A multidimensional statistical study of similarities between 74 notes used in perfumery

From a data bank of 2467 odoriferous products, the similaritiesbetween 74 notes used in perfumery were calculated. The similaritymatrix (74,74) contains – 63% of zero values and showsthat only a few pairs of notes present high similarity coefficients.A fine analysis using ascending hierarchical taxonomy with thecomplete linkage procedure shows that 14 notes are isolatedwhile 60 notes are regrouped in 27 groups containing two tofour notes. The isolated notes correspond to well-defined structuralparticularities. Some pairs of notes or groups present similaritiesin their chemical structures but some groups are built on thebasis of semantic processes. This study shows that the notesare generally independent, with no strict hierarchy among them,and rules out the existence of primary odors.     

Portal imaging: Performance improvement in noise reduction by means of wavelet processing

This paper discusses the suitability, in terms of noise reduction, of various methods which can be applied to an image type often used in radiation therapy: the portal image. Among these methods, the analysis focuses on those operating in the wavelet domain. Wavelet-based methods tested on natural images – such as the thresholding of the wavelet coefficients, the minimization of the Stein unbiased risk estimator on a linear expansion of thresholds (SURE-LET), and the Bayes least-squares method using as a prior a Gaussian scale mixture (BLS-GSM method) – are compared with other methods that operate on the image domain – an adaptive Wiener filter and a nonlocal mean filter (NLM). For the assessment of the performance, the peak signal-to-noise ratio (PSNR), the structural similarity index (SSIM), the Pearson correlation coefficient, and the Spearman rank correlation (ρ) coefficient are used. The performance of the wavelet filters and the NLM method are similar, but wavelet filters outperform the Wiener filter in terms of portal image denoising. It is shown how BLS-GSM and NLM filters produce the smoothest image, while keeping soft-tissue and bone contrast. As for the computational cost, filters using a decimated wavelet transform (decimated thresholding and SURE-LET) turn out to be the most efficient, with calculation times around 1 s.     

Larval fish assemblages and geostrophic circulation in Bahia de La Paz and the surrounding southwestern region of the Gulf of California

We analyze spatial–temporal relationship between larvalfish assemblages and geostrophic surface flow in Bahíade La Paz and the neighboring Gulf of California (May, Julyand October 2001 and February 2002). The analysis of fish larvaedistribution in relation to geostrophic circulation and hydrographyis an innovative interdisciplinary approach for the understandingof fish larvae ecology. The Bray–Curtis Index definedtwo larval fish assemblages with spatial–temporal variations:coastal assemblage dominated by epipelagic coastal species (e.g.Sardinops caeruleus) and oceanic assemblages—oceanic andtransitional oceanic assemblages both dominated by mesopelagicspecies (e.g. Vinciguerria lucetia and Benthosema panamense)but with different larval abundance. The assemblage variationsappear to be related to water exchange between the bay and theGulf through the North Mouth. During July–October, thegeostrophic flow through the entrance is strong, and the oceanicassemblages spread in the whole bay, whereas during February–Maywhen the geostrophic transport is weak, the coastal assemblageis distributed over the whole bay. The strong summer–autumnwater interchange between bay and gulf is in agreement withthe annual evolution of the surface water properties insidethe bay, from high-salinity Gulf of California Water duringwinter–spring to fresher Tropical Surface Water duringsummer–autumn, when the highest species number was recorded.     

Comparison of Four Methods Calculating the Seasonal Pattern of Plant Growth Efficiency of a Kiwifruit Berry

