The uptake of Manganese-54 by green algae (Protococcoidal chlorella), Daphnia magna,and fathead minnows (Pimephales promelas)

Concentration factors (CF) of ⁵⁴Mn for three aquatic species: green algae (Protococcoidal chlorella), Daphnia magna, and fathead minnows (Pimephales promelas) were determined following direct exposure to the isotope in solution. The maximum accumulation (CF = 911) in P. chlorella was reached at 48 hours of exposure; the maximum uptake (CF = 65) in Daphnia was reached at 8 hours of exposure; and the maximum accumulation (CF = 22.6) in fathead minnows was at 128 hours of exposure. The data indicate that ⁵⁴Mn accumulation decreases with ascent up a theoretical aquatic food chain when water is the only source of contamination.     

Effects of simulated acid precipitation on the cycling of manganese under Sitka spruce (Picea sitchensis)

Simulated acid rain at pH 3.3, 4.3 and 5.3 has been applied by overheadspraying to reconstructed soil profiles and young Sitka spruce (Piceasitchensis) trees in pots to assess the effect of rainfall acidityonthe fate and recycling efficiency of Mn from ⁵⁴Mn-labelled Sitkaspruce litter. The concentration of ⁵⁴Mn in throughfall wassignificantly increased by acidification of the rain, but ⁵⁴Mnactivity in drainage water remained low and was not significantly affected. Thefact that only < 1.5% of ⁵⁴Mn was lost in drainage water suggeststhat forest ecosystems efficiently retain Mn, at least in the short term. Thebulk of ⁵⁴Mn (approximately 60 to 70%) was retained in the litterlayer. Compared with rain at pH 4.3, rain at pH 3.3 acidified L/F and H/Ahorizon soils. The amount of ⁵⁴Mn found in the different soilhorizons was only influenced significantly by rainfall acidity in the E horizonsoil, however, where ⁵⁴Mn activity was increased by the most acidtreatment. The activity of ⁵⁴Mn in Sitka spruce needles did notdiffer significantly with treatments.     

The Shrimp Caridina nilotica in Lake Victoria (East Africa), Before and After the Nile Perch Increase

The shrimp Caridina nilotica is a major prey of the introduced Nile perch in Lake Victoria. In spite of heavy predation, the density of shrimps increased after the Nile perch boom and the concomitant disappearance of the haplochromine cichlids. In the same period, the mean size of gravid shrimps and the size at first maturity declined. This seems to indicate an increased predation pressure on adult shrimps. Before the Nile perch upsurge, specialised shrimp eaters and piscivores, among the haplochromine cichlids, only took adult shrimps, whereas we assume that most haplochromines used to include juvenile shrimps into their diet. Another important predator on adult shrimps was Bagrus docmak. The combined density of predators on adult shrimps in the pre-Nile perch era was estimated at 10 kg ha⁻¹ and the potential predators on juveniles were estimated at 170 kg ha⁻¹. After the Nile perch upsurge, only Nile perch up to 10 cm TL and Rastrineobola argentea fed on juvenile shrimps (ca. 36 kg ha⁻¹) and Nile perch from 10 to 50 cm TL (ca. 13 kg ha⁻¹) fed on adults. These rough estimates of the biomass of predators on shrimps before and after the Nile perch upsurge indicate a reduced predation pressure on juvenile shrimps. The disappearance of the haplochromines may have released competition with small Nile perch for juvenile shrimps, thus enhancing the recruitment of Nile perch.     

<Emphasis Type="Italic">Aeromonas salmonicida</Emphasis> infected fish transfer disease to healthy fish <Emphasis Type="Italic">via</Emphasis> water

Experimental studies of infection transmission via water from infected to healthy fish were conducted. The dark-brown bacterial colonies typical for Aeromonas salmonicida on tryptone soya agar (TSA) have been isolated and counted (from 3.0±0.6×10² to 3.5±0.5×10⁵ c.f.u. g⁻¹) from the internal organs of naturally infected (NI) and experimentally infected (EI) perch and sea trout. No significant differences in dark-brown bacterial counts were detected between EI perch and EI sea trout. The assessment and comparison of the alterations of the biological parameters of EI European perch and sea trout with bacterium Aeromonas salmonicida subsp. salmonicida with naturally infected perch were conducted. No mortality was recorded in groups of EI perch and sea trout. Whereas, the mortality of NI perch (collected from the main sites of outbreak of disease) was observed from the second day of the experiments. Changes in morphophysiological parameters of EI perch and sea trout were similar. Different alterations in blood cell parameters of EI fish were observed, and the most noticeable was the decrease (P≤0.01) in white blood cell count (WBC) of EI perch and sea trout. Based on these results it can be deduced that there is infection transmission of bacterium A. salmonicida from European perch via water to other fish species.     

