Because of limited time available for the field survey, only two days were spent at Kosrae and six days at Pohnpei. Prospective areas, mostly those located outside of fringing or barrier reefs, were selected based on recommendations made by Marine Resources Divisions of Kosrae and Pohnpei. Windward reefs were not accessible by small boats and therefore the survey areas were limited more or less to the lee sides of both islands.
The reef physiography was considered to be most important in evaluating habitats of both juvenile and adult green snail. Inclination of reef slopes, development of spur and groove systems, and existence of terraces were the points of primary concern. Also considered were algal covers and micro-refuges formed on the limestone substrates.
The senior author (MY) snorkelled and examined shallower segments of the reef slopes up to the top of the reef flats whenever possible, and the junior author (KK) explored using SCUBA along the deeper parts of the slopes down to ca. 20 meters depth. Both took underwater photographs to record major topographic and biological features of the survey areas.
Larger gastropod snails such as Trochus niloticus, Tectus pyramis, and Turbo argyrostomus found along the survey areas were examined and sampled for further study on their feeding habits. These snails might compete with green snail for food within the same type of habitat, occupying similar ecological niches. Stomach contents of specimens of the congener Turbo argyrostomus and some of Trochus niloticus were examined after being extracted by boiling and removing the animal body from the shells.
In Pohnpei, one lot of such specimens was collected at night to see more fresher stomach contents, and also to observe feeding activities of the snails in the field.
Collected shell specimens were measured and examined to record abundances of hipponiciid, parasitic gastropods to turbiniids which might constitute a potential nuisance to green snail.
References were made to useful publications such as Pohnpei Coastal Resource Inventory, Pohnpei Coastal Resource Atlas, and Kosrae Coastal Resource Atlas. Report on surveys of many reefs including those of Pohnpei with regards to previous trochus transplantation were very informative and useful (Asano, 1963). For Kosrae, the final Environmental Impact Statement (Kosrae Airport, published in 1977), was also a useful reference.
Fig. 2 Map of Kosrae, indicating five survey points (X) and the two stations of transect lines (arrows) for Fig. 3
LEGEND
Fig. 3 Cross-sections of fringing reefs along transect lines at two stations indicated in Fig. 2. (modified after Federal Aviation Administration final EIS, 1977)
The Island of Kosrae has ca. 50 Km of coastline, surrounded by a series of fringing reefs which are cut by river mouths (Fig. 2). The reef flats are much wider along the northern coast. The majority of shorelines are fringed with mangrove or swamp forests.
We examined five locations indicated on the map. They are confined to the northern half of the island, because of prevalent southely and eastely winds during our survey period. It was impossible to examine the reef crests at the two eastern locations of Foko Finaunpes and Foko Puku, because of heavy surf.
Reefs at Foko Finaunpes and Foko Puku were covered with rich coral growths over rather gentle slopes, but those of three other sites were less so with more crustose coralline-algae substrates over dead corals. It is likely that predation by the crown-of-thorns starfish was responsible for such difference. The starfish was abundant a long the reef off Okat and common at the reefs off both Walung and the trochus sanctuary, while it was rare along those at Foko Finaunpes and Foko Puku.
Figure 3 illustrated general features of reef topography at two contrasting sites, representing northern and southern reefs. There are gentle reef slopes down to 12 to 20 meters and then the slopes drop sharply to the abyssal depths at both sites. However, the shelf-like slope is wider at the Okat reef than at Malem reef, probably reflecting difference in width of reef flats.
All the ocean-facing reef slopes of Kosrae may be considered as suitable for accommodating green snail populations. However, the reef designated by Kosrae State as a trochus sanctuary appeared to be the best area for transplanting the snail because of superior topographic complexity and wider reef shelf area. The southern half of the coastline from Walung to Lelu may be less favorable because of narrower reef zones for growing the snail. The sanctuary is most favorable also because the established management system for trochus will be applied for green snail as well.
The Island of Pohnpei is surrounded by a lagoon and series of barrier reefs which is narrower along the southeast coast, merging with fringing reefs. There are many islets with fringing reefs and also extensive patch reef systems inside the lagoon. We examined five locations along the western barrier reefs from offshore and made a few spot checks inside the lagoon (Fig. 4). The eastern and southern reefs were not accessible by small boats because of heavy surf generated by prevailing eastely winds.
Fig. 4 Map of Pohnpei, indicating survey points (X)
The seaward reef shelves of the western barrier reefs were in general fairly narrow (around 100 m) with steep drop-off slopes to abyssal depths commencing at about 10 fathoms. The reef flats behind the crests were scattered with large boulders and coral rubble which were exposed during low tides.
