著者

日下部 台次郎

出版者

広島大学水畜産学部

雑誌

広島大学水畜産学部紀要 (ISSN:04408756)

巻号頁・発行日

vol.2, no.2, 1959-12

---

(1) INTRODUCTION(i) During the last war, propagation of ark shell (Anadara subcrenata (LISCHKE)) culture was planned as a part of the national program for increasing food production. The plan consisted in exploiting those potential culture grounds for this clam which were found in various parts of the country by planting them with the seed clam collected by artificial methods. Since methods of collecting ark shell seed on a large scale had not been known by that time, national research fund was defrayed in 1942 for the development of such methods. And I was engaged in this project as the researcher in charge.(ii) This paper presents the results of the research carried out under this project, together with the results of the studies which I have done thereafter. I wish to extend my sincere gratitude to Mr. S. FUJIMORI, Dr. K. UCHIDA, Dr. M. TAUCHI, Dr. T. HANAOKA and Dr. Y. OSHIMA for their kind advice, and to Mr. F. MIZUNO, Mr. M. KAWAJIRI, Mr. K. TANAKA, Mr. H. TOKUNAGA and the late Mr. K. HATA who were engaged in the research with me.(2) ECOLOGICAL STUDIES(i) The ark shell, called "sarubo" or "mogai" in Japanese, is a bivalve inhabiting the bottom of calm bays and is widely distributed along the coasts of our country. Its annual production amounts to about 18,000 metric tons, and its principal production centers are Tokyo Bay, Osaka Bay, Kasaoka Bay, Ariake Bay and the Nakanoumi.(ii) This clam usually inhabits those shallow waters of a well-protected bay which are comparatively calm and whose bottom is composed of mud or sandy mud. It lives best in the depths of I to 7 m and buries itself shallow in the bottom, attaching itself to the sand or sandy mud by its byssus. The optimum range of specific gravity (σ15) of sea water is from about 18 to about 22.(iii) The breeding season is between early July and middle September, with the peak from middle July to the end of August. The spawning begins when the water temperature on the bottom rises to about 25°C, and reaches its peak at about 27°C. The spawning occurs with the rapid rise of water temperature one to three days before the spring tide, so that, we can forecast its occurrence.(iv) The development of this species was traced by artificial fertilization. Spawning was induced artificially. The egg is 50 to 60JJ> in diameter and the spermatozoon, 36 to 38JJ> in total length. At 26-27°C, the larval shell is completed and a larva becomes a D-form veliger in 16 to 17 hours after fertilization; one week after, the larva attains the umbo stage, when the shell becomes yellow; two weeks after fertilization, the larva becomes a full-grown veliger 280 to 300JJ> in length and 200 to 220JJ> in height, and is ready to settle. The shell has such a characteristic appearance at this stage, being very long, yellow in color, inflated in the umbonal region, and thick and robust, that we can easily distinguish a full-grown veliger of this species from those of other clams.(v) The full-grown veliger creeps over the surface of a substratum by extending and contracting its foot until it finds a suitable spot, where it attaches itself with the byssus which it secretes. Upon settlement the larva becomes the young. The shell that is secreted thereafter differs from the prodissoconch, being greyish-white and composed mainly of calcium carbonate like the shell of the adult. The radiating ribs characteristic of the shell of this species become distinct when the shell attains a length of about 4mm, and about 30 of them are recognizable on a shell reaching a length of about 1 mm.(vi) At the shell lengths of 10 to 15mm, i.e., 3 to 6 months after settlement, the young detaches itself from the substratum and creeps into the bottom of the sea, where it lives for the rest of the life. Even after it has crept into the bottom mud, it attaches. itself to such solid material as sand grains by means of its byssus.(vii) According to the measurements made on alternate days after attachment, the shell length reaches 0.8mm in ten days, 1.6mm in twenty days, 2.5mm in thirty days, 3mm in fourty days, 3.5mm in fifty days and 4mm in sixty days. The best season for the growth is May and June, when water temperature ranges from 18° to 23°C. The shell length reaches 23mm at the age of one year, 32mm at one and a half year, 37mm at two years, 42mm at two and a half years and 46mm at three years.