Four methods of determining the substrate requirements for synthesisof a kiwifruit [Actinidia deliciosa (A. Chev.) C. F. Liang etA. R. Ferguson var. deliciosa cv. Hayward] berry were comparedusing data derived from common kiwifruit berry samples collectedfrom anthesis to fruit maturity. The four methods were basedon fruit proximal analysis, elemental analysis, heats of combustion,or tissue carbon content. All methods gave similar patternsof seasonal costs and values of final cost to the plant (mean1.21 g glucose g^–1 season^–1) but there was lessagreement for growth respiration (mean 0.147 g glucose g^–1season^–1). This is the first time that a continuous recordof growth cost over the course of development has been presented,and the trends in seasonal cost reflect the uptake into andsynthesis of the different biochemical constituents in the fruit.The differences between the results of each method reflect theunderlying assumptions used in their development. It appearsfrom this work that the method of McDermitt and Loomis (1981),utilizing elemental analyses, is most preferred. Actinidia deliciosa (A. Chev.) C. F. Liang et A. R. Ferguson var deliciosa cv Haywood, kiwifruit, true growth yield, plant growth efficiency, production value, glucose value, bioenergetic cost     

Comparison of Glycolipids and Plastids in Callus Cells and Leaves of Alnus and Betula

The fine structure of plastids and fatty acid composition ofglycolipids (e.g. monogalactosyl diacylglycerides, MGDG; digalactosyldiacylglycerides, DGDG) in callus cells of Alnus glutinosa,A. incana and Betula pendula cultured in light was comparedwith that in intact leaves. The tissues were qualitatively verysimilar but a rather high amount oflignoceric acid (24:0) wascharacteristic for the callus of A. incana. This fatty acidwas found only in trace amount in other tissues. Linolenic (18:3)and palmitic (16:0) acids are the most abundant (25–65%and 17–27% respectively) fatty acids in all tissues studied.The proportion of 18:1 and 18:2 was much higher in the calluscompared with corresponding intact leaves, which are especiallyrich (48–65%) in 18:3. In callus cultures a higher proportion(17–19%) of linoleic acid (18:2) is found in both Alnusspecies than in the two callus strains of Betula (9–12%). All leaf and callus samples contained esterified steryl glycosidesand two cerebrosidelike spots in thin-layer chromatography,but they were more prominent in callus cultures than in leaves.The callus cells have plastids with rather well developed thylakoidswhich explains the similarity of the main glycolipid components(MGDG and DGDG) to that of leaves. (Received April 23, 1984; Accepted August 17, 1984)     

Sulcal patterns of fossilTheropithecus baboons: Phylogenetic and functional implications

Sulcal patterns were determined from endocasts of fossilTheropithecus (T. oswaldi andT. darti) and compared to the sulcal pattern of extant geladas (T. gelada). A comparison of the configurations of central and arcuate sulci suggests that the cortical motor face representation is relatively enlarged in extantTheropithecus, whereas the cortical motor and sensory face representations were enlarged inT. oswaldi. A relatively expanded sensory and/or motor cortical face representation may be related to masticatory (e.g., “seed-eating”) and gestural (e.g., “lipflip”) functions. SinceT. oswaidi is more derived (expanded) in sensory face representation thanT. gelada, it seems unlikely that it was a direct ancestor of modern geladas. Rather, the two groups were separate by 1.8 million years ago and may have had a most recent common ancestor similar toT. darti from the Hadar Formation of Ethiopia between 2 and 3 million years ago.     

Trophic links in the lowland River Meuse (Belgium): assessing the role of bacteria and protozoans in planktonic food webs