Bioaccumulation of mercury in yellow perch (Perca flavescens) and common loons (Gavia immer) in relation to lake chemistry in Atlantic Canada

Mercury biomagnifies in aquatic foodwebs in freshwater lakes, and common loons (Gavia immer) breeding in eastern Canada can be exposed to reproductively toxic concentrations of mercury in their fish prey. We assessed the bioaccumulation and biomagnification of mercury in juvenile and adult common loons, and their preferred prey: yellow perch (Perca flavescens) in Kejimkujik National Park (KNP), Nova Scotia by measuring mercury levels and stable isotope ratios in tissues. Total mercury levels and stable-carbon (δ¹³C) and nitrogen isotope ratios (δ¹⁵N) were determined in composite whole-fish samples from lakes in KNP and blood samples from juvenile and adult loons captured on lakes in KNP and southern New Brunswick. Geometric mean mercury concentrations were 0.15 and 0.38 μg/g (wet wt.) in small (9-cm fork length) and large (17-cm fork length) yellow perch, and were 0.43 and 2.7 μg/g (wet wt.) in blood of juvenile and adult common loons, respectively. Mercury concentrations in perch and loons were positively associated with body mass and δ¹⁵N values. Juvenile loons and large yellow perch had similar mercury levels and δ¹⁵N values, indicating similar trophic status despite their 22-fold difference in body mass. Mercury concentrations were higher in yellow perch and common loons in acidic lakes. Our findings highlight the importance of both chemical and ecological factors in understanding mercury biomagnification in lakes and associated risks to fish-eating wildlife. Electronic supplementary material Electronic supplementary material is available for this article at and accessible for authorised users.     

Microdynamics and seasonal changes in manganese oxide epiprecipitation in Pinal Creek, Arizona

In Pinal Creek, Arizona, Mn oxyhydroxides (MnO_x) collect as thick precipitates on surface sediment, within the streambed, beneath algal mats, and on submerged and emergent plants and mosses. The proximate source of Mn is a thick, alluvial alkaline aquifer that was contaminated by past acid mine waste disposal practices associated with copper mines located upstream in the Globe–Miami area. Almost every organism in Pinal Creek is coated with MnO_x. Some are actively precipitating manganese, and others are doing it passively. The variety and seasonality of epilithic biological processes resulting in Mn oxidation (epiprecipitation) was studied for more than a year by analyzing artificial substrates placed in surface water having different flows and different vegetation types and densities. Most epiprecipitation took place on the holdfasts of the green alga, Ulothrix sp., and the iron bacterium, Leptothrix discophora. Extensive patches of MnO_x also coated extracellular polymeric substances of fungal hyphae and bacterial filaments. The dominant macroscopic precipitation was in the form of MnO_x clumps on mosses, green algae, and cyanobacterial mats, consistent with precipitation by pH elevation during photosynthesis. Most oxidation occurred in the spring and summer, in agreement with thermal, biological, and chemical activity models. More biological oxidation occurred in swifter water, consistent with oxygen elevation models. The efficiency of this naturally occurring, diverse ecosystem suggests that remediation efforts to remove metal contaminants such as Mn should focus on creation of habitats that raise biodiversity.US Geological Survey, retired     

Use of Biochronology to Examine Interactions of Freshwater Drum, Walleye and Yellow Perch in the Red Lakes of Minnesota

Growth histories for freshwater drum, Aplodinotus grunniens, walleye, Stizostedion vitreum, and yellow perch, Perca flavescens, were constructed using calcified structures for the period 1947 through 1996 for the Red Lakes, Minnesota. Increased walleye growth and decreased yellow perch growth were observed over the period from 1983 to 1996, which are attributed to intensive fishing resulting in decreased intraspecific competition in walleye and increased intraspecific competition in yellow perch through release from predation. Strong year-classes of yellow perch were positively correlated with walleye growth (r=0.57, p-value=0.042). There was no evidence for interactions of walleye or yellow perch with freshwater drum. Freshwater drum growth (r=0.680, p-value=0.0001) was more highly correlated with temperature than were walleye (r=0.386, p-value=0.006) and yellow perch growth (r=0.303, p-value=0.036).     