(3) THE HYDROLOGICAL PROPERTIES OF THE NAKANOUMI(i) The Nakanoumi is a lagoon enclosed by the Shimane Peninsula and the Yumigahama Peninsula, measuring 8.5km in circumference and 152 km2 in area. Two islands, Daikon-jima and E-jima, are in the northern part of it. It communicates with the sea (Miho Bay) by a channel called Nakae-seto which is 200 to 400m in width, 7 km in length and 7m in depth. It is shallow with the average water depth of 4.6 m: it is not deeper than 2m in its one-fourth, and is 2 to 8 m deep in the other three-fourths. Three rivers, namely, the Ohashi, the Iu and the Iinashi, pour into it in all seasons. The exchange of water with the outside is seldom effected by the tide, but is influenced chiefly by the atmospheric pressure and the direction and velocity of the wind, because the range of tide is very small (less than 30cm) in the Japan Sea. In winter, when the water level of the Japan Sea falls because of high atmospheric pressure, the fresh water of Lake Shinji flows into the lagoon to lower the salinity of its surface layer. In summer, atmospheric pressure is low and sea level rises in the Japan Sea, from which the sea water of high salinity flows into the bottom of the Nakanoumi to form a sharp discontinuity of salinity at the depths of 3.6 to 4.8m. Since water is shallow, it sometimes occurs after a heavy storm that the upper and the lower layer of water intermingle to make the salinity uniform throughout the depths. The water temperature is much influenced by the air temperature: it is 2° to 5"C in the surface and 3° to 5°C on the bottom in winter (January and February), and is 28° to 30°C at all depths in summer (August). The specific gravity of the water (σ15 ) is 10 to 15 in the surface and 20 to 23 on the bottom both in summer and in winter. In June and July salinity rises in spite of much rainfall, owing to the influx of the sea water of high salinity. In a droughty year salinity is low in spite of little precipitation, because the high atmospheric pressure which prevails in such a year lowers the sea level and prompts the fresh water to flow down from Lake ShinjL(ii) The water of the Nakanoumi contains twice or thrice as much silicate and nitrate as that of Tokyo Bay. The concentration of silicate ranges from 5 to 10 mg/L, and that of nitrate from 600 to more than 900 mg/m3 • Soluble organic matter is present in the concentrations over 5 mg/L. Thus the Nakanoumi has the properties of an eutrophic lake. Consequently, planktonic diatoms are comparatively abundant; the species represented are those which are common in shallow seas, and belong to such genera as Chaetoceros, Ske/etonema, Coscinodiscus, Rhizosolenia, Thalassiothrix, Nitzschia, Navicula, P/eurosigma, etc. It is not seldom that red tide is caused by the heavy growth of dinoflagellates or other planktonic organisms. There has been tendency that fine drifted sand deposits in the strait of Nakaeseto. Though a long embankment has been built in order to protect the entrance of Sakai-minato (the harbor situated in the strait) from being blocked by depositing sand, deposition of sand proceeds at such a rate that the harbor entrance has becotne shallower year by year and the inflow of sea water into the Nakanoumi has been hindered increasingly. As a result, the trend for the Nakanoumi to become brackish has been accelerated in recent years, and the plankton community has been changing correspondingly.(iii) The bottom consists, in the main, of more or less blackish soft mud, rich in humus and mingled by dead shells. It is so soft that one can easily thrust a bamboo-pole to a depth of several feet. The mud is greyish blue underneath, and is overlain by a thin, blackish brown layer. Generally, it becomes increasingly blackish toward, the shore. Mud is more blackened in Yonago Bay than in other parts of the lagoon. In summer, the water of the bottom layer contains little, and in some places no, oxygen.(iv) The famous "red tide" is brought about when the plankton organisms which have been nourished by the nutrient-rich sea water multiply into great numbers owing to the high summer temperatures. Death of these organisms, coupled with the rapid decomposition of humus due to high temperatures, produces a water mass devoid of dissolved oxygen. When this water mass moves about, it kills benthic animals and occasions heavy damage not only to the ark shell on culture grounds but also to such aquatic resources as eel, sea bass (Lateolabrax), octopus, blue crab, etc. The red tide that I witnessed in the middle of August, 1942 was caused by the heavy growth of Ceratium tripos. Red tide usually originates in Yonago Bay. Since the aforementioned embankment was constructed, red tide has been occurring more frequently in the Nakanoumi than before, because the exchange of water with the open sea has been more limited.(4) SEASON OF OCCURRENCE OF ARK SHELL LARVAEIf one attempts to collect the ark shell spat by artificial method, he should first find out in what season the free-swimming larvae of this species occur, and thus become able to foresee the time suitable for spat collection. Observations were therefore conducted at Station 1 in the center of Ara-shima Bay (water depth 5.5m) in the summer of 1943. The observation covered the period from July 10 to September 30, 1943. During the period the plankton was collected every other day with both the plankton net and the hand pump. The plankton net was hauled vertically from the bottom to the surface three times and the catches were combined. In the collection with the hand pump, 30 L of water was drawn up from different depths of the water column at 60 em intervals and filtered through a plankton net to recover plankton. The plankton samples were fixed in 4% formalin. After taken back to the laboratory, each sample was stirred gently, the diatoms that floated up were discarded, the material settling on the bottom of the ve3sel was transferred onto a ruled counting plate, and the ark shell larvae of umbo and full-grown stages were counted. The results indicated that there were two major peaks of the occurrence of the ark shell larvae: the concentration of larvae reached a maximum during August 4-7, gradually decreased thereafter, and attained another maximum during September 5-7.