The commercially important sergestid shrimp, Sergia lucens (Hansen), has long been considered an endemic species of Japan because it had been found only in Suruga Bay and neighbouring waters. Recently, however, a considerable amount of a similar shrimp was caught by trawl nets off Tung-kang, southwestern Taiwan. This shrimp is distributed at depths of 100-300 m on the continental slope, around a deep submarine canyon adjacent to the mouth ofthe Kao-ping River. A morphological comparison with specimens from Suruga Bay reveals thatthe shrimp is identical toS. lucens.However, a slight difference in the patterns ofwater-soluble proteins was observed in a thin layer isoelectrophoretic examination. A difference was also observed in the spawning season of the two populations, indicating sexual and geographic isolation.
Considering the distribution and phylogeny of the family Sergestidae, based on a possible evolutionary development from a benthic neritic organism to apelagic oceanic one along the generaSicyonella-Sergestes-Sergia, it is assumed thatS. lucensentered a lower epipelagic habitat in the coastal waters from the warm oceanic mesopelagic habitat of the original stock. A hypothesis is proposed that speciation ofS. lucensfrom the original stock occurred when it was trapped in a semi-enclosed inlet (the paleo-East China Sea Gulf) that existed at the present Okinawa Trough during the late Pliocene to early Pleistocene. The inlet was deep, but had a neritic environment due to drainage from ancient large river systems, including the paleo-Yangtze River. The species expanded its distribution to the neighbouring waters during the warm interglacial period. However, a rapid rise in sea level after 14, 000 years significantly changed the environmental conditions in the distributional area and the species could not expand into a neritic environment, which was too shallow for survival. Accordingly, S. lucenspopulations remained only in Suruga Bay and Tung-kang waters, where the environment has remained stable for the last 17, 000 years or more. The two areas have the following common characteristics:
1. A large amount of fresh water is discharged into the deep submarine canyon adjacent to the river mouth.
2. The northeastern part is surrounded by land so that a direct iflow of boreal water into the environment is blocked.
3. Warm Kuroshio extension water intrudes and mixes with water from the rivers, and for Suruga Bay there is a current system that keeps distribution of the eggs and larvae restricted to the local habitats.
Taiwan annually exported some tens of metric tons of the dried S. lucens to
Japan for the last few years, but the standing stock of Tung-kang waters is probably not as large as that in Suruga Bay.

View full abstract

Size-fractionated primary productivity and chlorophyllaconcentration were studied at two stations in the temperate neritic waterof Funka Bay, Japan, from April 1984 to May 1985. Size distributions of phytoplankton were discussed in relation to nutrient availability. In the central part of the bay, 66% of the annual primary production occurred during the spring phytoplankton bloom with 95% of the spring production being accounted for by thegreater than 10μm size fraction, which was dominated by diatoms. The increase in this large fraction was enhanced at both stations when nutrient concentrations increased in the bay's upper layer. Under low nutrient concentrations during summer, small phytoplankters (<2μm) accounted for 40 to 75%(average 60%) of the total¹⁴C uptake atthe central station, and from 25 to 59%(average, 45%) at the coastal station. However, a sudden nutrient enrichment at the coastal station during the summer triggered the growth of the large size fraction. These seasonal and regional changes in total¹⁴C uptake were attributed to the large size fraction, composed mainly of diatoms. From the decreases in various nutrients during diatom blooms, it was further suggested that the predominance of diatoms was determined, not only by nutrient concentrations, but also by their relative availability.

View full abstract

Measurements of phosphorus content were made for eleven species of copepods, two species of mysids, shrimps (including mysis stage) and chaetognaths, and one species ofNoctiluca, cirriped (naupliar stage) and crab (zoeal stage) from the Inland Sea of Japan. The body phosphorus content (P, inμg) is highly correlated to prosome length (L, inμm) for copepods, which is expressed as logP=2.90×10gL-9.34. Meanphosphorus content expressed as Dercent of drv weight is 1.30% for copepods, 1.39% for mysids, 0.63% for a decapod shrimpAcetes japonicus, and 0.80% for chaetognaths. Noctiluca contains phosphorus of only 0.36%. The comparatively higher phosphorus content of suspension-feeding copepods may be related to the constant supply of phosphorus-rich phytoplankton in this region.