Trophic interactions within the plankton of the lowland RiverMeuse (Belgium) were measured in spring and summer 2001. Consumptionof bacteria by protozoa was measured by monitoring the disappearanceof ³H-thymidine-labelled bacteria. Metazooplankton bacterivorywas assessed using 0.5-µm fluorescent microparticles (FMPs),and predation of metazooplankton on ciliates was measured usingnatural ciliate assemblages labelled with FMPs as tracer food.Grazing of metazooplankton on flagellates was determined throughin situ incubations with manipulated metazooplankton densities.Protozooplankton bacterivory varied between 6.08 and 53.90 mgC m^–3 day^–1 (i.e. from 0.12 to 0.86 g C^–1bacteria g C^–1 protozoa day^–1). Metazooplankton,essentially rotifers, grazing on bacteria was negligible comparedwith grazing by protozoa (1000 times lower). Predation of rotiferson heterotrophic flagellates (HFs) was generally low (on average1.77 mg C m^–3 day^–1, i.e. 0.084 g C^–1 flagellatesg C^–1 rotifers day^–1), the higher contribution ofHF in the diet of rotifers being observed when Keratella cochleariswas the dominant metazooplankter. Predation of rotifers on ciliateswas low in spring samples (0.56 mg C m^–3 day^–1,i.e. 0.014 g C^–1 ciliates g C^–1 rotifers day^–1)in contrast to measurements performed in July (8.72 mg C m^–3day^–1, i.e. 0.242 g C^–1 ciliates g C^–1 rotifersday^–1). The proportion of protozoa in the diet of rotiferswas low compared with that of phytoplankton (<30% of totalcarbon ingestion) except when phytoplankton biomass decreasedbelow the incipient limiting level (ILL) of the main metazooplantonicspecies. In such conditions, protozoa (mainly ciliates) constituted50% of total rotifer diet. These results give evidence thatmicrobial organisms play a significant role within the planktonicfood web of a eutrophic lowland river, ciliates providing analternative food for metazooplankton when phytoplankton becomesscarce.     

Carboxyarabinitol 1-Phosphate Phosphatase from Leaves of Phaseolus vulgaris and Other Species

Carboxyarabinitol 1-phosphate (CA1P) phosphatase activity occurredin leaves of 10 species examined, with the highest activityin leaves of Phaseolus vulgaris. Enzyme was purified from P.vulgaris 1,580-fold to a final specific activity of 6.1 µmolmin^–1 (mg protein)^–1. Structural characteristicsof positive effectors and substrate analogs for the CA1P phosphatasereaction were examined. Positive effectors were compounds thatcontained one phosphate group in close proximity to a secondphosphate or a carboxyl group (e.g. 2-phosphoglycolate, pyrophosphate,3-phosphoglycerate, and carboxyethylphosphonic acid). Many ofthe activators are structurally quite similar to CA1P, but werenot used as substrates. In addition to the natural substrateCA1P, carboxypentitol and carboxyhexitol bisphosphates wereshown to be good substrates (e.g. carboxyarabinitol bisphosphateand carboxymannitol bis-phosphate). A substrate arabinitol configuration(R) was preferred at C-2, and reactivity was lost when a hydroxymethylgroup was substituted for the carboxyl group. Despite structuralsimilarities to positive effectors, none of the tested reactionsubstrates could activate the enzyme. (Received November 11, 1996; Accepted February 3, 1997)     

Macroaggregates and their phytoplanktonic components in the Southern California Bight

The abundances of phytoplankton associated with scuba-collected,visible macroaggregates (i e. ‘marine Snow’) fromvarious euphotic zone depths of several nearshore and offshoresites sampled in late winter/early spring and summer were determinedby microscopic study. Such phytoplankters have a different potentialfor predator/prey interactions than they would as separate individuals.Mean macroaggregate concentrations of 1.7–7 6 1^–1and mean individual macroaggregate sizes of 27–175 mm³were observed at the different sites. Phytoplankters associatedwith the macroaggregates were generally a few percent or lessof the total phytoplankton (range for numbers. 0.2–2.5%.range for carbon, 0.2–7.1%. n = 11). Of the taxonomicgroups, pennate diatoms in general showed high relative associationwith macroaggregates. Compared to the mean abundance of phytoplanktonin an equivalent volume of surrounding water, macroaggregateswere algal-enriched by factors ranging from 6.2–1300 (median,65) for numbers and 6.3–2500 (median. 110) for carbon(n = 11). ATP and chlorophyll a showed degrees of associationwith, and enrichment in. macroaggregates similar to that ofphytoplankton carbon. Before assessing the overall importanceof phytoplanktonic associations with aggregates in terms ofpelagic food web consequences, the abundances and algal compositionof the smaller microaggregates must also be known     

Estimating nutrients recycling by tropical oceanic macroplankton-micronekton from biomass data

Simultaneous measurements of biomass and excretion of macroplankton-micronektonlead to similar results in tropical Atlantic and Pacific Oceans.Ratios are thus proposed to roughly estimate regenerated productionin oligotrophic tropical waters from biomass data: biomass figures(mg dry wt m^–2) should be multiplied by 0.32 to estimatethe amount of total nitrogen excreted (µg at m^–2day^–1) and by 0.027 to estimate the amount of total phosphorusexcreted (µg at m^–2 day^–1).     