(5) HORIZONTAL DISTRIBUTION OF ARK SHELL LARVAEFor the purpose of investigating the horizontal distribution of ark shell larvae, the Nakanoumi was divided into series of squares like a chess-board. The dividing lines were 1,090 m apart. Itohana was designated the base point, and the line passing this and Nyfiko Village on the island of Daikon-jima, the base line. At each crossing of the dividing lines a bamboo-pole was set up, from which a "test rope" (i.e., a rope of hemp palm 5 mm in diameter) was suspended with a brick attached to its lower end as a sinker. After one or two months, the test rope was taken up and the number and size of the attaching ark shell spat were examined in the following manner. The rope was cut off at every 30cm starting from its lower end. Within each 30cm section a portion showing the average spat attachment was selected, and two 3-cm pieces were cut off from this portion, one for examination and the other for preservation as the specimen. The spat attaching to the 3-cm piece were counted and their number was multiplied by 10 to obtain the number of spat attaching to the 30cm section. The latter was added up to give the number of spat attaching to an entire test rope. It was assumed that artificial spat collection can be conducted with profit in those waters where more than 1,000 spat attach to a test rope. Examination of the test ropes hung in July and taken up in August showed that such waters lay near Ara-shima and covered an area of 600 ha. The test ropes hung in July and taken up in September indicated that such waters lay south by east of Daikon-jima (bounded by the lines connecting Kame-shima, E-jima, Hanyu and Ronde) and in Yonago Bay, covering an area of I ,500 ha. Those hung in August and taken up in September indicated that the waters suitable for profitable spat collection extended from the line passing Tsuzuki-jima and the mouth of Iinashi River eastwards to Yumigahama and also in Yonago Bay, a total area of 2,000 ha. The test ropes hung in September and taken up in October showed that such waters measured 3,000 ha in area and covered all over the central part of the Nakanoumi (between the line passing Kame-shima and E-jima and the line passing Itohana and Nyuko).Putting these data together, ark shell larvae were densely distributed chiefly in Arashima Bay in the earlier part of the season; with the progress of the season, they became more abundant and the center of their distribution were shifted eastwards to be found along Yumigahama and in Yonago Bay; as the season advanced further, they were found in abundance near Daikon-jima. The data indicated also that spat collection could have been conducted profitably over an area of 4,000ha. It was indicated by these findings that the ark shell larvae of advanced stages were not abundant in the west of the Itohana-Nyuko line where the adult of this species were relatively abundant (this part of the Nakanoumi is now utilized as the culture ground of adult ark shell), and that the larvae were carried by tidal currents and accumulated in high densities in the vicinity of Yumigahama and in Yonago Bay where the parent stock was scarce.( 6) RELATION BETWEEN VERTICAL DISTRIBUTION OF LARVAE AND ATTACHMENT LAYER OF SPATOne can not collect spat successfully without knowing the proper depth at which his spat collectors are to be placed. In the summer of 1942 we investigated into how the attachment of ark shell spat varies according to the depth. The investigation took place at St. 4, 1.5km south of Watari-jima, and at St. 7, halfway between Ito Village and Tsuzuki-jima.The spat collector employed in this investigation was a series of six oblong screens of hemp palm fibre connected lengthwise. Each screen consisted of a sheet of hemp palm fibres stretched on a wooden frame 3 cm thick and measuring 9 x 90 cm inside.Two collectors were hung in the sea every week, one being taken up after two weeks and the other after four weeks. When taken up, the collector was cut off at 30cm intervals starting from its lower end, and two 3 x 3 cm portions showing average spat attachment were cut off from each 30cm section, one for examination and the other for preservation. The number and size of the spat attaching to either side of a 3 x 3cm piece were examined, and the former was multiplied by 30 to obtain the number of the spat attaching to the 30cm section (Text-fig. 12).This investigation revealed that in the collector suspended during July 22-August 5, 78% of the spat were concentrated within the 1.5m wide zone corresponding to water depths 4.5-5. 