View full abstract

In situmeasurements of the primary productivity of ice algae and phytoplankton werecarried out in the fast ice area near Syowa Station (69°00'S, 39°35'E) during the austral spring and summer of 1983/84. Standing stock of ice algae reached a maximum of 45.1mg chlam^-2in late October. Phytoplankton standing stock attained a value of 3.57 mg chlam^-2in mid-January. Primary production of ice algae in late October (7.64mgCm^-2hr^-1) was 14 times greater than that in mid-January (0.54 mgCm^-2hr^-1). Production in the water column in mid-January (3.46mgCm^-2hr^-1) was 50 times greater than that in late October (0.07 mgCm^-2hr^-1). These results indicate a substantial production by ice algae in the spring and by phytoplankton in the summer period.

View full abstract

Variations of current velocity of the Kuroshio are examined using the 1965-1983 sea-level difference between Naze and Nishinoomote, located on the offshore and onshore sides of the Kuroshio in the Tokara Strait south of Kyûshû.
Interannual variations of Kuroshio velocity are large, especially at periods longer than five years and around 2.1 years. They are almost determined by those of sea level on the offshore side of the Kuroshio.They are highly coherent with the offshore sea level at periods longer than 1.7 years, andincoherent with the onshore sea level at periods longer than 2.8 years.
The mean seasonal variation averaged for 19 years is at its maximum in July and at its minimum in the second half of October, with a sharp decrease in August and September. However, such a variation does not repeat every year. Amplitude, dominant period and phase are greatly differentby year, and they can be roughly divided into four groups: small-amplitude group, semiannual-period group, and two annual-period groups with different phases. The only feature found in almost all years is a weak velocity from September to December.
The amplitude of seasonal variation tends to be large in the formation years of the large meander (LM) of the Kuroshio and small during the LM period. It is also large in the years preceding El Nino, and diminishes remarkably in El Niño years.
Kuroshio velocity in the Tokara Strait is incoherent with position of the Kuroshio axis over the Izu Ridge, but highly coherent with 70-day variations of coastal sea levels which are dominant during the LM period.

View full abstract

Historical observations of the surface current and daily mean sea level during the period 1965-1985 are analyzed in order to examine the geostrophic balance of the Kuroshio currentin the Tokara Strait and near the Izu Islands. The variation in the sea level difference across the Kuroshio is associated with a variation in surface current velocity as predictedby the theory of geostrophic balance. However, the slope of the linear relation between the current velocity and sea level difference is smaller than the theoretically predicted value by about 30%. This disagreement may be ascribed to the effects of the centrifugal force and the occasional rise in sea level due to storm surges.
Absolute mean sea level differences between the tidal stations are estimated by making use of the empirical relationship between the surface current and sea level difference. Estimated differences are: 87.4±22.1 cm between Naze and Nishinoomote, 24.3±9.2 cm between Miyake and Minamiizu, 41.3±17.7cm between Miyake and Mera and 45.1±8.8 cm between Hachijyo and Miyake. The absolute value of sea level difference between Miyake and Minamiizu and thatbetween Miyake and Mera may be about 30cm, since geodetic levelling tells us that the meansea level at Minamiizu is nearly equal to that at Mera.

View full abstract

The present study investigates the way an ocean filled with homogeneous warm water is cooled by prescribing cold water formation inside the ocean in the southern part of the southern hemisphere using multi-level numerical models. Cooling of the whole ocean starts with introduction of the cold water from the formation region into the deepest part of the oceanin the equatorial and eastern boundary regions by Kelvin wave-type density currents. The cold water along the eastern boundary extends westward as a Rossby wave-type density current setting up an interior poleward flow, and hits the western boundary to form a northwardflowing boundary current in the northern hemisphere. Only then does the western boundary current cross the equator. Cooling of the rest of the ocean basin is accomplished by upwellings in the interior and also along the coasts. During this introduction the cold water is mixed with surrounding warm waters, and the thermocline, rather than forming just below the top level where heating is imposed, tends to spread down to deeper depths. Consequently the circulation at a steady state has a significant vertical structure such that the maximum upwelling in the interior occurs in the mid-depths, and only the deeper part of the deep ocean yields the Stommel and Arons circulation pattern. In the equatorial region higher vertical mode motions dominate, and a set of alternating zonal jets forms along the equator.

View full abstract