Live coral cover was not extensive over the reef shelves at a majority of survey stations. Limestone substrates covered with crustose coralline algae were prevalent. Development of spur and groove systems were apparent but surface topography was rather simple along the western barrier reefs, except for one station near Peina Islet (Kiparitik) where an excellent green snail habitat was found. This area is one of the seven trochus sanctuaries of Pohnpei (see Pohnpei Coastal Resource Atlas) and therefore would be the first choice for green snail transplantation.
In addition to the above, the westernmost barrier reef off Palikir Point may be a good site. This is because extensive patches of algae Laurencia cartilaginea were found on the outer reef flat zone. This species was also abundant at inner reef flat of Peina Islet station, together with another alga Hypnea nidulans. These algae are good food for juvenile green snail. Such extensive algal growths were not reported by the survey of reefs along the northern barrier reefs and lagoon reefs (Birkeland, 1980) and were not found by us from reefs other than the two western barrier reefs. Perhaps, these good algal growths are related to the trade wind which blows over the reefs and drives the flow of nutrient-rich fresh water from the main island.
A few lagoon patch reefs examined did not indicate any good habiats for green snail because silty and unconsolidated surfaces were common. Although limited to the northern lagoon, a survey report of patch reefs indicated similar topography and habitat type (Birkeland, 1980). Thus, lagoon reefs, including the fringing reefs of Pohnpei Island itself, may be unsuitable for green snail transplantation.
We snorkelled along the north-western side of Nankapenparam Reef at night (11 July), in order to observe and collect specimens of the two gastropod species which might become competitors against transplanted green snail. This northernmost barrier reef was one of the seven trochus sanctuaries. There, large trochus were abundant all over the reef shelf. We arrived at the survey site at about 20:30 hours and spent about two hours.
Almost all individuals of both Trochus niloticus and Turbo argyrostomus were actively feeding on the limestone substrates, probably grazing microscopic algae together with surface-covering silt. The stomach contents of these gastropods were later examined and they verified this observation. These gastropods were inactive and most individuals are found hidden inside crevices, under limestone ledges or coral colonies, during the daylight hours. One very large reef crab, Carpilius maculatus, was holding an adult Turbo argyrostomus with its two large chelae, although the shell was intact with its operculum closed. Two additional crabs of the same species were found not carrying any victims.
Large hermit crabs such as Dardanus spp. were fairly common at the majority of survey sites of both Kosrae and Pohnpei, occupying trochus shells. We believe that those large hermit crabs would constitute significant predators of green snail, trochus and other gastropods.
In Kosrae, we found two trochus which had been killed recently. Their opercula were removed but there were no major damages to the shells and parts of their flesh still remained inside, rotten. They were likely to be victims of hermit crab predation. Nash (1985) discussed details on hermit crab predation of trochus in the Great Barrier Reef.
Hermit crabs do not crush the shell to kill the gastropods. On the other hand, we found crushed remnants of both trochus and silver-mouth turban (T. argyrostomus) scattered on the bottom at most survey sites. It was not possible to identify the specific predators of the shells but larger crabs such as Carpilius maculatus were probably involved with these crushings.
There were many (about one dozen) dead trochus shells piled at a spot in about 5 meter depth of reef shelf off Walung, Kosrae. We can not figure out what caused this underwater graveyard of trochus. Some of the shells have been consolidated with the limestone substrate, while others were more freshly killed in their appearences.
Table 1 Abundance of parasites on Turbo argyrostomus
| Locality | N | Number of Sabia sp. on Turbo | Shell length of Turbo | ||||
| 0 | 1~2 | 3~8 | 9~18 | mean | (range) mm | ||
| Kosrae | 9 | 1 | 3 | 3 | 2 | 67.9 | (54.4 ~ 79.1) |
| Pohnpei | 2 7 | 5 | 7 | 9 | 6 | 62.2 | (55.3 ~ 72.3) |
| Okinawa | |||||||
Maeda | 23 | 2 3 | 0 | 0 | 0 | 66.6 | (58.4 ~ 78.5) |
Sunabe | 1 1 | 1 0 | 0 | 1 | 0 | 75.2 | (68.3 ~ 84.2) |
Toya | 2 7 | 2 3 | 4 | 0 | 0 | 73.0 | (67.2 ~ 80.5) |
Yomitan | 1 0 | 7 | 3 | 0 | 0 | 65.9 | (60.5 ~ 71.3) |
| Tokunoshima | 1 5 | 9 | 6 | 0 | 0 | 80.6 | (74.7 ~ 89.0) |
note: Number of Sabia sp. attached directly on shells of Turbo argyrostomus were counted by naked eye, excluding those attached on the shells of the conspecifics. Those larger than 2 mm shell length of Sabia sp. were detectable. The maximum length of Sabia sp. was 15.0 mm and 17.8 mm for shells of the FSM and the Ryukyus, respectively.