7m, while 77% of the attaching spat were concentrated within the 1.5m wide zone corresponding to water depths 3.3-4.5m in the collector suspended during July 31-August 12 (Text-fig. 13).In 1943 we investigated into why the width and level of the attachment zone vary like this. 30 L of sea water were drawn up with a hand pump from different depths of the water column at 60 cm intervals, the number of the ark shell larvae contained in the water was determined, and this was correlated to the specific gravity (σ 15) of the sea water. The results indicated that the ark shell larvae of advanced stages mostly occurred in the layer of the specific gravities of about 1.020 within the range of 1.018- 1.022, and that the level of such water layer perfectly coincided with the level of the zone of concentrated attachment observed on the spat collector (Text-fig. 14).If atmospheric pressure is relatively low and it is rainy in the summer, relatively large volume of sea water flows into the Nakanoumi, where a marked discontinuous layer of salinity is developed. In such a year (e.g., 1943) the ark shell larvae are concentrated within a narrow zone, and, consequently, artificial spat collection is very successful. On the contrary, if it is dry in the summer with high atmospheric pressures, discontinuous layer of salinity does not develop and salinity remains low in the Nakanoumi. In such a year (e.g., 1944) artificial spat collection is not very successful (Text-figs. 16 - 17).(7 ) ARTIFICIAL COLLECTING OF SPATIn the Nakanoumi the upper layer is occupied by much diluted sea water, and the saline sea water of the bottom pushes upwards against it, creating inbetween a zone with salinities suited to ark shell larvae. The narrower is this zone, the more densely are the larvae concentrated within it.After the aforementioned investigations were carried out, it became a practice to locate the discontinuous layer of salinity prior to the spat collecting season, to forecast the probable depth of the attachment zone of spat from the vertical distribution of specific gravity, and to set out the spat collectors in conformity with such information. This practice almost eliminated the chances of failure, and made it possible to collect the spat with economic profit. In the succeeding years the relationship between the swimming layer of larvae and the vertical distribution of specific gravity was further investigated, and it was found that, from practical point of view, the width of the attachment zone may be regarded as 75 cm. The spat collector 75 cm in length was therefore designed. The collector of this type, suspended at the depths of 3.6-4.8 m, has ever since been used very successfully.(8) NEW TYPES OF SPAT COLLECTOR(i) The ark shell larva swims about moving its foot and searching for an attachment surface. When it arrives at an attachment surface, the presence of which it detects from the changes in current velocity, it crawls over it and attaches itself to a safe spot. The larvae tend to attach collectively to the hollows of a surface, for example, the furrows between the strands of a rope. The reason why they preferably attach to hollows is not clearly understood. Some of the tentative explanations are that attachment may be securer in the hollows than on other parts of a surface, that in a hollow the larvae may be hidden better from enemies, or that food particles may be gathered near the hollows owing to eddy currents.Taking advantage of the tendency for the spat to gather in hollows, I devised the spat collector of Wara-mabushi type, which were made of rice straw, the easily obtainable material. This collector was put into practical use. It consisted of a two-stranded rice straw rope bearing the "bristles", 6 cm long, of rice straw around it. This bristled rope, reminiscent of a bottle brush without a handle, was prepared by placing pieces of rice straw cut to 12cm between and across the entire lengths of two fine rice straw ropes when the latter were twisted together into a rope with a rope-making machine. The bristled rope was cut into the pieces 1.8 m long, and each piece was looped. Three loops were tied together to make up a "bunch" and used as a collector. The collectors were stained with coal tar before use, in order to prevent their decay in the sea. However, a large number of collectors coated with wet coal tar proved to be a problem, because they required a large space for drying and were very embarassing to handle.In order to remove this difficulty, I devised the collector of Shida-mabushi type. It was of the same structure as its predecessor, the Wara-mabushi collector, but was made of# 19 galvanized wires and 12cm long palm fibres instead of rice straw strands and 12cm long rice straw. The advantages of this collector are:(a) It collects twice or thrice as many spat as the Wara-mabushi collector.(b) It is light and convenient for handling.(c) It sinks in the water when a small sinker is attached. As the weight of the collector including the sinker is small, supporting material such as piles and rope is much saved.