind the coralline ridge. To a lesser extent the group is common in surge: channels and on lagoon shore reefs, but is very rarely seen on lagoon coral heads. The natives fear all of them as being venemous, but actually only Pterois, Scorpaenodes, Scorpaenopsis gibbosus and Synanceja appear to really belong in this category. The natives do not eat any of then. Pterois volitans (Linnaeus) .--Life colors of specimen from station 61: Head and body with brown vertical bands separated by white or reddish-—white narrow interspaces. Head dark brown above, dorsally covered with reddish-white narrow lines. Swpraocular flap striped black and white. Snout white except for two brown spots bordering upper jaw. White flap at side of mouth distally and posteriorly edged with orange. Lower preopercular edge with two orange cirri with a white line dom the center. Jris black above and below, otherwise golden to brownish- golden. Underside of head white. Breast white, crossed by two pale lines. Belly white except for bands continuous across ventor. Spin- ous dorsal membranes pale brown; dorsal spines crossed by broad black and reidish-white bands. ‘Soft dorsal hyalin except for approximately five black cross bands on rays. ‘Caudal fin hyalin except for 5-6 vertical series of spots as on soft dorsal. Anal fin similar to caudal, with white dots more prominent. Pelvic fins with reddish- white rays; membranes solid pale black. Pectoral fins with rays reddish-white; besal section of rays with three black to brown bands; outer portion of membranes with large round black spots, not more than two per membrane. Stations: 10(1), 16(1), 22(1), 44(1), 6(1), 66(2). Habitat: Outer reef flat under soe Observed nowhere else at Raroia. _ Raroian name: TATARATHAU. orpashodes parvipinnis (Garrett). ——Dorsal spines very short. Scales ex- tending onto soft dorsal and anal fins. Head and body coloration mottled like coralline algae. Life colors of specimen from station 17: Ground color light brown to.lavender and pink. Head with dark brown stripes. Iris intricately’ striped with orangish-red and brown. Eyeball yellow bordered by greenish=brom. Body with a broad ver- tical brown band that is bordered above and below by pink markings. Caudal peduncle with a dark brow vertical stripe. Soft dorsal, anal and caudal fins with yellow spots. Caudal fin bright yellow. Pelvic Dania -156- fins dark brown with pink and yellow markings. Pectoral fins crossed by a vertical browmish-purple stripe near base; remainder of fins yellow with a brignt yellow spot. Stations: 17(1). Habitat: Observed in sea urchin holes behind coralline ridge on outer reef flat. Raroian name: none. Scorpaenodes sp. one.--Thirteen dorsal spines present. Two black spots on soft dorsal. Long flap on side of upper jaw. Life colors of speci- men from station 61: General color bright orange with bromnish to jet black markings. Head bright orange on interorbital region, tip of snout and above upper iip. Jaw flap striped black and orange. Iris marked with 6 broad black bands that extend almost to lens border. Lens border narrowly ringed in black crossed by numerous brown lines. Top of snout dusky black. Side of operculum also dusky black. Lower helf of head orange mottled with dusky black. Lins dusky orange. Body crossed by five broad suffuse vertical bands; various scales pure black to orangish-browmm; interspaces orange, except white be- tween last two vertical bands. Spinous dorsal basally orange where body interspaces continue on fin; membranes bluish-gray, distally with one or two longitudinal white lines; spines distally crossed by three or four black bands. ‘Soft dorsal with two brilliant black spots sur- rounded by orange and black rings; basal region of soft dorsal scaled and colored. like body. Caudal fin membranes hyalin; rays crossed by approximately six black spots. Anal fin edge white; middle of fin , crossed by a white longitudinal line; basal region of fin gray, darker anteriorly. Outer region of fin with approximately six suffuse - black spots. Pelvic fins dusky black, edged in white, and with a _ basal white spot posteriorly; middle of fin shades to white on pos- terior rays. Pectoral fin brilliantly marked with four vertical black lines alternating with orange; lower simple rays crossed with white and black bands; base of fin orange-—brom. Stations: 10(1), 61(1). Habitat: Known only from outer reef flat under rocks. Raroian name: TATARAIHAU. Scorpaenopsia gibbosus (Schneider) .--Large pit present under eye. Head and body mottled grey and white. A vertical grey band present on caudal fin. Two dark bands extend across pectoral fin. Stations: 12(1). Habitat: Only one specimen noted from shore reefs next to Garu- maoa Village in lagoon. Natives say the dorsal spines are poisonous. Raroian name: PUGA PUGA VEVE. -157- Scorpaenopsis sp. one.~--Lateral line with 35 scales. Palatine teeth absent. No pit below eye. Ground color mottled overall, blending with coral- line algae of coralline ridge. Upper region of back green. ‘Fins reddish. Head mottled with grey, lavender, green, red and _ ahgilitig No distinct bands on body. All fins also mottled. ‘Stations: 7(2), 8(2), 9(2), 17(2), 61(3).. Habitat: Hides in holes in coral along reef flat and in lagoon. Raroian name: none. Scorpaenopsis sp. two.--Palatine teeth absent. Thirty lateral line scales. Pit indistinct below eye. hi Stations: 17(2). Habitat: Noted from. sea urchin holes behind coralline ridge on reef flat. Reroian name: none. Scorpaenopsis sp. three.--Body with broad’ light brom vertical band. Brown streaks radiating from eye. Caudal peduncle crossed by a vertical “brown band. Life colors of specimen from station 58: Ground color light brow. Head pale whitish-brom, somewhat darker on top of head. Eye with four dusky black lines radiating from its lower margin across lips end suborbital region. First three lines ending on upper jaw. Mid-dorsal line with a black spot on snout, a brown spot anteriorly on interorbital, and three suffuse vertical lines across posterior re- gion of es oe Occiput with a suffuse vertical line and spots on nape. Body with two suffuse superior bands ending above lateral- line. A brownish black spot is present behind upper angle of opercu- lum. Most of body covered with a single broad brown vertical band that extends to belly; posteriorly it ends below last soft dorsal rays at end of anal fin base. Base of caudal fin crossed by an irregular vertical brown band. Spinous dorsal coloration similar to body, with brown spots on the spines. Soft dorsal reddish-browm basally, with six series of oblique brown spets on rays; membranes hyalin at tip. Caudal fin regularly spotted with brown on rays; membranes hyalin. Anal fin basally red, edged in brown; middle of fin with eight brom spots; edged hyalin. Pelvic fins with a few scattered brow spots, pale red to hyalin at tip. Pectoral fins with 2-3 vertical rows of brown spots; base yellow; membranes hyalin. Stations: 45(6), 58(1). Habitat: Common in surge channels. Raroian name: none. -158- Scorpaenopsis sp. four.--Suborbital spine extending down over side of upper - jaw. Head and bor body covered with numerous black spots. Life colors of specimens from station 58: Ground color brownish-grey. Body: covered by brownish-black to black spots, irregularly arranged so that there are no lines or bends. Belly white except for a few brown spots. Head covered by tiny black spots.. Iris gold, outiined with an. orange-red . ring. Dorsal fin mottled like body, edged in orange-black. Soft dor- sal shading to hyalin distally. Caudal fin with approximately four. . vertical series of:-browmish-black spots on rays.and membranes; edge of fin pale red... Anal fin with scattered browmish—black: spots, tip of fin hyalin.: Pelvic fins with suffuse brown spots, remainder of fin hyalin, Pectoral fins with six series of brow spots following the contour of the fin. Stations: 45(2), 58(3). Habitat: Observed in surge channels. Raroian name: none. Taenianotus triacarnthus Lecépéde.—-One specimen solid black. The other specimen taken at station 66 was mottled like coralline algae. Stations: 66(1). Habitat: Two specimens collected from under rocks on outer reef flat aid on shore reef in Garve Pass. Raroian name: none. Synenceja verrucosa Schneider.