Food and habitat partitioning among juveniles of three fish species in the pelagial of a mesotrophic lake

An analysis of the diet of 0⁺ perch, smelt and roach caught at night of the pelagial of a mesotrophic lake showed that their food was composed mainly of herbivorous and predatory cladocerans, copepods and Chaoborus larvae during summer, and of herbivorous cladocerans and copepods during October. An analysis of habitat use by juveniles revealed separation among the species: roach occurred in the upper, perch and smelt in the deeper water layers. Food and habitat were alternatively partitioned among the juveniles. In early summer the lowest food overlap between perch and smelt coincided with high habitat overlap. In October high food overlap between perch and smelt corresponded with low habitat overlap. Smelt and roach fry used common food resources throughout the season, but they were segregated in habitat.     

Three new interstitial dorvilleids (Annelida: Polychaeta) from the Cymodocea nodosa meadows of the Canary Islands

Resource–consumer relationships in Lake Victoria were investigated by use of stable isotope data. ¹³C and ¹⁵N signatures were determined for organisms at a deep (22 m) and a littoral (5 m) site in the Napoleon Gulf near Jinja, Uganda. Results suggest that two food chains operate at the deep site, one leading from a shrimp (Caridina nilotica) to juvenile Nile perch (Lates niloticus), the second leading from zooplankton (copepods and cladocerans) to a cyprinid (Rastrineobola argentea) and lake flies (Chaoborus). Isotopic evidence suggests that shrimp eat suspended particulates and benthos, not crustacean zooplankton or water hyacinth (Eichhornia crassipes). Resource–consumer relationships revealed in this study have implications for understanding future yields of the economically important Nile perch fishery.     

Book reviews

The numbers of perch Perca fluviatilis in the Cikola backwater system of the river Danube were estimated by mark-recapture techniques using multiple fishings. Fish were caught by fyke nets and electrical fishing. The latter method was unselective for sex and the catch data could be used to correct the sex bias in the fyke net catches. A population estimate in April of 6059 ind ha^–1 (95% C.L. 5135–7386) for perch > 5 cm was obtained for a 0.6 ha area of backwater and 1665 ind ha^–1 (95% C.L. 1204–2692) in May for a 0.4 ha area of a small bay.     

Cloning and characterization of endosymbiotic algae isolated fromParamecium bursaria

Summary The green parameciumParamecium bursaria has many endosymbiotic algae in its cytoplasm. Here, we cloned and characterized endosymbiotic algae fromP. bursaria and examined in detail the interaction between the cloned algae and algae-free paramecia. Homogenates ofP. bursaria were cultured on agar plates containing various kinds of media to establish clones of the endosymbiotic algae. Many algal colonies were obtained from poorly nutritious medium (CA medium) after one month in culture. Algae were picked up from these colonies and inoculations were repeated 9 times on agar plates containing CA medium. On enriched media including bacto-peptone, glucose, proteose-peptone and/or yeast extract, however, bacteria and mold grew rapidly and no algal colonies were formed. When the cloned algae were cultured in liquid CA medium, they grew faster than on agar plates and the numbers stayed constant at 1 × 10⁷ algae/ml after 7 days in culture. They revealed high infectivity to algae-free paramecia, and an incubation period of 24 h and at least 1 × 10³ algae/paramecium were required to achieve successful infection (80–90%). The growth and infection rate did not change through 74 repeated inoculations of algae in liquid CA medium. Optical microscopic observations revealed marked morphological similarity between endosymbiotic algae and free-livingChlorella, but the latter showed no infectivity to algae-free paramecia. The cloned endosymbiotic algae presented here will provide an excellent opportunity to examine the mechanism of symbiont-host interaction.     

Gastric evacuation rates of white perch,Morone americana,determined from laboratory and field data

Synopsis Gastric evacuation rates (R) of white perch,Morone americana, were determined in laboratory experiments and by using field data. The resulting relationship ofR and temperature (T) for white perch wasR = 0.028e^0.106T (r² = 0.98). The high r² of the regression indicates good agreement of the combined laboratory and field data. Our rate compares well with those available for other species; and especially for Eurasian perch, which has a similar thermal existence to white perch in Lake Erie.     