(d) As it is light, the shipping cost of spat is cut down. As an example, 1,500 collectors ("bunches") attached by 60 million spat were transported from Shimane Prefecture to Okayama Prefecture on a single truck.(ii) Method of spat collectionBamboo poles I 0 m long are set up 4.5 m apart in the spat collecting area, and a cross piece, which is a bamboo pole about 4 m long and to which ten bunches of Shida-mabushi collectors have been fastened at equal intervals, is suspended horizontally from and between two neighboring poles with ropes.The collectors are lowered to the depth which has been determined in advance by examining the vertical distribution of the specific gravity of sea water as well as that of the ark shell larvae of advanced stages.At the beginning of the season, a few collectors are suspended every day for trials; they are taken up on the next day and the number of the attaching spat is examined. Once the attaching spat rapidly increase in number, all the collectors are set out as quickly as possible; the work is continued day and night, since the peak of attachment lasts for only 3 or 4 days. The collectors with which spat were collected in the summerof 1953 were attached by an average of 97,515 spat per collector. 90% of the spat were below 2 mm in shell length, when they were shipped for the nursery beds in Okayama and Hiroshima Prefectures.(9) CULTURE OF SEED(i) If the spat are left in the Nakanoumi as they are attached to the collector, they grow to the shell lengths of 2-3 mm in one month and to 5-l 0 mm in three months, and then detach themselves from the collector in November. Once they are detached from the collector and fall to the sea bottom, their recovery is not assured, because they may be damaged by red tide or washed away by storms. In order to prevent such loss, the spat are removed to safer places before their detachment from collectors, and raised there under proper protection until they grow to the size usable as the seed. In practice, the collectors carrying the spat 2-3 mm in shell length are shipped to the seed culture grounds in Okayama and Hiroshima Prefectures by rail or truck after September. If the spat are transported by truck, as is usually done in recent years, 60-70% of them fall off from collectors during the transport.Culture grounds for the ark shell seed are chosen from among such places where the following conditions are fulfilled:(a) The weather is comparatively calm during the winter, the northwesterly wind (i.e., the monsoon) being not very strong.(b) The sea floor is either about 30-60cm above the mean low water level to be exposed during the low waters of spring tide, or about 30-60cm below mean low water springs.(c) The bottom is composed of sandy mud and is hard enough to allow a man to walk about without stepping deep into it. The KMn04 consumption of the bottom sediment is below 20 mg/L.(d) Good circulation of the water is provided by tidal currents, and the velocity of the flood current exceeds 20m/min.(e) The site is not far from a village or town, and is convenient for the caretaker's frequent visits and for the recovery of the spat that have fallen to the sea floor.(ii) Method of seed cultureOn the seed culture ground, which is usually a part of the tidal flat of the level of the low water springs, racks of bamboo poles 60cm high are set up. The spat collectors attached by spat are hung over the top of the rack. The spat that have fallen off from collectors during transportation are sown on the tidal flat between the racks and nearby at the rate of 15,-30,000 spat per m 2 • In one night they attach themselves either to the dead shells and gravels scattering on the tidal flat or to the spat collectors hanging from the racks, exhibiting a tendency of crawling up to a level a little above the sea floor. When a storm is expected, a fence of twigs 60cm high is set up around the culture ground; the fence serves as a wind and wave break.Transferred to a favorable habitat, the spat grow rapidly and, when they reach a certain size, detach themselves from the substrata one by one to begin their life in the mud. At the beginning, the young ark shell live partially buried in the mud, so that part of their shell remains exposed. Gradually they move downwards, until they are completely buried in the mud.The spat transferred to a seed culture ground mostly grow to young clams 12-15 mm in shell length by next March or April, when their population density is5,500-6,600 per m2 of the sea floor in average, with the maximum of 13,200-16,500 per m2 , and 2,000-3,000 of them fill a liter measure. The young ark shell can be used as the seed when they attain the shell lengths from 10.3 to 12.9mm. At this size 1,100-1,700 clams fill a liter measure. After the clams reach nearly this size, they are harvested to be used as the seed. The harvesting is done on several suitable occasions; each time the clams are taken up in such a way that their density is evened all over the culture ground. The seed are sown on the culture ground for adult clam at the rate of 100,000 per are. 10- 20% of the spat transferred to a seed culture ground in October of the previous year are recovered as the seed. The major part of the loss is ascribable to such predators as sea bass (Lateo/abrax japonicus), eel, octopus, blue carb, drills, starfishes, etc. Yet another important enemy of the young ark shell is the bivalve Brachidontes, which suffocates the clam by secreting a thick mass of byssus over the bottom. Methods to protect young ark shell from these enemies should be found out through the studies in the future.