--Grey in color. Body as broad as Cceep. Head broader than deep. Stations: 12(2). Habitat: Observed only in the lagoon on the western shore reef flat. This species hides in holes under rocks and in the coral dur- ing the day; at night it comes over the flats near shore to feed. They are completely unafraid and refuse to move, even when poked. When disturbed they remain perfectly stiff.and can be: pushed or hit without making any movement. However, when they have:had: gree they make a slow cameuanins retreat. Aiubg Raroian name: PUGA PUGA. FAMILY CARACANTHIDAE This family is one of the most abundant in the cracks and holes of the outer reef flat, but is restricted to very shallow water between the inner ~159- edge of the coralline ridge and inner beachrock tide pools. It is found | in close association with Paragobiodon in this reef flat region. Occasion— ally, specimens are taken elsewhere around the atoll on lagoon shore reefs or on lagoon coral heads. It is rarely seen in a habitat over two feet deep. The natives have no name for this group, being too small to be of any consequence to them. Only two species seen at Raroia. Caracanthus unipinna (Gray).--Dorsal fins fully united without a notch. Life colors of specimens from station 57: General color grey to black-— ish-brown, head and breast lighter. Dorsal and anal fins dark grey at base, sheding to light grey distelly. Caudal fin grey, somewhat lighter at tip. Anal and pectoral fins light grey. Stations: 17(1), 57(3),. 58(3). Habitat: As described under the famify. Caracanthus maculatus (Gray).--Dorsal fins separated by notch. Life colors of specimens from station 57: General color of head and body grey, fading to greyish-white below. Spots on head and body reddish—trom anteriorly to brow posteriorly. Lower side of head and body without spots. Iris gold around iens to black distally. 411 fins pale at base, shading to dusky black at tip, except caudal which is pale black With a hyalin edge. Pectoral fin with reddish-brown spots on base and lower part of fin. Stations: 17(2), 57(4), 58@) jwowa). Habitat: As described under family. FAMILY BOTHIDAE Flat Fishes Only one species of flatfish was seen at Raroia, and the natives said there was only this one there. Bothus pantherinus (Ruppell).--Eyes on left side. Life colors of speci- mens from station 58: Extremely variable in color but with stable pattern. Ground color ranging from brownish-white to brown, gray or black. Head and body covered with irregularly outlined circles with pale centers. Head orange to reddish-brown between eyes. Numerous small blue spots around eyes. Fins mottled like body with smaller circles. Edge of dorsal and anal fins hyalin. Pectoral fin with long filamentous upper rays in larger specimens. Three black spots on body on horizontal line from eye. Anteriormost black spot large, approxi- mately same diameter as interorbital width, situated immediately above pectoral base. Second black spot approximately as large as eye, sit- vated equidistant between first spot and end of dorsal fin. Third black spot often faint, near caudal peduncle below end of dorsal fin. -160- ae Stations: meen TA 6)i,) PECL psZo C2 ii23( a ipiQ2(G)rw23( a inies (2), BALE) :\'45(2) 5 AL), ‘58(4), 65(1). Habitat: Abundant on the outer reef flat around the atoll and to a lesser extent on the gravel and sand inter-islet channels. Rare in the lagoon, but occasionally occurring on the cones tops of lagoon coral heads and shore reef's. FAMILY BALISTIDAE Trigger Fishes Eight species of trigger fishes were taken and I doubt if more than one or two other species occur at Raroia. The natives asserted that we had col- lected all the different kinds occurring at this atoll. All species except those of the genus Balistapus are abundant at Garve Pass and immediate vi- cinity; they also abound in deeper water on the outside deeper reef shelf, but are rare in the lagoon. The genus Balistapus is very abundant on the outer reef flat around the atoll, and to a lesser extent is found wherever there is coral in the lagoon. Almost all the trigger fishes make grunting sounds when caugnt, the genus Balistapus making the loudest noise. All species, exceot those of the genus Balistapus, take a hook and are eaten by the natives. . Balistes vidua Richardson.--Distinctive coloration of caudal fin rose-red. Pectorel fin yellow. White dorsal and anal fin edged in biack. Life colors prepared at station 35: Head and body dark brommish-black overall. Iris dusky golden. Lips dark brown. Teeth white. Eyeball jet black. Spinous dorsal brownish—black. Dorsal and anal fins edged anteriorly and distally by narrow black band. Remainder of fin pure white. Caudal fin narrowly edged dorsally and ventrally by black. Caudal fin rose-red, somewhat fading at posterior'edge. Base of fin and posterior part of caudal peduncle white. vi Stations: 33(1). ‘Habitat: Only one specimen in Garve Pass taken. The natives say it is fairly common in the Pass and along the outer deeper coral shelf on the west side of the atoll. None seen other than the example taken. Raroian name: POPOGA. Balistes undulatus Mimgo Park.--Distinctive coloration of black around caudal peduncle spines. Orange streak extending from mouth to below pectoral fin. Life colors taken at station 33: Stripes on body bright orange, those around mouth brightest. Lower jaw with bright orange around mid- dle. Tip of upper jaw orange. Otherwise lips blackish~brom. Iris pale green, eyeball brow. Lens narrowly ringed in gold; radiating lines from lens; on pale green eight suffuse orange lines; on browm -161- 10 blue lines bordered by black lines. Generel color pattern of num- erous orange lines: 6 lines from eye (counting one above and below) across interorbital; 16 lines from orange spots below to middle dorsal rays. Spinous dorsal brom, black at tip of first membrane. Soft dorsal and anal with orange longitudinal line at base. All rays pale bluish gray at base, remainder orange. Memlargnes hyalin. Caudal fin with outer rays orange, middle rays yellow. Caudal peduncle with ver- tical orange lines. Pectoral fin with orange rays, pale black at tips, membranes hyalin. Stations: 33(3). Habitat: Collected only in Gerue Pass. An adult was observed on a lagocn shore reef on the northeast side of the atoll. Raroian name: KOKIRI KARAVA. Balistes fuscus Schneider.--Distinctive colors consist of black fins with yellow-orange edges (except spinous dorsal which is solid black). Life colors prepared from an example taken at station 33. General color dark grey. Iris greenish gold. First dorsal dusky black. Dorsal and anal black to brovmish black, fin edge yellow-orange. When viewed through light dorsal has a basal pale browm longitudinal line. Lower half of fin has series of four spots on each membrane in a vertical line that tend to form into lines posteriorly. Anal fin similar. Caudal fin black to dark brown, the postermigr edge of the middle rays are yellow-orange to hyalin at tips. Pectoral fin black with paie browmish yellow margin. Stations: 23(2). Habitat: Scattered individuals seen around lagoon coral heads; it is most abundant in Garue Pass.. Raroian name: KUTARO. Balistes capistratus Shaw.--Distinctive coloration of solid brow overall except lower lip orange, with a pale brown line above and below lips. Numerous rows of small spines on posterior part of body. Life colors taken at station 33: Lips grayish purple. Posterior edge of lower lip orange. Pale line around iips brow, slightly orange below. Iris brow, somewhat golden anteriorly. Pectoral fin green, outer edge abruptly dusky biack. Stations: 33(2). Habitat: Seen and taken only in Garve Pass. The natives very rarely see this fish and know little about it. Raroian name: none, other than the general trigger fish name of KOKIRI. -162- Balistes sp. one.--Characterized by seven horizontal rows of spines on caudal peduncle. Ali fins with black band near edge. Black:streak around eye. Life colors of a specimen from station 60: General color slate-grey with the white in the caudal showing up as the most distinctive marking in life. Head with black streaks bordering upper jaw, and medially on snout through eye, gill opening, onto base of pectoral fin. Lower lip brom. Eye distinctly marked with numerous black lines. Lens golden green with 6 broad lines radiating from lens. Eyeball dark grey with 20 wavy black lines radiating from cornea. Spinous dorsal dusky black. Soft dorsal and anal white with a broad black band near tip. Last rays dusky black at base. Caudal fin edged with a broad black band. . Upper and lower long rays black; entire ba- sal area dusky black; middie of caudal fin white. Pectoral fin white, a broad black band near tip of fin. Edge of pectoral fin white. Stations: 15(1), 27(1), 33(4), 60(1). Habitat: This is the largest trigger fish at Raroia, and indivi- duals are found in Garue Pass, in. inter-islet channels with swift moving water {at high tide) and along the outer Ceeper coral shelf on the west side of the atoll. The natives spear it for food and cc- castionally take it on hook and line. Raroian name: KOKIRI PAGO. Balistepus rectangulus (Schneider) .