ABSTRACT. Little IS known about the relative significance of heterotrophic bacteria in open-ocean oligotrophic environments. The pelagic waters of the Levantine Basln of the eastern Mediterranean Sea are among the most oligotrophic on record. We surveyed the spatial distribution of bacterial abundance, biomass and production along 2 transects of the pelagic waters of the southern Levantine Basin to assess which changes in these parameters may occur in association with varylng physlcal structure and c.hlorophyl1 concentrations, to calculate the relative b~o m a s s contl-ibutions of bacter~a and phytoplankton, and to estimate the magnitude of carbon flux from phytoplankton to bacteria. Chlorophyll had an average concentration of 134 * 85.4 n g 1-' and was relatively uniform throughout the upper 200 m Bacterial numbers ranged from 0.40 to 3 90 X 10H cells 1.' and were generally highest above l10 m. Cocci cells comprised 87";1 of the population with an average volume of 0.049 pm" Bacterlal numbers and biomass were notably high in the Ierapetra Eddy and Mersa Matruh Gyre. Although bacterial numbers and chlorophyll concentrations were not generally correlated, the mean bacterial number was accurately predicted from a regression equation uslng chlorophyll. Over the upper 200 m , bacterial b~omass (x= 603 mgC m-?) was on average about 50% of phytoplankton biomass (2 = 1235 mgC m l ) , which is contrary to other published studies reporting bacterial biomass equalled or exceeded algal biomass in oligotrophic marine waters. Bacterial production ranqed from 0 to 3.91 pm01 TdR 1-'

Prochlorophytes, cyanobacteria and eukaryotic ultraphytoplankton from the southern Levantine Basin of the eastern Mediterranean Sea were analyzed by flow cytometry to obtain measurements of cell abundance, relative cellular fluorescence and relative cellular hght scatter. Assuming that fluorescence is a proxy for chlorophyll and that light scatter is a proxy for cellular carbon, phytoplankton biomass can be expressed as the sum (over all cell groups) of adaptive cellular characteristics (i.e. chorophyll and carbon) weighted by cell abundance. On this basis, much of the carbon appeared attributable to eukaryotic ultraphytoplankton, but chlorophyll was more evenly partitioned such that the contributions from prochlorophytes and cyanobacteria were also significant. The subsurface chlorophyll maximum coincided with the maximum in total fluorescence but not with the maxlrnum abundance of cells nor with the presumed maximum in the carbon biomass of ultraphytoplankton.

The Cyprus Eddy, a warm-core eddy southeast of Cyprus, was sampled towards the end of an exceptionally cold winter in early March 1992, within 4 d of a storm and w~thin 24 h of an intrusion of cold air. Depth profiles of temperature, salinity and dissolved nutrients showed an active deep mixed layer from the surface to ca 500 m at the core of the eddy, while at the eddy boundaries the mixed layer extended only to 150 m. Microb~al populations were evenly distributed over the entire upper 500 m at the core station, as iiidicated by chlorophy!! and high performance liquid chromatography (HPLC)-determined pigment composition, by flow-cytometric analysis of the ultraphytoplankton, by direct counts of 4',6-diam~dino-2-phenylindole (DAP1)-stained bacter~a and %-thymidine measurements of bacterial activity. As far as we know, thls is the first deta~led description of the microbial populations in a warm-core eddy during the bloom season. The integrated water column chlorophyll content, 59 mg m-2 at the core and 45.5 mg m-2 at the boundary, was more than double the typical late autumn values, suggesting a bloom was occurring. Noticeably, this bloom was not delayed until the establishment of summer stratification as has been observed previously in warm-core eddies. While theoretical considerations based on the calculated critical depth at the core of about 300 m suggested that a bloom should not have occurred, our data jointly with previous data from the Cyprus Eddy support the hypothesis that interim periods of quiescence between mixlng events enable bloom development even when the mixing depth is greater than the critical depth. Added nutrients and dilution of grazers, both resulting from the deep mixing, probably contributed jointly to the enhanced productivity. Based on phytoplankton 11ght-shade adaptation features and cellular chlorophyll fluorescence per cell, we calculated that the rate of vertical mixing in the core was at least 30 m h-'

No abstract