(iii) Culture of adult ark shellThe seed clam, 10 mm in shell length, that are sown on the culture ground grow to a shell length of 32mm in one year, and is ready to be marketed. 30-40% of the seed survive to attain this size. When the clams grow rapidly on the culture ground, they are harvested within one year after they are transplanted as the seed. The meat of the ark shell is marketed raw, or canned after properly seasoned.In Kasaoka Bay, Okayama Prefecture, 500,000 kan (1,875 metric tons) of the ark shell (shell included) worth 30 million yen are produced annually by the culture method decribed above. The Konoshimauchi Fishermen's Co-operative, which is the most active producer in this area, has been producing this clam at the rate of 886 kan (3.25 metric tons) or 55,824 yen from each 10 ares of the culture ground per year.(10) PROBLEMS OF ARK SHELL CULTURE IN FUTUREAs was decribed in the foregoing sections, the ark shell can be cultured intensively on a commercial scale by using the spat collected in the Nakanoumi. However, many problems must be solved through future studies for the culture of this clam to be practiced in many localities along our coasts. Some of these problems follow.(i) Spat collection: Considerations should be made from both technical and economic point of view, so that the spat may be collected more surely and abundantly. For example, scientific investigations are needed to determine whether spat collecting area can be further extended in the Nakanoumi and to find out the maximum number of the collectors that can be placed profitably in a unit area of spat collecting ground. Since the density of ark shell larvae is higher in the east of the present spat collecting area than in the present area, it seems advisable to extend spat collecting area in that direction, although the proposed area is farther from Ito Village, the base of the operation, than the presently utilized area.(ii) Possibility of seed culture in the Nakamoumi: At present, the ark shell spat collected in the Nakanoumi are transported to Okayama and Hiroshima Prefectures and raised there to the size usable as the seed. There are such areas in the Nakanoumi which are free from the harmful effects of the red tide and where the exchange of water is good owing to tidal currents. It may be possible to raise spat to the size of seed in these areas.(iii) Culture ground for adult ark shell: Commercial culture of the ark shell may be unsuccessful in the waters inhabited by a natural population of this species. Success or failure of the ark shell culture depends not only on the natural conditions of the culture ground but also on the method of transplanting the seed and on the care with which the clams are tended during the culture period. The seed of the ark shell are transplanted for either of the two different purposes: in some cases, they are expected to establish and reproduce themselves in the transplanted area, so that their offspring may be harvested after reaching the marketable size; in other cases, the clams sown as the seed are harvested when they reach the marketable size and their natural breeding is not taken into account. Although ark shell seed have been transplanted to many localities, success has not been reported in many cases. A plan for the transplantation of seed should be based on adequate scientific knowledge concerning the ark shell culture, and not on subjective judgement or mere experience.(iv) Offshore extension of culture ground: The possibility of extending the culture ground offshore should be investigated. One of the disadvantages of an offshore culture ground, as compared with the present ground, would be that the clam can not be harvested with the inexpensive hand dredge owing to the greater water depths. However, cost of harvesting may be cut down considerably by using the dredge that can be pulled by a boat. It is necessary to study the growth rate and the survival rate of the ark shell in the offshore waters and to estimate how much production of this clam can be expected from a unit area.(v) Careful tending: Success or failure of the culture of an aquatic animal largely depends on whether the animal is tended with sufficient care or not. It is therefore very important to guide the culturists in such a way that they take more interest in shellfish culture and take better care of their clam. One of the suggested methods is to have each culturist make it a practice to count, every month, the number of the clam necessary to fill a vessel of a fixed capacity. By doing so, he can tell whether his clams are growing or not, since the decrease in the number means the increase in the size of the clam. Moreover, this practice will induce him to take more interest in the clam on the culture ground and to tend them with better care.