-~Distinctive coloration of black diamond on tail and black oblique stripe on body.’: Life colors will have to be prepared from kodachrome Seal taken mg stations 31 and 45. Stations: monte 2) m(1), 45) Habitat: This species is ae ee restricted to the outer reef flat around the atoll on which it feéds at high tide. At iow tide it hides under rocks or in holies in the dead coral. It is about one-third as abundant as Balistapus aculeatus, and much more restricted in distribution than the latter species. The natives take them un- intentionally oe line but do not SH eer Raroian name: none, other than the general trigger fish name KOKIRi. Balistepus aculeatus (Linnaeuvs).--Distinctive coloration consists of oblique vertical stripes without black patch on gaudae peduncle or black ob- ligue band as in B. rectangulus. Stations: 10(1), 14(4), 22(8), 24(1), 31(19), 45(3), Seth); 60(2).. Habitat: A dominant species of the outer reef flat around the atoll. Scattered individuals occur around all the coral heads in the lagoon and on the lagoon shore reefs. This is a strictly shallow water form. tei Raroian name: KOKIRI HORATAHORA. -163~ Melichthys buniva (Lacépéde).--Distinctive colors blue lines on dorsal and anal fin bases. Pectoral fin black. Life color notes prepared from specimens taken at station 33: Solid black over all. A narrow bril- liant blue line along base of dorsal and anal fins. A thin blue line is submarginal to middle caudal rays. Iris golden black. Lines across interorbital faint. Teeth white, tinged with purplish brom along middle. Pectoral fin solid black. Stations: 33(4). Habitat: This species is fairly abundant in Garve Pass and along the outer deeper coral shelf on the west side of the atoll. Raroian name: HOPUPU. FAMILY MONACANTHIDAE File Fishes Three species were noted from Raroia and probably several other species occur there in deexer water. This family prefers hiding places under rocks and coral on the outer reef flat. It is expected that it is also common on the corallifcrous outer bench. It is uncommon in the lagoon, although the natives say tney are abundant in Garue Pass. The natives do not eat them and they are usually referred to as KOKIRE, the goneral Raroian term for trigger. fishes. The application of the name KOKJPE KARAVA to two of the monacanthids and not to the very sin- ilar third species seems strange. This discrepancy was checked with sev— eral natives and appears to be correct. i Cantherines pardalis (Ruppell).--Head, body and caudal fin solid dark brom. Dorsal, anal and pectoral fins with brom rays and pale brow membranes. Caudal fin brownish black, distally shading darker than body. Lips pale brown. Teeth pure white. Iris blue. Eyeball tan. Stations: 27(1).: Habitat: Observed on outer reef flat of west side of atoll. Raroian name: KOKIRE KARAVA. Cantherines sp. one.--Life colors of specimen from station 46: Head, body ‘and all but posterior margin of caudal fin uniform slate grey without spots or markings. Dorsal, anal and pectoral fins completely hyalin. Posterior edge of caudal except upper and lower lobes white (not hy- alin). .Iris light green. Lips light grey. Life colors of specimen from station 58: Dorsal, anal and pectoral fins pale yellow. Posterior ~L64- region of caudal fin yellowish before hyalin edge. Iris yellowish- green. Faint yellowish lines from eye extending obliquely forward. Numerous faint yellow spots from snout. Stations: 15(1), 16(1), 26(1), 45(5), 46(1), 58(1). Habitat: A few individuals noted in lagoon but primarily observed in holes in the coral and under rocks on the cuter. reef flat. Reroian name: KOKIRE. Cantherines sp. tvo.--Solid brow with orange fins; four forward projecting scaies on caudal peduncle. Life colors of specimen from station 45: Body solid brom. Lips pale yellowish-brom. Iris pale blue ringed with orange. Dorsal, anal and pectoral fins pale orange. Caudal fin browm, posterior edge orange. Caudal peduncle spines orange. Stations: 33(1), 45(1). Habitat: Noted hiding in holes on shore reefs in Garue Pass and under rocks on outer reef flat. Reroian name: KOKIR=E KARAVA. FAMILY TETRACDONTIDAE Puffer Fishes Seven species of puffer fishes occur at Raroia and apparently all spe- cies existing there were taken. The genus Tetreodn is most common on the outer reef flat and to a less extent in the inter-islet channels. The genus Canthigaster is mostly confined to lagoon western shore reefs and inter-— islet channels. Occasional individuals were noted on lagoon coral heads. The family Canthigasteridae is included under the family Tetraodontidae in the field notes and has not been separated in this report. Tetraodon meleegris Schneider.--Solid black with light blue spots on head, body and fins. Edges of fins hyalin. Life colors of specimen from station 52: Edge of pectoral fin bright yellow-orange. Edge of dorsal fin light yellow to hyalin. Edge of caudal narrowly white, remainder black with light blue spots as on other fins. Anal fin edge yellow- orange. Teeth white. | Stations: 10(1), L7G), MB) 2208), BIC). SC le AS aya ead 52(1), 60(1). Habitat: Common on outer reef flat particularly of west side of atoll. At low tide they hide among coral and rocks and come out to feed at high tide. Raroian name: KOPARIPARI MORORI. -165- Tetraodon regani Gunther:--Life colors of specimen from station 52: Ground color black with numerous light blue spots over head, body and caudal fin. Dorsal fin pale yellow-orange, with obscure pale black reticu- lations; edge of fin hyalin. Anal fin similar to dorsal. Caudal fin with spots and black background almost out to edge of fin. Stations: 11(1), 19(1), 33(2), 52(1). Habitat: Not as common as Tetraodon melLeagris; almost entirely confined to inter-islet channels. Occasionally noted on outer reef Piet. Bia Reroian name: KOPARTPARI. Tetraodon sp. one.--Head and body yellow above to white below. Vertical fins pale yellow. Pectoral fins orangish-yellow with brow streaks between rays. Left side with a browm spot above gill opening. Above notes of specimen from station 22. Stations: 22(1). Habitat: Only one observed and captured on outer reef flat, west side of atoil. Raroian name: KOPARIPARI. Canthigaster cinctus (Richardson).—Top of head and body crossed by four vertical black bands. Stations: 9(1), 23(1). Habitat: Moderately uncommon at Raroia. Noted only on shore reefs, west side of lagoon, in region of Garumaoa Village. Raroian name: HUE TETE. Canthigaster solandri (Richardson).--Head, body and caudal fin covered with blue spots edged in black. Ground color brow shading to whitish be- low. Throat yellow. Blue lines edged in black radiating from eye. Eyeball green. Iris ringed with three metallic orange and blue bands. Two of the lines radiating from the eye cross the interorbital. Stations: 9(5), 11(1), 12(1), 25(1), 57(1), 58(1). Habitat: Moderately common at Raroia; definitely the most abundant species of .Canthigaster at this atoll. It was mainly noted in the western inter-islet channels and to a lesser extent in the lagoon on the west shore reefs and coral heads. One specimen was taken on the outer reef flat. Raroian name: HUE TETE. ae) ; } 1 | . Canthigaster bennettii (Bleeker).--Ground color greenish-brom. Snout and belly with indistinct yellow spots. Most prominent marking on body is a black blotch at the base of the dorsal fin. A broad yellow-orange band extends from behind pectoral fin to lower region of caudal peduncle. Blue spots present on upper side of caudal peduncle. Blue lines radiating from eye but none extending across snout or interorbi- tal. Stations: 9(12), 12(2), 23(1). Habitat: Uncommon at Raroia. Noted only on lagoon shore reefs in vicinity of Garumaoa Village. Raroian name: HUE TETE. Canthigaster janthinopterus (Bleeker).-~Sides of body brow with large blue spots. These spots become larger on belly and underside of head until there are brown reticulations instead of spots. Life colors of speci- men from station 66: Top of head with suffuse markings. Two longitu- dinal brown stripes on middle of interorbitai. Head with brown reti- culations {somewhat orange around eye) that become narrower ventrally. Between reticulations pale blue. Iris ringed with orange around lens; remainder blue-green. Body gradually changes from reticulations an- teriorly to spots on a browm background dorsally and posteriorly. Dorsal base black. Fins hyalin with pale black rays. “Stations: 66(1). Habitat: Several specimens noted on shore reefs in Garue Pass. Raroian name: HUE TETE. FAMILY DIODONTIDAE Porcupine Fishes Apparently only one species occurs at Raroia, but.there was some dis- agreement among the natives on this point. The natives ‘relish it for food and prefer to cook it whole in a fire, spines and all. Then they peel off the skin, and eat the flesh. . Diodon is not poisonous at Raroia, nor did the natives know it to be poisonous on the neighboring islands. ~ Diodon hystrix Linnaeus.