Use of Poisons in Determination of Microbial Manganese Binding Rates in Seawater

A method was developed to determine whether microorganisms mediate the precipitation of manganese(II) in the marine environment. Radioactive ⁵⁴Mn(II) was used as a tracer to measure the precipitation (binding and oxidation) of Mn(II) [i.e., the ⁵⁴Mn(II) trapped on 0.2-μm membrane filters] in the presence and absence of biological poisons. A variety of antibiotics, fixatives, and metabolic inhibitors were tested in laboratory control experiments to select poisons that did not interfere in the chemistry of manganese. The poisons were deemed suitable if (i) they did not complex Mn(II) more strongly than the ion-exchange resin Chelex 100, (ii) they did not interfere in the adsorption of ⁵⁴Mn(II) onto synthetic δMnO₂ (manganate), (iii) they did not cause desorption of ⁵⁴Mn(II) which had been preadsorbed onto synthetic manganate, and (iv) they did not solubilize synthetic ⁵⁴manganate. In addition, several known chelators, reducing agents, and buffers normally added to microbiological growth media or used in biochemical assays were tested. Most additions interfered to some extent with manganese chemistry. However, at least one inhibitor, sodium azide, or a mixture of sodium azide, penicillin, and tetracycline was shown to be appropriate for use in field studies of ⁵⁴Mn(II) binding. Formaldehyde could also be used in short incubations (1 to 3 h) but was not suitable for longer time course studies. The method was applied to studies of Mn(II) precipitation in Saanich Inlet, British Columbia, Canada. Bacteria were shown to significantly enhance the rate of Mn(II) removal from solution in the manganese-rich particulate layer which occurs just above the oxygen-hydrogen sulfide interface in the water column.     

Some life history responses of the cladoceran Ceriodaphnia cf. dubia to variations in population density at two different food concentrations

Atrazine is the most commonly used agricultural herbicide in the midwestern cornbelt of the U.S.A. This project investigated atrazine bioconcentration in Cladophora glomerata (L.) Kütz, the dominant algae by biomass in nutrient enriched rivers. C. glomerata concentrates many organic and inorganic materials from solution, and by considering the partitioning coefficient of atrazine (Log K _ow=2.56), the uptake of atrazine by this green algae was expected. The bioconcentration factor of C. glomerata ranged from 29 to 5223 over a one year sampling period, and bioconcentration was found to be a function of maximum biomass increase (due to high nutrients) and discharge. Uptake increased at low discharges as a result of increased contact time between cells and water. We found that C. glomerata was able to bioconcentrate a yearly average of 54% of the available atrazine from the water. Efficiency of uptake increased during the summer growing season of the algae (81%) and was 34% during the winter months. This field study demonstrates the ability of C. glomerata to bioconcentrate atrazine to a high degree and suggests an alternative and quicker biological removal of atrazine from agriculturally polluted rivers.     

Quantifying the activity and movement of perch in a temperate lake by integrating acoustic telemetry and a geographic information system

The horizontal movement patterns of perch (Perca fluviatilis L.) were studied by means of an automatic positioning system in a small Mediterranean temperate lake in northeastern Spain. Seven perch (247–265 mm in fork length, 243–271 g in weight) were implanted surgically with ultrasonic transmitters (38 mm long, 3.5 g in water) and their positions monitored approximately every 180 s using a radio-linked acoustic positioning (VRAP) system. Fish positions were integrated into a geographic information system database with information on date, time, slope and bottom depth. As a measure of the intensity of movement, the swimming speed was calculated. The Minimum Convex Polygon area was obtained as a measure of the area of activity. The data dispersion was calculated as r ² (the mean squared distance from the center of activity). The tagged perch showed individual variations in activity levels, but all had a circadian pattern, being less active at night (average speed 6.0 cm s^–1) than during the day (average speed 10.4 cm s^–1). They exhibited two peaks of activity, at sunrise and sunset. Two patterns of movement emerged, the first characterised by movement between the limnetic and the littoral zones, the second restricted to the shoreline. Fish appeared to follow the same path, reducing the area of activity to a limited part of the lake. Also, fish returned to the same place after movements, showing site fidelity but only for a few days. Overall, although perch could be very abundant in open waters (limnetic zone), they demonstrated a high dependence on the littoral zone.     

Influence of culture temperature on the growth,biochemical composition and fatty acid profiles of six Antarctic microalgae