Simulated annual plankton production in the northeastern Pacific Coastal Upwelling Domain

A microcomputer simulation model is presented that describesthe generalized plankton production dynamics, in the surfacemixed layer, of the Juan de Fuca Eddy located on the southwesternBritish Columbia continental shelf. The Juan de Fuca Eddy simulationmodel evaluates how the annual biomass production of diatoms,copepods and euphausiids is forced by plankton feeding interactions,seasonal variability in upwelling, water temperature and solarradiation, and generalized fish predation. The model estimatesannual primary production of 345 g C m^–2 year^–1and secondary production of 19.4 g C m^–2 year^–1for copepods and 6 g C m^–2 year^–1 for euphausiids,during 1985–89; -90% of the annual plankton productionwas generated during the April-October upwelling season. Perturbationsof 22 abiotic and biotic parameters, one at a time by ±10%of nominal values, indicated that oceanic variability (e.g.upwelling rate) most strongly affected primary production. Conversely,zooplankton production was most sensitive to variability inbiological parameters describing zooplankton grazing potentialand growth (e.g. gross growth efficiency). Simulated seasonalbiomass patterns of diatoms, copepods and euphausiids were foundto closely match empirical data. However, euphausiid biomassproduction in the Juan de Fuca Eddy alone was unable to meetthe demands of estimated pelagic fish consumption. Local Eddyeuphausiid populations had to be supplemented, from regionaleuphausiids. by a mechanism that is proposed to be linked tothe seasonal pattern and intensity of positive Ekman transport(upwelling).     

Perfusion of charophyte cells: a critical analysis of the method

The data obtained by different types of intracellular perfusionwere compared. As the ligated cells cannot be space-clamped,the efficiency of compartment-clamping was evaluated, showingthat the difference I/V (current-voltage) profile between space-clampedand compartment-clamped data could be approximated by a straightline. The time-dependence of the clamp currents was not affectedby the clamp technique. The comparison of different sets of data was quantified by fittingthe I/V curves with a mathematical model (Beilby and Walker,1996). The I/V curves of ligated cells perfused with 1 mM ATPshowed the closest similarity to intact cells with resting potentialsof –22010 mV (7 cells) and similar model parameter values.The cells under open-end perfusion with ATP showed less hyperpolarizedresting p.d.s (potential differences): –17512 mV (4 cells).For both preparations the —ATP data were similar withresting p.d.s at –80 12 mV (5 ligated cells, 7 open-endcells). The excited state was more pronounced in open-end cells(resting level: –5912 mV, 5 cells) than in ligated cells(resting level: –6512 mV, 7 cells). In open-end cellsthe pump responded faster to changes of ATP concentration thanthe excitation channels. The cells stabilized with Pb(NO₃)₂were strongly depolarized both with ATP: –8010 mV (6cells) and without: 010 mV (6 cells). Most model parametersdiffered from those in the intact cells. The excited state wasabolished. Key words: Intracellular perfusion, current-voltage characteristics, Chara, stabilization with Pb(NO₃)₂, ATP effects, voltage clamp techniques     

Non-Local Means Denoising of Dynamic PET Images

Objective

Dynamic positron emission tomography (PET), which reveals information about both the spatial distribution and temporal kinetics of a radiotracer, enables quantitative interpretation of PET data. Model-based interpretation of dynamic PET images by means of parametric fitting, however, is often a challenging task due to high levels of noise, thus necessitating a denoising step. The objective of this paper is to develop and characterize a denoising framework for dynamic PET based on non-local means (NLM).