--Life colors of specimens from station 47: general Color light below, white shading to dusky black above in one specimen and to browm above, in the other. Spots in both specimens solid black. Stations: 47(2), 60(2). -167- Habitat: This species prefers the broad sand flats at the southern end of the atoll and usually occur in pairs. They are uncommon at Raroia. Raroian name: TOTARA. FAMILY OSTRACIIDAE Box Fishes Two species were collected at Raroia and I doubt that others occur there, except natives say the genus Lactoria occurs in the lagoon. Box fishes are moderately uncommon at Reroia and all examples were seen deep in holes in the coral or under rocks. Rotenone did not kill them. The najority of the field records note them from inter-islet channels. The natives do not use them for food, but do not consider them poisonous. Ostracion sebae Bleeker.--General color black with orange spots and markings. Life colors of specimen from station 52: Head and body solid black, . grey beneath eye to a vertical from pectoral fin base. Large blotch of orange on interorbital and superior orbitai socket. Iris black, with large orange spots, superiorly metallic blue spots between orange markings. Body with orenge spots on side, a few on bottom of caranace neer lateral angle in region of pectoral fin and anal fin. The orange spots tend to become irregularly elongate near edge of upper carapace angle. Top of body with small evenly spaced white spots from posterior margin of eye to end of dorsal. fin. Caudal peduncle with large irregu- lar orange spots that tend to join together. Dorsal fin with black rays, hyalin membranes. Caudal fin black. except for a crescent shaped posterior hyalin area. Anal fin with bromish-black rays and hyalin mem- branes. Pectoral fin same as anal. — ae Stations: 52(1). Habitat: Oniy one specimen observed and collected in inter-islet channel north of Garue Pass. Raroian name: TUTUKE. Ostracion lentiginosus Schneider.--General color brown with yellow fins. Blue spots on body ringed in black. Life colors of specimen from sta- tion 51: General color brown to yellowish-brow ventrally and poster- iorly. Head with scattered black spots, particularly around orbital border. Iris silver. Body with small white to bluish—white spots surrounded by a black ring. Blue spots confined to sides and top of carapace. Bottom of carapace in head region with black spots, other- wise carapace bottom markless, except for a spot on each side before © anus. Black spot present on dorsal fin base. Caudal peduncle uniform yellowish-brom. Fins yellow-orange except for dusky irregular black stripe in upper posterior section of tail; tips of outer caudal rays dusky black. -168- Stations: 36{1), 38(1), 42(1), 51(1), 60(2). Habitat: Moderately common at Raroia. Apparently confined to lagoon shore reefs, inter-islet channels and occasionally on outer reef flat. Raroian name: TUTUKE. FEMILY ANTENNARITIDAE Angler Fishes Two antennariids (apparently different species) taken at Raroia under rocks in channeis between islets. The natives seldom see this species and confuse it with Synanceja, calling it PUGA PUGA. Antennerins sp. one.--This species has a long slender ilicium. Caudal fin with vertical black stripes. Head and body mottled white, grey, green and black without spots, olending with the gravel and dead coral. Stations: 22(1). Habitet: This frogfish was found among dead coral fragments in shallow water (6" deep) under a rock in an inter-islet channel. It remained perfectly still when disturbed, and blended perfectly with its background. ~169- CHECK LIST OF THE ISLANDS OF THE TUAMOTU ARCHIPELAGO While attempting to prepare a bibliography of all known fishes collec- ted in the Tuamotus it became apparent that there was considerable confusion concerning the names of the islands and island groups. When we attempted to check the standard sources we discovered that none were completely reliable, and many were filled with confusing errors. As a result we found it neces- sary to prepare a new list based on H. 0. chart number 0077 (January 27, 1951) and H. 0. Gazetteer number 7 (1944). Brigham (1900), Wilkes (1845), Picquenot (1900), and Freeman (1951) are the principal sources utilized. In the course of compiling this check list we learned that the confu- | sion was greatest east of the cannibal line (see Wilkes, 1845, chart of | Paumotu Group prepered in 1839) northwest to southeast through the center of the Tuamotus. The probabie reason for this situation is because few ex- peditions have entered that area of the Tuamotus and hostility of the natives in the nineteenth century prevented landing parties from examining most of the islands. ; ‘I. Island Groups. Actaeon Is. (21 20 S 136 30 SS is. Maturei-Vavao (Atoll, 21 22S 136 23 W) = = elibunace Acteon I. (Wilkes Exped.), Estancelin.— Tenararo (Atoll, 21 20 S 136 43 W) =Acteon i. “(Wilkes Exped.), Bedford. Tenarunga (Island, 21 20 S 136 30 W) = Minto, Acteon I. (Wilkes Exped.). Vahanga (Atoll, 21 20 S 136 40 W) = Bedford, Vehanga (Brigham). Disappointment Is. (14 08 S 141 20 W) =TIles du Desappointement, Disappointment Is. of Byron: Napuka (Atoll, 14 08 S 141 14 W) = Wytoohee (Wilkes Exped.). Tepoto (1405 S 141 24 W) = Otooho (Wilkes Exped.), Tetopoto. Duke of Gloucester Is. (20 40S 143 20 W): Anuanuraro (Atoll, 20 25 S 143 33 W) = Archangel (of Quiros), Anu-anuraro, Anu-Anuraro, San Miguel Archangel, Heretutea, Heretua (Wilkes Exped.). Anuanurunga (Atoll, 20 38 S 143 19 W) = Teku (Wilkes Exped.), Four Crowns (of Quiros), Quarto Coronados. ~170- Nukutipipi (Atoll, 20 42 S. 143.05 W) =Margaret, Nukutavake - (lapsvs, Brigham chart). Gambier Is. (23 10 5. ee aeaed Agakavitai (Island, 23 10 S 135 01 W) = Agakanitai (Brigham) . ieee (Island, 23 12'S’ 134 rh W) = Wainwright. ‘Aukena (Island, 23 08 S 134 54 W) = Elson. | Kamaka (Island, 2315 S 134 57 7) s Collie, Colles Makapu (Islet, 2312S 134 55 W). "eae (Island, 2314S 134 58 W) = Marsh. Mangareva (Island, 23 07S 134 58 W) = Peard, Manga Reva. Manui (Island, 2314S 134 56 W). _ Taravai (Islend, 23 09 S 135 02 W) = Belcher. King Georges Gp. (14 35 S 145 00 W) =TIles du Roi George: Takapoto (Atoll, 14 38 S 145 11 W) = Oura (of Cook), Taputa, Ura. Takaroa (Atoll, 14 27S 144 58 W) = Takoroa, Tiokea. Palliser Is. (15 31 S 146 38 W) = Shadelyk, Pernicious: Apataki (Atoll, 15 25 S 146 20 W) = Hegemeister. Arutua (Atoll, 1518 S 146 44 W) = Rurick (of Kotzebue). Kaukura (Atoll, 15 43 S 146 40 W) = Aura. | Toau (Atoll, 15 54S 146 01 W) = Elizabeth. Sea Gull Group (16 44S 14415 W) = Raeffsky Is.: Hity (Atoll te" 44"S” 174° 08 3h) = Eliza, Tipotu or Bacon (Wilkes Exped.). Tepoto (Atoll, 16 49 S 14417 W) =Eliza (of Mauruc), Ofiti, Ohiti or Clute (Wilkes Exped.). Tuanake (Atoll, 16 41 S 14414 W) = Reid, Tuinaki, or Ried - (Wilkes Exped. ya -171- Two Groups (18 07 S 142 10 W) =Manaka: Marokau (Atoll, 18 07 S 142 13 W) = Daushaida, Dawhaida (Wilkes Exped.). [North Atoll of Manakal. 3 Heeb ee 1148 = 10 W) = Dawhaida, Manaka ~~ (Wilkes Exped.). [South Atoll of Manakal._ Vairaatea (19 20 S 139 20 W) = Egmont, Tatakoto (lapsus, Wilkes ny Pukararo. (Atoll, 19 20 S 139 20 W) =Tres Cootiers, Tatakotu (iebeus, Wilkes er ), @res Cocotiers (du Mauruc). Pukarunga (Atoll, 19 20 S 139 20 W) = Tatakoto (lapsus, Wilkes). | II. Islands not in Groups. Ahe (Island, 14 30 S 146 20 W) = Peacock, Ahii. , Ahunui (Atoll, 19 40S 40 25 W) = Byam Martin Pinaki or Whitsunday (lapsus, Buicnen), Nganati (Brigham), . Anuni (H. 0. Gazetteer). Aki-Aki (Atoll, 18 30 S 139 13 W) = Thrum Cap, Lancier, Vahitahi (lapsus Brigham, descr.), Les Lanciers , teeters Pukerua (lapsus, Wilkes Exped.). Amany (Atoll, 17 50 S 140 45'W) = Moller. Anaa (Atoll, 17 25 S.145 32 W) = Chain I. Angatau (Atoll, 15 50S ‘140:50 W) = Arakchev, Arakcheef, Ahangatiu, or Arackchuff (Wilkes bia e . Aratika (Atoll, 15 32 5 M45 31 W = Karlsholl, Carlshov, Carlshoff, Karlshoff. | ur Encarnacion (Approx. 24 44 S 136 40 W) = Wilkes Exped. Faaite (Atoll, 16 45 8 145 13 W) = Mylardowitich, Faiti. Fakarava (Atoll, 1618S 145 36 W) = Wittgenstein. Fangahina (Atoll, 15 59 S 140 07 W)'= Fakahina, Fakaina, Akahaina, Predprieetige (Wilkes Exped.), Predpriatie. Fangatau (Atoll, 22 15 S 138 45 W) = Cockburn, Fangataufa, Ahunui (lapsus, H. 0. Gazetteer and Brigham). Hao (Atol1, 18 13 S 140 54 W) = Bow, La Harpe, Harp, Hau. -172- Neanaiti (Wilkes Exped.), Haraiki (Atoll, 17 30 S 143 30 W) = Croker, St. Quentin, Heraiki.- Hereheretue (Atoll, 19 50 S 145 00 W) = St. Paul, San Pablo (Wilkes Ex— ped.), Tematangi (lapsus, Brigham descr.) Hikueru (Atoll, 17 36 S 142 40 W) = Melville, Bird, Tekukota (lapsus, Wilkes Exped.) . aes Kativ (Atoll, 16 26 S 144 21 W) = Saken. Kauehi (Atoll, 15 51 S 145 10 W) = Vincennes, Kaveh, Kawehe. Makatea (Island, 15 52 5 148 13 W) = Metia, Aurora, Dageraad. Makemo (Atoll, 16 37S 143 41 W) = Phillips, Makima, Makemu (Wilkes Exped.), Koutousoff (of Bellingshausen). Manihi (Atoll, 14 20S 145 59 W) = Wilsons, VWaterlandt (of Schouten and Tenaire), Oahe, Manhii (Wilkes eped. Vis Manuhangi (Atoll, 19 11 S 141 16 W) = Cumberland, Menuwangi (Wilkes Ex- ped.), Mehuagi. Maria (Atoll, 22 00 S 136 10 W) = Moerenhout, Maria Reef (Wilkes Exped.). Matahiva (Atoll, 14 50 S 148 40 W) = Lazareff, Mataiwa .