Table 2 Abundance of parasites on Turbo marmoratus in Tokunoshima
| shell width mm | N | Number of Sabia sp. on Turbo | |||
| 0 | 1~2 | 3~8 | 9~18 | ||
| 85~160 | 9 | 7 | 2 | 0 | 0 |
| 161~198 | 8 | 3 | 0 | 2 | 3 |
| Total | 17 | 10 | 2 | 2 | 3 |
note: Number of Sabia sp. attached directly on shells of Turbo marmoratus were counted by naked eye, excluding those attached on the shells of the conspecifics. Those larger than 2 mm shell length of Sabia sp. were detectable.
Internal parasites and commensals were not examined specifically, but one species of ectoparasite to T. argyrostomus was conspicuous. It is probably significant as a potential nuisance in the transplantation of green snail. The majority of adult turban were infested with a number of conical shells of a hipponicid Sabia sp. which were attached around the shell apertures of the host. This parasite is problematical because it digs into the shell of the host species and ruins its quality. Fortunately, trochus shells appeared to be free from this ectoparasite.
We compared numbers of parasitic Sabia attached to the shells of T. argyrostomus from Kosrae and Pohnpei against those on the same species from the Ryukyu Islands. Table 1 indicates clearly that the shells of this turban snail were more heavily infested with the parasites in FSM than those in the Ryukyus.
Forty-two T. argyrostomus in three samples and 12 Trochus niloticus in two samples were collected from Nankapenparam Reef. They were boiled in hot water and their soft bodies removed. Materials contained in their esophagus, stomach, and intestine were examined under dissecting microscope. The results are indicated in Table 3, with information on sex ratio, sizes etc.
Table 3 Stomach and gonad conditions of Turbo argyrostomus in Pohnpei.
| date (1988) | stomach conditions | gonad conditions | ||||
| all jelly | part jelly | no jelly | ripe female/male | intermediate female/male | spent female/male | |
| July 9 | 9 | 3 | 1 | 3/4 | 1/3 | 0/2 |
| 10 | 0 | 17 | 0 | 3/1 | 7/4 | 0/2 |
| 11 | 1 | 7 | 4 | 1/1 | 3/4 | 0/3 |
| total | 7/6 | 11/11 | 0/7 | |||
note: samples were collected in the afternoon, in the morning, and at night on 9th, 10th, and 11th, respectively.
The first lot of T. argyrostomus was examined for stomach contents only. Nine out of thirteen individuals from this sample revealed stomachs fully occupied by gellatinous material which resembled carrageenan or agar jelly of creamy white color. Remaining four stomachs were part-occupied by the jelly mixed with sand and silt as well as a few fragments of algae of the Family Squamariaceae which encrust limestone substrates.
Because the shells were collected in the daylight hours when they were not actively feeding, the jelly-filled stomachs did not indicate the nature of fresh food taken. Therefore, the second sample was collected in the morning (0800 to 0830 hours) and boiled as soon as being landed before noon.
The stomach contents of the second sample, 17 T. argyrostomus, all indicated uniform condition of half-jelly and half-silty sand grains. The gellatinous material was found enclosing sand grains and not consolidated in one block, which condition was common in the first sample. Both esophagus and intestine contained silty sand with occasional fragments of algae, mostly those of Squammariaceae which were carried through to the anus without being digested.
Then the third sample was collected by diving at night when they were actually feeding and were boiled at about midnight. 12 specimens of T. argyrostomus thus collected revealed similar condition to the second sample but four individuals little or no jelly (sand and silt only). Jelly was formed around sand grains and also along stomach walls. Both esophagus and intestine showed similar conditions to those of the second sample, containing silt and sand.
The above may indicate that this turban snail would consume microscopic algae (not identifiable by dissecting microscope) growing on the limestone substrates together with detritus, silt and sand plus occasional fleshy algae including undigestible Squammariaceae. Jelly was found in the stomachs of live animals so that it might be produced by digestive activities by the stomach itself rather than the artifact of boiling and cooling. It might be retained inside the stomach to be digested slowly while the undigestible matters passed through the digestive system.
Tsuda and Randall (1971) reported on stomach contents of Turbo argyrostomus and T. setosus. They found the stomachs were filled with detritus with small quantities of a number of algal species.
The stomach contents of 12 trochus examined also revealed gellatinous matter mixed with silty sand, similar to those of turbans. It is evident that the two species co-habit with each other and that they grazed limestone substrates in a similar fashion.