Theory

NLM denoising computes weighted averages of voxel intensities assigning larger weights to voxels that are similar to a given voxel in terms of their local neighborhoods or patches. We introduce three key modifications to tailor the original NLM framework to dynamic PET. Firstly, we derive similarities from less noisy later time points in a typical PET acquisition to denoise the entire time series. Secondly, we use spatiotemporal patches for robust similarity computation. Finally, we use a spatially varying smoothing parameter based on a local variance approximation over each spatiotemporal patch.

Methods

To assess the performance of our denoising technique, we performed a realistic simulation on a dynamic digital phantom based on the Digimouse atlas. For experimental validation, we denoised PET images from a mouse study and a hepatocellular carcinoma patient study. We compared the performance of NLM denoising with four other denoising approaches – Gaussian filtering, PCA, HYPR, and conventional NLM based on spatial patches.

Results

The simulation study revealed significant improvement in bias-variance performance achieved using our NLM technique relative to all the other methods. The experimental data analysis revealed that our technique leads to clear improvement in contrast-to-noise ratio in Patlak parametric images generated from denoised preclinical and clinical dynamic images, indicating its ability to preserve image contrast and high intensity details while lowering the background noise variance.     

Shape, size and density of daytime surface swarms of the euphausiid Meganyctiphanes norvegica in the Bay of Fundy

Daytime surface swarms of ^{Meganyctiphanes norvegica} in the Bayof Fundy were examined using a variety of techniques to providemeasurements of their shapes, sizes and densities. Shapes andsizes were determined from two aerial photographs: swarms werespherical, ribbon-like or amorphous. were up to 28 6 m longand ranged in area from 0.4–111 7 m². Densities were measuredby a bag-sampling device which gave figures of up to 41 000animals m^–3 by photographic methods which gave figuresof up to 770 000 animals m^–3 and by a plankton net whichgave maximum values of six animals m^–3 Using the photographicmethod the maximum euphausud biomass was estimated to be 154kg m^–3 within swarms and the largest swarm measured wasestimated to contain up to 2 1 tonnes of M. norvegica. Meanpatch biomass estimates for the two aerial photographs rangedfrom 77.8–778 g m^–3 and 15 6–155.6 g m^–3which are similar to figures obtained by other authors usingintegrating sampling techniques at depth.     

Characterization of NADH-dependent Fe3+-chelate reductases of maize roots

Iron-deficient maize seedlings exhibit a starvation syndromecharacterized by an increase in different parameters such asroot fresh weight (+ 30%), protein (+ 25%) and plasma membrane-associatedNADH Fe³⁺ –EDTA reductase (NFR; +45%). NFR activity wasfound associated with 9 000g (20 min) and 110 000 g (1 h) sediments,purified plasma membrane and 110000 g supernatants. No differenceswere observed between the properties of reductases from Fe-starvedversus Fe-sufficient roots. The characterization of NFR wasundertaken. Low Mr forms (46 and 28 kDa, as detected by size-exclusionchromatography) were present in all fractions whereas 210 and110 kDa forms were unique in membranes and 110 000 g supernatants,respectively. The 210 kDa form was solubilized from microsomes and characterized.The enzyme is cetone-resistant and appears to be comprised largelyif not totally of the low Mr forms (46 and 28 kDa, correspondingto 30 and 32 kDa bands, respectively, in SDS-PAGE). The 210kDa form tended to break down to subunits following dilution,and the effect could be prevented by addition of 10% (v/v) glycerol.A three-step purification procedure for microsomal NFR was devised,consisting of acetone fractionation of lysophosphatidycholinesolubilized microsomes, Blue Sepharose CL-6B affinity chromatographyand a final size exclusion chromatography in the absence ofdetergent, resulting in a 700-fold purification of the 28 kDaprotein. The best electron acceptor for the purified 28 kDaform was ferricyanide (400µmol min^–1mg^–1 protein)followed by Fe³⁺–chelates (up to 200µol min^–1mg^–1 protein) and other compounds to a lesser extent (cytc, DCPIP).The 46 kDa form, on the other hand, had high ferricyanidereductase activity (about 300µmol min^–1mg^–1protein) and relatively low Fe³⁺–chelate reductase activity.The properties of NFR (high M, active forms, donor and acceptorspecificity, purification behaviour, large hydrophilic domains,size of subunits) suggest a relationship with the NADH-cyt b5reductase family of FAD-containing proteins. None of the latterflavoproteins is a transmembrane enzyme. Key words: Maize roots, Fe³⁺–reductase, ferricyanide reductase, iron     