(Wilkes Exped.). | Minerva (Reef, 22 40 S 133 30 W) = Ebrilles, Bertero, Berbero. Morane (Atoll, 23 00 S 13710 W) = Cadmus, (Wilkes Exped.). Motutunga (Atoll, 1705 S 144 22 W) = Motw Tunga, Adventure. Mururoa (Atoll, 21 50 S 138 50 W) = Matilda, Osnaburgh. N. Marutea (Atoll, 1700 S 143 11 W) = Furneaux, Marutea (Wilkes Exped.). Nengo Nengo (Atoll, 18 46 S 141 48 W) = Pr. William Henry, Nae een nos L'Ostange, Lastang (Wilkes Exped.). Niau (Atoll, 1610S 146 21 W) = Greig. Nihiru (Atoll, 16 43 S 142 50 W) = Nehirwu or Niori (Wilkes Exped.), Negeri, Nigeri, Niheris Nukutavake (Island, 19 40 S, 138 42 W) = Queen Charlotte, Nukutapipi (lapsus, Brigham chart), Vahitahi (lapsus, Brigham descr.), Akiaki (lapsus, Wilkes Exped.). Paraoa (Atoll, 19 08 S 140 40 W) = Hairi, Gloucester. -173- Pinaki (Atoll, 19 20 S 138 40 W) = Whitsunday, Tematu- leiwowau (Wilkes Exped.), Nukutavake (lapsus, Brigham chart), Ngenati (lapsus, Brigham descr.). Portland Reef (Reef, 23 40 S 134 30 W). Puka-Puka (Atoll, 14 50 S 138 50 W) = Pukapuka, Honden, Hondau, Henauke, — Dog, Tatakoto (lapsus, Brigham), Clerke (lapsus, Brigham). Pukaruha (Atoll, 18 19 S 137 02 W) =Serle, Searle, Pukerua or Thrum Cap Akiaki (lapsus, Wilkes Exped.), Reao (lapsus, Brigham chart). Rangiroa (Atoll, 15 03 S 147 40 W) = Rahiroa, Vliegen, Deans, Nairsa. Raraka (Atoll, 1611S 144 53 W). Raroia (Atoll 16 03 S 142 23 W) = Barclay de Tolley. Reao (Atoll, 18 30 S 136 23 W) = Clermont-Tonnere, Natupe, Clermont de Tonnere, Pukaruha, Serle (Lapsus, Brigham). Reitoru (Atoll, 17 49 S 143 08 W) = Bird, Hekueru (lapsus, Wilkes Exped.). Rekareka (Island, 1650S 141 54 W) = Good Hope. St. Juan Baptist (Approx. 24 S. 38 59 W) = Wilkes Exped. S. Marutea (Atoll, 21 30 S 135 40 W) =Lord Hood (s), ieruten (Brigham). Taenga (Atoll, 16 20S 143 17 W) = Holt, Yermaloff (of Bellingshausen). Tahanea (Atoll, 16 52S 144 45 W) = Tchitschagoff, Tchigschagoff. Taiaro (Atoll, 15 44S 144 36 W) = King Takume (Atoll, 15 48 S 142 12 W) = Wolkonsky, Wolchonsky, Takurea or Wolconsky (Wilkes Exped.). Tatakoto (Island, 1720 S 138 22 W) = Narcissus, Clerke, San Narcisso, Pukapuka or Clerke (lapsus, Wilkes Exped.) Tauere (Atoll, 17 22 S na 28 W) =St. Simeon, Resolution, Tawere (Wilkes Exped.), Tandrec, Taveri. Tekokota (Atoll, 17 20 S 142 37 W) = Tekokoto, Doubtful, Tekareka (Wilkes chac age Exped. ) Tematangi (Atoll, 21 40 S 140 40 W) = Blighs, Hereheretua (lapsus, Wilkes Exped.), Bligh lagoon. -174- Tikahau (Atoll, 15 00S 148 10 W) = Krusenstern,- Krusenshorut, Tikehau. Tikei (Atoll, 14 58 S 144 32 W) =Romanzoff, Tike or Romantzoff (Wilkes Exped.). Timoe (Island, 23 20 S 134 30 W) = Crescent, Temoe (Wilkes Exped.). | Tureia (Atoll, 20 48 S 138 32 W) Exped.). i Papakena, Carysfort, Tureie, (Wilkes Vahitahi (Atoll, 18 44 S 138 50 W) = Lagoon, Four Facardins, Cook, Nukutawake or Teay (lapsus, Wilkes Exped.). Vanavana (Island, 20 45 S 139 10 W) = Kurateke, Barrow, Teku (Wilkes Exped.). ~175- NATIVE USES OF FISHES Food Resources: While the 127 Raroians have an unlimited supply of food in the sea only approximately 50 species of fishes are commonly used for food and the natives are rather selective even among these preferred species. The nunm- ber of kinds of fish that the natives will occasionally eat totals higher, but acculturation gradually makes fishing seem less desirable to them. According to the natives many of the food fishes at Raroia are sea- sonal. Also, in their detailed accounts of the individual fishes they in- dicate that many fishes have seasonal movements around the atoll. ‘The period at Raroia was too short for us to. check these notes further. The natives use no conservation measures and are under the impression that there are just as many fishes as ever before. There are very few tabus relating to fishes (except concerning one species of carangid and several tetraodontids ané balistids to be reported upon by Bengt Danielsson). No seasonal restrictions or minimum size limits are involved. Food fishes are as plentiful in the immediate vicinity of the main village (Garumaoa) as elsewhere and the natives have no trouble catching then. Catch Statistics: While Hiatt and Strasburg(1951) discuss catch statistics for Arno Atoll, Marshall Islands, it was not possible to obtain similar data at Raroia. Fish- ing activity varies greatly between families and with time, and as found by Hiatt and Strasburg the interviews on such subjects were not satisfactory. The Raroians live to a great extent on fish while away from the main vil- lage (Garumaoa), but the rate of fish consumption while in Garumaoa is much lower; at home they prefer canned foods. Fishing Methods and Localities: Bengt Danielsson has prepared a report on native fishing techniques but a brief summary of current methods is presented here. The greatest bulk by volume of fishes are taken with single and four pronged spears in the inter- islet channels and along the shore, both along the outer corailine ridge and lagoon shore. Scarids, labrids, carangids, lutianids and serranids comprise practically all the fishes taken with this means. The natives hook and line fish to a lesser extent, preferring to troll along the outer reef edge over the coralliferous outer bench or in Garue Pass. They usually make their own metal hooks and pearl shell lures. Thunnids, carangids and sharks make up most of the catch. Stationary hook and line fishing is primarily done in Garue Pass, inter-islet channels and to a lesser extent around lagoon coral heads. A short line with a hook is attached to any handy pole. Serranids, acanthurids, and balistids comprise the major part of the catch. No reels are used. The hooks and lines are purchased from the trading schooners. -176- j 1 | Stone traps are often erected arownd schools of fish in shallow water ' on the outer reef flat end in inter--islet channels; these are only temporary and are destroyed quickly by wave action. Only one permanent trap is main- tained at Raroia. This is a stone trap approximately 50 feet in diameter next to shore in the lagoon south of Garue Pass. The natives occasionally take various food fishes out of it, but usually do not bother to keep it re- paired. No stick traps are used. No nets of any kind are now used at Raroia. All of the seining methods are accomplished by coconut fronds. The only fishing at night is done on the outer reef flat at high tide. The natives slash at the fishes in the shellow water with machetes or other long knives. They use torches or kerosene lamps for light. They primarily take holocentrids and lethrinids (Monotaxis). Usually the boys and young © men do this very successful type of fishing. & Poisonous species: The natives say that the fishes at Raroia poisonous to eat are Meromero (Epinephelus bohar), Tero (Lutjanus marginatus), Tonu (Paracanthistius maculatus), Ono {Sphyraena snodgrassi), Maito (not identified) and Oiri (not identified). Notes on their poisonous qualities are presented under each species in the systematic account. There are numerous records of ichthyotoxism at Raroia, since the na- tives sometimes cannot resist eating the particularly fat fishes, even | though they know they are poisonous. There was one mild case among the natives while the team was at Raroia which Bengt Danielsson attended. Mr. Danielsson has notes on the cases of fish poisoning at Raroia and will pre- sent them in his reports. The natives stated that there were usually about six cases of ichthyotoxism a year, but mentioned no deaths. They also as- serted that the number of poisonous fishes varies from atoll to atoll and from season to season. They stated very definitely that certain fishes are poisonous in one locality of an atoll, but completely edibie if occurring in another region of the same island. They claimed that the poisonous proper-— ties were due to what the fishes eat, but were not sure what the food was’ that caused the damage. They said that some atolis in the southern Tuamo- tus had very few inhabitants, since practically all the fishes were poison-— ous, and the natives could eat only a few of the species. The venomous fishes are Pugapuga (Synanceja), Tataraihau (Pterois, Scorpaenoges sp. 1), Pugapuga veve (Scorpaenopsis gibbosus), Marava (Siganus), some Acanthurus, all Ctenochaetus, Tapereta (Aetobatus) and the large moray eels (according to the natives). The natives especially fear Pugapuga. Tuna Baitfish Resources: Hiatt and Strasburg (1951) heve pointed out that baitfish resources may become the most significant marine income to natives on atolls, although this does not appear likely for French Oceania. The only baitfish at Raroia -177- during June-August was Stolephorus celicatulus. Lerge schools of young