(vi) Joint operation: The commercial culture of the ark shell is now operated jointly by the culturists in Shimane Prefecture, who take charge of spat collecting, and those in Okayama, Hiroshima or Saga Prefecture, who take charge of the latter phases of the culture. The proceeds from the sale of the clam of marketable size are divided between the two parties in the ratio of 5 to 5 or 6 to 4. This type of operation, as long as it is based on the common consent of the concerned parties, can be regarded, at present, as one of the rational systems of managing the commercial culture of the ark shell, either from the viewpoint of the progress of the techniques or from that of the propagation of the culture method.(11) PRINCIPLES FOR ARTIFICIAL COLLECTING OF BIVALVE SPAT AND THEIR APPLICATIONIn order to collect bivalve spat successfully, it is necessary to make clear the morphology and ecology of the planktonic larvae, especially the season of appearance of the full-grown larva and the depth of the attaching layer. Since the attaching behaviors of the full-grown larvae of various bivalves generally have a common feature, the same principles of spat collecting apply successfully to many species. These principles may be summarized as below:i) The collectors should be arranged to each other in such a way that the passage of water is relatively free between them, so that the larvae, being carried by the water current, may arrive at the collector quickly.ii) The surface of the collector should be provided with numerous hollows of appropriate size, since the larvae tend to attach to the hollows or the spots sheltered by protrusions on the surface of the collector.iii) The spat should be raised under adequate protection from predatory enemies until they attain the size usable as the seed. Otherwise, their survival rate is very low. The items mentioned above are explained further in detail.(i) The full-grown larvae, which are ready for attachment, swim protruding their foot from the shell. These larvae, as observed in table aquaria, swim with their foot vibrating to the right and left as if searching for proper spots for attachment. The foot, being covered all over with numberless cilia, serves as a kind of sense-organ to perceive the water current. While swimming by its own force with its foot as a rudder and by taking advantage of the water current, the larva finds its way toward the periphery of the water path and finally arrives at the collector. Therefore, the collectors should be arranged in such a way that the circulation of the water is not hindered, in order to collect spat efficinetly.(ii) The larva, after attaching to the collector, seeks for the best spot for its settlement by crawling about with its foot. If no proper spot can be found, it detaches itself and swims to another attachment surface, where it again searches for a spot suitable for settlement scrupulously. Observation of the oyster spat, soon after its attachment on the collector suspended in the sea for 24 hours, reveals infallibly that it is the concave spots where the larvae attach.As a whole, it is in the hollow or behind the protrusion that shellfish larvae prefer to attach, and this is also the case in the cypris larva of Balanus.These hollows, when cast in paraffin, measure 0.4-1.8 mm in depth and. 1.2-4.0 mm in width. The shell of Pecten a/bicans is provided with numerous depressions, and this may be the reason why the shell of this scallop is preferably used as the collector of the oyster spat. Therefore, I devised the "double-net collector", which has an ideal structure full of hollows. It is made of the two or three sheets of nettings of fine mesh, which are placed one over another and sewed together into a sheet. The size of the mesh is about 6 mm as stretched. When one netting is placed over another, the meshes of the former is shifted by half a mesh from those of the latter, so that each knot of one netting falls in the center of a mesh of another netting. With this collector, bivalve spat can be collected in abundance. For example, an average of 2,263 ark shell spat of 2 mm in shell length attached to each 10 x l0cm area of this collector. It has been shown also that this collector can be used practically for collecting the spat of the pearl oyster, Pinctada martensii.(iii) The spat grow rapidly after attaching to the collector, secreting a hard shell containing calcium carbonate. During the growth, however, they reduce in number owing to their own death and the attack by predators (fishes and crabs). It is most important in the production of bivalve seed to raise the survival rate of the spat by preventing such death and attack. The double-net collector, with its numerous hollows formed by staggered meshes, not only serves as a good collector, but also has a superb advantage in protecting the spat from predatory enemies, and moreover is quite suitable for the culture of juveniles. So that, it may be called an appropriate collector for shellfish culture.