Role(s) of nucleoli and phosphorylation of ribosomal protein S6 and/or HSP27 in the regulation of muscle mass

Effects of 14 days of hindlimb unloading or synergist ablation-related overloading with or without deafferentation on the fiber cross-sectional area, myonuclear number, size, and domain, the number of nucleoli in a single myonucleus, and the levels in the phosphorylation of the ribosomal protein S6 (S6) and 27-kDa heat shock protein (HSP27) were studied in rat soleus. Hypertrophy of fibers (+24%), associated with increased nucleolar number (from 1–2 to 3–5) within a myonucleus and myonuclear domain (+27%) compared with the preexperimental level, was induced by synergist ablation. Such phenomena were associated with increased levels of phosphorylated S6 (+84%) and HSP27 (+28%). Fiber atrophy (–52%), associated with decreased number (–31%) and domain size (–28%) of myonuclei and phosphorylation of S6 (–98%) and HSP27 (–63%), and with increased myonuclear size (+19%) and ubiquitination of myosin heavy chain (+33%, P > 0.05), was observed after unloading, which inhibited the mechanical load. Responses to deafferentation, which inhibited electromyogram level (–47%), were basically similar to those caused by hindlimb unloading, although the magnitudes were minor. The deafferentation-related responses were prevented and nucleolar number was even increased (+18%) by addition of synergist ablation, even though the integrated electromyogram level was still 30% less than controls. It is suggested that the load-dependent maintenance or upregulation of the nucleolar number and/or phosphorylation of S6 and HSP27 plays the important role(s) in the regulation of muscle mass. It was also indicated that such regulation was not necessarily associated with the neural activity. rat soleus muscle; functional overload; deafferentation; 27-kDa heat shock protein; ubiquitination of myosin heavy chain     

Determining the mass density of marine copepods and their eggs with a critical focus on some of the previously used methods

We collected Calanus finmarchicus copepodites CIV, CV and CVIfemales in a deep fjord on the west coast of Norway during April1996, May 1997 and November 1998. Eggs of C. finmarchicus andCalanus glacialis were collected during May 1989 in the BarentsSea. The sinking speeds of animals and eggs were measured ina homogeneous column with seawater of known density, and Stokeslaw was applied to estimate their mass density. Also the densitycontrast between the organisms and seawater was calculated.The mean mass density of C. finmarchicus ranged from 1.0274to 1.0452 g cm^–3. During spring copepodite stage CV hada significantly lower mean mass density (1.0345 g cm^–3)compared to CIV (1.0381 g cm^–3) and CVI females (1.0408g cm^–3). Copepods collected during winter had a distinctlylower mass density. The sinking speed of C. glacialis eggs followeda unimodal distribution, with a mean of 25.9 m day^–1,while sinking speeds of C. finmarchicus eggs were bimodal, thetwo groups of eggs having a mean sinking speed of 23.3 m day^–1and 35.4 m day^–1 respectively. Correspondingly the meanmass density was 1.0556 g cm^–3 for C. glacialis eggs andfor the two groups of C. finmarchicus eggs 1.0639 g cm^–3and 1.0812 g cm^–3. Results of earlier work, particularlyusing density gradient methods to determine mass density ofzooplankton, are critically reviewed, and it is suggested thatthis method should not be used to determine the mass densityof small organisms the size of C. finmarchicus.     