con- gregated around lagoon coral heads. According to the natives, various Clupe- oids come into the lagoon at definite seasons and form huge schools along shore. At such times the baitfish resources would be ample for a tuna fish- ery. On the off seasons, however, it would be difficult to get even a few scoops of bait. French officials at Papeete, Tahiti, did not believe that there is an adequate source of baitfish anywhere in French Oceania, Shark Menace: The natives are afraid of the sharks at Raroia and always leave the water whenever sharks come into the region in which they are swimming. ‘Sharks were particularly ebundant and aggressive in Garue Pass and around the outside of the atoll over the coraliiferous outer bench and in the surge channels. I was charged twice by sharks while in surge channels, but there were no attacks while we were at Raroia. One woman in the village has an arm missing that was bitten off by a shark when she was a girl. -178- RAROIAN VOCABULARY FOR ANATOMICAL FEATURES OF FISH While the information is not complete, some data was compiled for me by Miss Aurore Natua of the Raroiian general terms for the parts of a fish (see fig. 2). The natives consider a fish in three basic sections corres-— ponding roughly to head, (Homua), trunk (Koronaega) and tail (Mahiga); the top half of Koronaega and Mahiga is termed Kotika and the bottom half Kopaki and Takere for the respective sections. All of these terms refer to the way a fish is cut up for food. The remaining terms obtained refer to definite anatomical features. GOREGORE I ROTO.--Skin GOREGORE I VAHO.--Scales GUTU.--Tip of snout and lower jaw HANAA.--Fin; finlet HAVANATRARO.--Pelvic fin HAVENE.—-Fat, plump 3 HOMUA.~-Head section including paired fins KAEHORA.~-Gall bladder KAMIKAMS .--Gill filaments KEIGA.--Spiny rays lin particular, but in general refers to all fin rays\ KEIGA HAVANA.--Dorsal fin | KEIGA KOPAKI.--Pelvic fin rays KEIGA MAKUI.--Dorsal fin rays KEIGA TAKERE.--Caudal fin rays KOEA.--Flesh KOHIRI.--Fat about the digestive tract KOPAKT.-—-Stomach or belly region KOPANI.--Breast region between gill openings and pelvic fins KOROVAEGA.--Trunk region between paired fins and anus Lag KOTIKA.-~Back or dorsal region MAHIGA.—-Tail region of fish behind anus; caudal fin MOKA.—-Milt NIHO.--Tooth " 'NOHI.—Eye iy GPUPUy<-cuim Bladder, poaey SAFIN ness PRET Pt gee tne ne en een “haath Led PEPENU.--Side of forehead behind eye fextent of region not clear to author} PEPERERAU.--Pectoral fin pe in | PUTOHE.-——Anus RAE.--Occipital region ROEROE.--Stomach and intestine; belly region . TAGA RARO.-~Lower jaw TAGA RUGA.--Upper jaw TAKERE.—-Ventral region over anal fin; anal fin (particularly referring to the spines; 5 TAPAU.—--Nape eee ; - TEKETEKE.--Roe © TUA.--Back or dorsal region of body TUKETUKE.--Top of head UPEREPERE.--To descaie UPOUPO.--Heart VAEGA HUA MAU.—-Mid-lateral line of body VAHA.-—-Mouth ny 7180- NOTES ON SPEAR FISHING EQUIPMENT _ While matching wits with the fishes in the field trying to obtain ade- quate collections the ability of the collectors and the type of equipment used are all important. Fish collecting in shallow water has advanced tre- mendously within the past few years, and an amazing arsenel of skin diving equipment is now readily available in sport gcods stores. It is a lament— able fact that very few ichthyologists are really aware of this advance and even fewer utilize the gear that wovld greatly increase the ease and ability of collecting coral reef fishes. The best of this equipment is discussed below. Swimming gear: Around a coral atoll practically all of the worthwhile collecting has to be done by swimming. A swimmer will find Japanese Tabe coral resistant shoes or the Italian Salvani rubber soled swim fins combined with Japanese Tabe socks the most convenient footwear. It is generally recog- nized that a swimmer should carry a knife strapped to the trunks; the french plastic handled stainless steel dagger that floats is considered the best now on the market. Duli colored (so sharks will not be attracted) nylon trunks are sturdy swimming attire. In rough coral areas deep sea diving underwear, or even regular coveralls, save a lot of coral cuts and canvas-leather gloves offer good protection for the hands. Squale or soft rubber Champion face masks and Squale goggles are by far the best available in this line; the Italians have a better soft rubber face mask but few are obtainable in this country. Constant use day after day will wear the Squale mask out in two months time. Abercrombie and Fitch offer an ingenious crystal Lucite frame made in France for prescription lens that fits in the Squale, Champion and Italian masks. Nose clips are recommended as a necessity for those who have trouble clearing their eustachian tubes. Snorkels are often a great help while swimming along the surface. The Abercrombie and Fitch snorkel seems to be the best, and is made to fit most masks. Seasoned skin divers prefer the lightest and simplest snorkels. ‘The best rubber suit is the Pirelli, but some new foam rubber suits are appearing that are warmer. Diving gear: Aqua-lung one or two cylinder compressed air rigs are uni- versally accepted as the best for spear fishing. Ichthyologists usually prefer the two cylinder rigs since they last a longer time underwater (one hour) but sport spear fishermen prefer the me tank rig (half hour) since it is much lighter and easier to handle. Fenjohn Co. makes a canvas vest for the one tank rig that is a big improvement over the original straps, and also sell a small gasoline powered compressor for charging cylinders. Spear fishing gear: Ichthyologists have not utilized the tremendous advances that have been made with this type of equipment, particularly in regard to compressed air guns. The best commercially available hand spear is a sea net A 77 with an H 23 needle point head; this is a 103 foot alumi- num spear that unscrews into sections; it has a rubber loop at the end of the handle. There are many good rubber sling guns: The best of the Hawaiian type are the sea net pistol grip (but should be modified), and the Abercrombie and Fitch plastic models. Hawaiian slings can be easily made in the field ~181- providing rubber tubing is available. The best cocking rubber sling gun is probably the Italian Cressi, since it is especially made for close work around coral. Also very good is the French single sling Arbelete (this gun is sold under a variety of names, but most are the same); a front grip can be added for greater accuracy. The best steel spring gun is the Italian Cressi, but the noise of the spring alerts the fish. There are other elaborate Italian and French spring guns, but I find them too complicated and too hard to cock. It is questionable as to what air guns on the market are the best, since new ones are coming out every month. Apparently tne Fisher CO2 GN - 6 (uses li- quid COo; fires 10 shots per load; for large fish only) and the light Barra- cuda (uses CO> capsules; single shot) are the most popvlar. Both are easy to load and handle, especially the latter. I have just tried the Neptune, which has recently come out in California. It is almost as good as the Parracuda and is lighter. Most of the compressed air guns on the market are dangerous and awkward to handle, particularly when loading, so that it is necessary to be cautious when obtaining this type of gun. They have an accurate range of 50-80 feet underwater and open up a relatively untapped region of fish col- lecting. Many fishes stay just out of range of rotenone end hand spears that can easily be obtained with compressed air guns. Safety gear: As previously mentioned under swimming gear, it is advis— able to wear a dagger while swimming. Inexpensive depth gauges worn on the wrist are an aid for measuring depths and for decompressicn. Shark repel- lent should be used in tropical regions. Despite a few accounts to the con- trary, it has repeatedly been proven that the copper acetate repellent is very effective against sharks. Under some conditions a Res-Q-Pak is worth- while having. This is a small plastic package that is blown up by a COa cartridge when squeezed. A good collecting and safety device is an inner- tube float covered with canvas to form a receptacle for fish. A line is fastened around the outside and an anchor line is attached to it with a small crow bar on the end. It is important that speared fish be taken out of water as fast as possible so that sharks will not be attracted; the canvas floats are perfect for this. Aliso they can hold the dipnets, swim fins, bottles, compressed air cylinders for Aqua-lung, CO2 master tanks for compressed air guns, and can be used as a place of refuge in case of shark attack. Since a swimmer cannot normally handle a speared fish larger than himself the line | can be attached to the float and the fish played from the float. Nets: In this category dipnets are the most important. The best I have found is the 20" diameter Ocean City, N. J., aluminum dipnet, with screws for taking off the hoop, and removable handle. These are fitted with nylon © netting bags, which stand up very well in the tropics. Small aquarium nylon dipnets, common seines, gill nets, trammel nets and throw nets are all worth- while for coral atoll work. Ederer Net Co., Philadelphia, is most cooperative to expeditions and quite reasonably priced. Probably a dome-shaped fine mesh net approximately 25' in diameter’ would be useful in poisoning coral heads.’ The dome should be buoyed and the entire net could be placed over the region to be poisoned, and few of the fish would escape. -182- Ichthyocide: Powdered rotenone is the most commonly used. It must be freshly ground as fine as possible and preferably packaged in 5 or 10 1b. air tight packages, or better yet -- canned with a home ¢an sealer. Highly con- centrated liduid rotenone is far more effective in tropical regions but higher priced. The importance of the quaiity and effectiveness of underwater equipment cannot be overemphasized and can make a tremendous difference in the success of field work. Since skin diving is progressing so rapidly, atoll team mem- bers should check with Florida or southern California sport fishing stores for the latest advancements. -183- CONCLUSIONS Raroia Atoll supports a very large marine fauna both in the lagoon and on the outer reefs, that is not appreciably depleted by the native popula- tion. The relatively large size of the atoll allows for a great range of physical and ecological variation and the time and resources of the team were taxed to the limit trying to study even the representative situations. It would seem from the literature, however, that such atolls in the Tuamotus as Makemo, Tahanea, Anaa, Takaroa and Hao must have richer marine faunas, although the fish population of some of these islands may be affected by the fishing intensity of larger native populations than are on Raroia. The outstanding features of Raroia are (1) the presence of a moderately broad coralliferous outer bench around the atoll that supports the heaviest coral growth and largest fish fauna of all the major ecological zones; (2) the development of surge channels completely around the atoli that are ex- tremely porous and cavernous on the southwest side of the atoll; this eco- logical zone contains a specialized fish fauna that for a great part was found nowhere else; (3) a broad outer reef flat exposed at low tide that is a feeding ground for many groups of larger fishes at high tide; (4) the single deep pass into the lagoon on the west side of the atoll that sup- ports an amazingiyv iarge fisn fauna and allows for the free passage of larger fish in and out of the lagoon; (5) the numerous inter-islet channels around the atoll that sustain a moderate fish fauna of smaller species; (6) the deep lagoon with numerous flat-topped coral heads extending to the surface; while there appears to be a relatively small fish population in the open water of the lagoon there is a large fish fauna around and on top of the coral heads. In summary it should be noted that the coralliferous outer bench probably supports almost half the fish fauna at Raroia, and this zone plus the surge channels, deep pass, and lagoon coral heads support about three-fourths of the fish population at Raroia. Currents are an important factor in the distribution of fishes on the atoll. There is a lag between tides inside and outside the lagoon, causing a constant flow of water across the lagoon towards the west and south. At extreme high tide water flows into the lagoon through almost all the channels. At low tide by far the greatest volume of water flows out through the deep pass and over the broad southern flat. In the lagoon the coral heads near the deep pass and in the southwest region near Oneroa maintain the largest fish populations, apparently receiving the greatest flow of nutrient mater- ials. The character of coral heads and other coral growth in the calm pro- tected waters next to the east and north islets are strikingly different than elsewhere around the atoll. It is probable that the lagoon coral heads could be divided into several ecological umits depending upon their position in the lagoon and resulting modification of habitat. Also the volume of water flow in the inter-islet channels has a profound effect on the life of this re- gion, there being a noticeable change between the various types of channels and between channels of different parts of the atoll. One of the most interesting aspects of the Coral Atoll Project was the comparison of the ecology of Raroia and Takume Atolls, situated approximately -~184- fifteen miles apart. Raroia has a deep lagoon, major passes to the outside, and supports a large coral reef fauna in its lagoon. Takume has a shallow enclosed lagoon with minor channels to the outside and supports a small coral reef fauna in its lagoon remarkably different in balance than that of Raroia. While these two atolls may have the same species, their relative abundance and ecological associations are clearly very different. Fishes that were seen only once or twice during the entire period at Raroia were noted to be common at Takume and vice versa. These facts do not refer only to the lagoon, since even certain physical aspects of the outer reefs were strikingly dif- ferent and must have had a corresponding modification of ecological condi- tions. In many ways a comparative ecological study of atolls in one group would be more revealing than comparing islands from widely separated regions. It is difficult to estimate at this time the relative success of the Raroian ichthyologicali study. Somewhat less than twice as many species were collected at Raroia Atoll than ere known in the literature for the entire Tuamotu Archipelago. But it seems safe to state that at least 100 more spe- cies would have been collected, as well as much more useful data obtained, if the bulk of the ichthyological equipment had been available. Probably the main contributions that can be obtained from the collections and field data are the (1) determination of the relative abundance of species and .z00- geographical relationships of the fishes of this area, (2) establishment of identifications for approximately 150 Tuamotuan native fish names, (3) eluci- dation of some economic aspects, such as identification of poisonous fishes and food fishes and their relative importance, (4) establishment of a basis for a systematic study of the fishes of the Tuamotu Archipelago, (5) contri- bution of data on the habits, habitats, distribution, and other ecological aspects of coral atoll fishes. No small factor is how successfully the data is utilized for final publication. The author is greatly indebted to Mr. George Vanderbilt for generously sponsoring both the field work and subse- quent research. It is planned to complete investigations on the Raroia ma- terial along with the Vanderbilt 1951 Expedition collections family by family, culminating in a general systematic and ecological monograph. -185- AGASSIZ, 1903 BANNER, 1952 BRIGHAM, 1900 _ REFERENCES | ath Alexander Reportar on the scientabie esau ts of the vemeeahids to the Tropical - Pacific, in charge of Alexandér Agassiz, by the U. S. Fish Commis- sion steamer "Albatross" from August, 1899, to March, 1900, Commander Jefferson F. Moser, U.S.N., commanding. IV. The Coral Reefs of the Tropical Pacific. Mem. Mus. Comp. Zool. Harvard Coll., vol. 28, haribo gipiiahingt: PP+s ai plates. A. H. Preliminary Report on Marine Biology Study of Since atoll, Gilbert ae Islands. Part. I. Atoll a Bulletin no. 13, iit4e PP+s 6 tabls., 1 map. William T. An index to the Islands of the Pacifie Océan. A Handbook to the chart on the Walls of the Bernice Pavahi Bishop Museum of Polynesian Ethnology and Natural History. Mem. Bernice Ps ae Mus. Vol. Ube Art. 2, pp. 87-256, ner CLOUD, Preston E. Jr. 1952 FOWLER, 1928 1931 1934 1938 1940 Brelanadeny, Report on the Geology and Marine Environments of Onotoa Atoll, Gilbert Islands. Scientific Investigations in Micronesia. — Pacific Science Board, National Research Council. Atoll Research - Buldssi nos! 12, py: i- 73, 8 oriclont: 5 tables. ; . Henry W. The fishes of Oceania. Mem. Bernice P. Bishop Mus., vol. 10, i-iii+ 540 pp., 82 text-figs., 49 plates. The fishes of Oceania. Supplement 1. Mem. Bernice P. Bishop Mus., vol. 11, no. 5, pp. 313-381, 7 text-figs. The fishes of Oceania. Supplement 2. Mem. Bernice P. Bishop Mus., vol. il, no. 6, pp. 385-406, 4 text-figs. The Fishes of the George Vanderbilt South Pacific Expedition, 1937. Forming Part III of the Zoological Results of the George Vanderbilt South Pacific Expedition. Mon. Acad. Nat. Sci. Philadelphia. No. 2, 11i1-v+349 pp., 4 maps, plates v-xii. The fishes obtained by the Wilkes Expedition. 