Chlorophyll standing crop and phytoplankton production in the western Irish Sea during 1992 and 1993

Chlorophyll standing crop and phytoplankton production werestudied in the western Irish Sea over a 21 month period during1992 and 1993. For both years, the start of the production seasonwas first observed in Dundalk Bay and occurred progressivelylater in more northerly coastal and offshore waters. Standingcrop and production exhibited marked spatial heterogeneity with12.5- to 19-fold differences in crop observed over distancesof 20–30 km. Distinct regional differences in the lengthof the production season were apparent. The longest season,6–7 months with a production of 194 g C m^–2, occurredin Dundalk Bay. The season lasted 3–4 months in the summerstratified region with a production of 140 g C m^–2. Northerly,offshore mixed waters and coastal waters of Northern Irelandsupported a short (2–3 months) season and production of194 and 140 g C m^–2, respectively. The similarity in seasonalproduction between Dundalk Bay and coastal waters of NorthernIreland, and between the summer stratified and northern mixedregions, is attributed to the intensity of production duringthe summer. Between 59 and 79% of seasonal production in thenorthern mixed region and coastal waters of Northern Irelandtook place during June and July, compared to 29–40% inDundalk Bay and the summer stratified region. Lower summer productionin the latter two may be due to nutrient limitation and thishas implications for the sensitivity of these two regions toanthropogenic nutrient enrichment.     

Repartition des Appendiculaires au large des cotes de Californie

Twenty-seven taxa of appendicularians have been identified fromthe cruise CALCOFI n° 7202 and their distribution has beenstudied by various numerical methods, with the purpose of definingsimultaneously the groups of associated species and the groupsof stations possessing similar characteristics. Two large recurrentgroups have been defined by the method of Fager and McGowan,corresponding approximatively to the major water masses present:central Pacific waters and equatorial Pacific waters. The methodof Williams and Lambert permits the separation of 11 southernstations and a northern group divided into 2 subgroups dependingon the presence or absence of O. dioica. The rank correlationsgive results very similar to those obtained with point correlations.Finally, principal component analysis allow good separationof the northern and southern zones by the factorial plan 1–3,while representation of the stations by the factorial plan 2–3separates the two northern subgroups depending on the presenceor absence of O. dioica. The results obtained by the differentmethods are therefore very much alike, but it is difficult toconclude they are a result of the specific cruise, or whetherthey represent a general phenomenon.     

Seasonal variations of bacterial abundance and biomass and their relation to phytoplankton in the hypertrophic tropical lagoon Cienaga Grande de Santa Marta, Colombia

The seasonal development of bacteria was studied in the hypertrophiccoastal lagoon Ciénaga Grande de Santa Marta (Caribbeancoast of Colombia). This large but only 1.5 m deep lagoon issubject to strong seasonal variations of salinity from almostfully marine (April/May) to brackish conditions in October/November.Chlorophyll ranged from 6 to 182 µg L^–1, and grossprimary production amounted to 1690 g C m^–2 per year.Total bacterial number (TBN) ranged from 6.5 to 90.5 x 10⁹ cellsL^–1 and bacterial biomass (BBM) from 77 to 1542 µgC L^–1, which are among the highest ever reported for naturalcoastal waters. Neither TBN nor BBM varied significantly withsalinity, phytoplankton or seston concentrations. Only the bacterialmean cell volume showed a significant relation to salinity,being highest (0.066 µm³) during the period of increasingand lowest (0.032 µm³) during decreasing salinity. Bacterialprotein accounted for 24% (19–26%) and phytoplankton proteinfor 57% (53–71%) of total seston protein. The ratio (annualmean) of bacterial carbon to phytoplankton carbon was 0.44 (range0.04–1.43). At low phytoplankton abundance [chlorophylla (Chl a) < 25 µg L^–1], bacterial carbon wasalmost equal to phytoplankton biomass (i.e. the mean ratio was1.04). In contrast, at Chl a > 100 µg L^–1, BBMwas low compared to phytoplankton biomass (the mean ratio was0.16). In general, BBM varied less than phytoplankton biomass.Most probably, the missing correlation between bacterial andphytoplankton variables was due to (i) organic material partlyderived from allochthonous sources serving as food resourcefor bacteria and (ii) a strong resuspension of bacteria fromthe sediment caused by frequent wind-induced mixing of the veryshallow lagoon.