1838-1842. Proc. Amer. Philosophical Soc. Vol. 82, no. 5, pp. 733-800, 76 figs. _~186- FOWLER, Henry W. (cont'd.) 1949 The fishes of Oceania. Supplement 3. Mem. Bernice P. Bishop Mus., ‘ vol e ae 9 no e 3 5 Pp ° 37-186 s ‘ 1952 The longnose black surgeon fish (Laephichthys rostratus). The Fish Cultusist, Vol. 325 No. As Pp. 25-27, 1,fige - FOWLER, Henry W., and Barton A. BEAN 1923 Descriptions of eighteen new species of fishes from the Wilkes Ex- ploring Exnedition preserved in the U. S. National Museum. Proc. U. S. Natl. Mus., vol. 63, Article 19, no. 2488, 27 pp. FREEMAN, Otis W. (Edited by). 1951 Geography of the Pacific. v-xii+573 pp., 156 figs. John Wiley & Sons, New York. GILTAY, Louis. 1939 Résultats scientifiques des croisiéres du Navire-Ecole Belge "Mercetor." Poissons. Mem. Mus. Royal d'Histoire Naturelle de Belgique, Sér. 2, fasc., 15, vol. 2, no. 2, pp. 29-45, 1 fig. GUNTHER, Albert 1873- Andrew Garrett's Fische der Sudsee. Vol. I--pt.l, 2, 3 & 4} vol. II-- 1910 pt. 5, 6, & 7; vol. III--pt. 8 & 9. Jour. Mus. Godeffroy, vol. 2, pt. 2,°5, 7, & 9, ivtiitl28 pp., 10 figs., 83 pls., 1873-1875; vol. 4, pt. 11, 13, & 15, pp. i-v4l29-260, 3 figs., pls. 84-140, 1876-1881; vol. 6, pt. 16 &17, pp. i-vit261-515, 17 figs., pls. 141-180, 1909-1910. 1887 Report on the Deep-Sea Fishes collected by H. M. S. Challenger dur- ing the years 1873-1876. Zoology, vol. 22, pt. 57, i+lxv+335 pp., pls. 1-73. Eyre & Spottiswoode, London. HALE, Horatio. 1846 United States Exploring Expeditions during the years 1638, 1839, 1840, 1841, 1842 under the command of Charles Wilkes, U.S.N. Ethnography and Philology. i-xii+666 pp., 3 maps. Lea and Blanchard, Philadel- phia,. | ; HASKELL, Daniel C. 1942 The United States Exploring Expedition, 1838-1842 and its publications. 1844-1874. xi-xiit+t18$ pp., Frontispiece, 4 plates. The New York Public Library. -187- HERRE, Albert Wi. 1935 New Fishes obtained by the Crane Pacific Expedition. Field Mus. Nat. Hist., Publ. 335. Zool. ser., iis el no. 12, pp. 383-438, figs. 31-33. 1936 Fishes af au Crane Pacific siehthsaiRcn Field Mus. Nat. Hist. Publ. 353, Zool. ser. vol. 21, 472 pp. figs. 1-50. HIATT, Robert, We, and Donald STRASBURG 1951 Marine Zoology Study of the Arno Atoll, Marshall Islands. Scientific Investigations in Micronesia. Pacific Science Board, National Re- search Council. Atoll Research Bulletin no. 4, i+13 pp. HYDROGRAPHIC Apes Ae 5.N. Dept. 1944, Islands of the Ban and South Pacific. (New Caledonia to Sala y Gomez, latitude 15°N. to latitude 60°S.). H. 0. Pub. No. 887 (Reprint Nov. 1944 of H. 0. Mise. no. 10,887). Gazetteer (no. 7), 114222 zt. KENDALL, William cas and Edmund L. GOLDSBOROUGH 1911 Reports on the scientific results of the aiid to the tropical . Pacific, in charge of Alexander Agassiz, by the U. S. Fish Commis-— sion Steamer "Albatross" from August, 1899, to March 1900, Commander Jefferson F. Moser, 0.S.N. commanding. XIII. The shore fishes. Mem. Mus. Comp. Jape Harverd Coll., vol. 25, no. 7; pp. 241-344, 7 gis. KENDALL, William C., end Lewis RADCLIFFE 1912 Reports on the scientific results of the expedition to the eastern tropical Pacific, in charge of Alexander Agassiz, by the U. S. Fish Commission steamer "Albatross," from October 1904, to March 1905, Lieut. Commander L. M. Garrett, U.S.N., commanding. XXV. The shore fishes. Mem. Mus. Comp. Zool. Harvard Coll. Vol. 35, no. 3, Pp. Ve «i Bag g pls. KNER, Rudolf. 1877 1879 Pisces. pp. 11-18. Im: Schmeltz, J. D. E. Jr. Museum Godeffroy Catalog VI, viitl07 pp. L. Friedrichsen & Co., Hamburg. Pisces. pp. 36-64. In: Schmeltz, J. D. E. Museum, Godeffroy Catalog VII. Wirbelthiere (Animalia vertebrata) und Nachtrage zu Catalog v & vi aus den Ubrigen Thierklassen. bahia ag pp. L. Friedrichsen & ' Co., Hamburg. : -188- MURPHY, Robert Cushman, 1922 The Whitney South Sea Expedition of the American Museum-of Natural History. Science, NS vol. 56, no. 1460, pp. 701-704. NICHOLS, J. T. eee Two new fishes from the Pacific Ocean. American Mus. Novitates, no. 94, S pp. 2 rigs. NICHOLS, J. T., and ©, M. BREDER, Jr. 1935 New Pacific Flying-Fishes collected by Templeton Crocker. American Mus. Novitates, no. 821, 4 pp., 3 figs. PICQUENOT, F. V. 1900 Géographie, Physique et Politique des Etablissements Francais de l'Océanie. Cours Supérieur (Troisiéme Année). 119 pp., 6 Maps. Augustin Challamel, Paris. RANDALL, John E. 1952 Preliminary report on marine biology study of Onotoa Atoll, Gilbert Islands. Part II. Investigations of the Ichthyofauna of Onotoa, Gilbert Islands. Atoll Research Bulletin no. 13, pp. 43-62. REGAN, C. Tate. 1917 A revision of the Clupeid fishes of the genera Sardinella, Harengula, & etc. Ann. & Mag. Nat. Hist., ser. 8, vol. 19, pp. 377-295. SCHULTZ, Leonard P. 1943 Fishes of the Phoenix and Samoan Islands collected in 1939 during the expedition of the U.S.S. "Bushnell." Bull. U. S. Nat. Mus., no. i80, iii-x4316 pp., 27 text-figs., 9 plates. SEALE, Alvin. 1906 Fishes of the South Pacific. Occ. Papers Bernice P. Bishop Mus. Vol. 4, no. 1, pp. 1-89, figs. 1-23. TINKER, Spencer Wilkie. 1944 Hawaiian fishes. A handbook of the fishes found among the islands of the Central Pacific Gcean. 404 pp., 506 figs., 8 plates. Tongg Publ. Co., Honolulu, Hawaii. -189- VAILLANT, Léon. 1887 Matériaux pour servir & l'histoire ichthyologique des Rrchipele de la Société et des Pomotouvs. Bull. Soc. iedetaiachanaeal de Paris. Sér. 7, vol. 11, pp. 49-62. WEBER, Max, and L. F. de BEAUFORT. 1922 The fishes of the Indo-Australian Archipelago. Vol. IV. Heteromi, Solenichthyes, Synentognathi, Percesoces, Labyrinthici, Microcyprini. t x1iii+410 pp., 103 figs., E. J. Brill, Leiden. i 1931 The fishes of the Indo-Australian anaes VI. Perciformes (continued) families: Serranidae, Theraponidae, Sillaginidae, Emmelichthyidae, Bathyclupeidae, Coryphaenidae, Carangidae, Bachy- centridae, Pomatomidae, Lactariidae, Menidae, Leiognathidae, Mullidae. i-xiit448 pp., 81 figs., E. J. Brill, Leiden. WILKES, Cherles. 1845 Narrative of the United States Exploring Expedition during the years 1838, 1839, 1840, 1841, 1842. Vol. 1, i-xxiit433 pp., Illus. Leaé& Blanchard, Philadelphia. -190- FIELD STATIONS OF THE ICHTHYOLOGICAL. SURVEY FIGURE | a Photogrammetric base, N. D. Newell @ Sunol VOIHVW VOAVAOYOY YANWOH [HOLA ds | —— OYVYIVNVAVH sy ,dJOYNOSY a 7 oe —-VHIOMOM > VHVA 7 I4VdO- YNH V93VA C _- OHIN | VOIHVW —— Nin = sy—- ON ~S —S S ~ Eset VNVAVH=—=-—- Lyy PP | x . F; yy ‘wantin “~ nNadad ANNU p) SS SN 9 SSS a ace } a “Mel edsid HSI3 WV SO SLYVd SHL YO4 S3WYN NVIOUVY € JHYNdI4 SQV4H 1VH¥O9 HLIM NOOSVT N3adoO TIOLY WIONYWHY 40 $371404¥d OANVISI NOOSV1 NO SAHSIS 4O NOILVNOZ 1V9190710943 J0il MO $100d JdIL HONS YALNO STANNVHO 49d1u SNOY¥3AdIT1VYOO 49YNS ANIT1IVeO9 JDM el elelets| = 2S2hfao WOLV Viddve 40 Saqis0ud GNVaqS| Soegd 2 0)SEnO NG; SaHSla ~430 NOMVNOZ = aAVOISO1004 _— > ait Gat 7 : i % i fi pe I i ‘ : Kl . Cate. | aye re j spall Ne Mh 8 " ( 7 . > AGANTHURUS TRIOSTEGUS LOW TIDE CTENOGHAETUS POMACENTRUS NIGRICANS LAGOON INCOMPLETE INTER-ISLET CHANNEL 2 OUTER REEF FLAT \ LABROIDES zs ‘SAND BAR CHAETODON ae INCOMPLETE ZANCLUS % 4% INTER-ISLET tla CHANNEL € ae ig SERRANIDS , aw HENIOCKUS OPEN. LAGOON CORALLINE BUTTRESS ns : SHORE TIDE ACANTHURUS ain POOLS ZEBR ASOMA qj yy Ons a sas COMPLETE INTER- Now AULOSTOMUS Ong, r ISLET CHANNEL EPIBULUS maT ya \ y PSEUDOCHEILINUS tie pea CORALLINE RIDGE Fino a fT DASCYLLUS << ‘ yon MORINGUA y ANN ZAM ygpy I SURGE CHANNEL SYNODUS EGOLOGIGAL ZONATION OF FISHES ON SHORE REEF FACE WEST SIDE RAROIA FIGURE 4 ECOLOGICAL ZONATION OF FISHES ATOLL AROUND ISLETS OF RAROIA ATOLL FIGURE 5 CORALLINE RIDGE SURF ZONE SURGE ZONE ENGHELYNASSA eon iy Tei lifeoanayees GYMNOTHORAX i CIRRHITUS PINNULATUS,” — a ~ 7 id ny t 4 ‘ 4 SCORPAENIDS UROPTERYGIUS \ “ee oa EE: BOSS Sie ae REGION 2 Nein REGION 3 aa DINEMATIGHTYS i Ne ABUDEFDUF PHOENIXENSIS SCARUS 1 i ‘ ~ SHARKS 1 Biase i Sig ACANTHURUS /GUTTATUS SCARIDS * (eee ape \ CARANGIDS Si BALARING i ACANTHURUS SHARKS \ THALASSOMA I 1 x POMAGENTRUS 20' FIGURE 6 DAYTIME ECOLOGICAL ZONATION OF FISHES IN SURGE GHANNELS RAROIA ATOLL ECOLOGICAL ZONATION OF FISHES ON SMALL LAGOON CORAL HEADS RAROIA ATOLL CROSS SECTION ACANTHURUS MYRIPRISTIS AGANTHURUS CTENOCHAETUS GOBIIDS , HALICHOERES CTENOCHAETUS ; CHINE TOES EPINEPHELUS, THALASSOMA CHAETODON LOW TIDE DASGYLLUS | DASGYLLUS CHROMIS CHROMIS ENCHELYNASSA GOBIODON EPINEPHELUS ACANTHURUS CTENOCHAETUS LABROIDES HENIOGHUS, GHAETODON ZEBRASOMA ,EPINEPHELUS THALASSOMA CHAETODON, AULOSTO MUS AGANTHURUS THALASSOMA, SCARUS SCARUS ACANTHURUS HOLOCENTRUS, APOGON DINEMATICHTHYS NASO EPINEPHELUS, GOBIODON CARANGIDS CEPHALOPHOLIS, YOUNG GHEILINUS IN HOLES GEPHALOPHOLIS EPINEPHELUS CHROMIS DASCYLLUS DASCYLLUS GHROMIS SCARUS SYNODUS FIGURE 7 “e ; a4 A al a "LALA 9088 01375 3561