The growth, biochemical composition and fatty acid profiles of six Antarctic microalgae cultured at different temperatures, ranging from 4, 6, 9, 14, 20 to 30 ^∘C, were compared. The algae were isolated from seawater, freshwater, soil and snow samples collected during our recent expeditions to Casey, Antarctica, and are currently deposited in the University of Malaya Algae Culture Collection (UMACC). The algae chosen for the study were Chlamydomonas UMACC 229, Chlorella UMACC 234, Chlorella UMACC 237, Klebsormidium UMACC 227, Navicula UMACC 231 and Stichococcus UMACC 238. All the isolates could grow at temperatures up to 20 ^∘C; three isolates, namely Navicula UMACC 231 and the two Chlorella isolates (UMACC 234 and UMACC 237) grew even at 30 ^∘C. Both Chlorella UMACC 234 and Stichococcus UMACC 238 had broad optimal temperatures for growth, ranging from 6 to 20 ^∘C (μ = 0.19 – 0.22 day^–1) and 4 to 14 ^∘C (μ = 0.13 – 0.16 day^–1), respectively. In contrast, optimal growth temperatures for NaviculaUMACC 231 and Chlamydomonas UMACC 229 were 4 ^∘C (μ = 0.34 day^–1) and 6–9 ^∘C (μ = 0.39 – 0.40 day^–1), respectively. The protein content of the Antarctic algae was markedly affected by culture temperature. All except Navicula UMACC 231 and Stichococcus UMACC 238 contained higher amount of proteins when grown at low temperatures (6–9 ^∘C). The percentage of PUFA, especially 20:5 in Navicula UMACC 231 decreased with increasing culture temperature. However, the percentages of unsaturated fatty acids did not show consistent trend with culture temperature for the other algae studied.     

The toxicity of phenolic compounds to soil algal population and toChlorella vulgaris andNostoc linckia

Summary p-Nitrophenol (PNP),m-nitrophenol (MNP), 2,4-dinitrophenol (DNP) and catechol were tested for their effects on algal population in a soil and on pure cultures of two algae isolated from soil. Both PNP and MNP, even at 0.5 kg ha⁻¹ level were toxic to the soil algae; high doses effected increase in toxicity. Inhibition of algae was relatively more with PNP compared to the other two nitrophenols. Catechol treatment up to 1.0 kg ha⁻¹ led to a significant initial enhancement of algae with a subsequent far less toxic effect. The toxicity of the phenolic compounds towardChlorella vulgaris, a green alga andNostoc linckia, a blue-green alga, decreased in the order: MNP≧PNP>DNP>Catechol. However, algicidal or algistatic effect of the test chemicals was fairly more againstC. vulgaris, suggesting that the eukaryotic alga is highly sensitive to such soil pollutants compared to the prokaryotic alga.     

Manganese dynamics in the rhizosphere and Mn uptake by different crops evaluated by a mechanistic model

Understanding of the mechanisms of Mn supply from the soil and uptake by the plants can be improved by using simulation models that are based on basic principles. For this, a pot culture experiment was conducted with a sandy clay loam soil to measure Mn uptake by summer wheat (Triticum aestivum L. cv. Planet), maize (Zea mays L. cv. Pirat) and sugar beet (Beta vulgaris L. cv. Orbis) and to simulate Mn dynamics in the rhizosphere by means of a mechanistic model. Seeds of three crops were sown in pots containing 2.9 kg soil in a controlled growth chamber. Root and shoot weight, Mn content of plants, root length and root radius were determined 8 (13 days in case of sugar beet) and 20 days after germination. Soil and plant parameters were determined to run nutrient uptake model calculations. Manganese content of the shoot varied from 25 mg kg^-1 for sugar beet to 34 mg kg^-1 for maize. Sugar beet had the lowest root length/shoot weight ratio but the highest relative shoot growth rate, resulting in the highest shoot demand on the root. This is reflected by the Mn influx which was 0.9 × 10^-7, 1.7 × 10^-7 and 2.5 × 10^-7 nmol cm^-1 s^-1 for wheat, maize and sugar beet, respectively. Nutrient uptake model calculations predicted similar influx values. Initial Mn concentration of 0.2 μM in the soil solution decreased to only 0.16 μM for wheat, 0.13 μM for maize and 0.11 μM for sugar beet at the root surface. This shows that manganese transport to the root was not a limiting step. This was confirmed by the fact that an assumed 20 times increase in maximum influx (I_max) increased the calculated Mn influx by 3.7 times. Sensitivity analysis demonstrated that for controlling Mn uptake the initial soil solution concentration (C _Li), the root radius (r₀), I_max and the Michaelis constant (K _m) were the most sensitive factors in the listed order. This revised version was published online in August 2006 with corrections to the Cover Date.     

Use of enclosures to detect the contribution of particular zooplankton to growth of young-of-the-year yellow perch (Perca flavescens Mitchell)

Summary Densities of the cladoceran, Holopedium gibberum, were manipulated in 18 enclosures containing juvenile (age 0+) yellow perch (Perca flavescens) and mean-lake densities of other zooplankton. In enclosures, where nearlake densities of all zooplankton species including Holopedium were maintained, young-of-the-year perch grew significantly heavier and longer than in experimental enclosures where Holopedium was excluded. Holopedium comprised between 15–45% of the diet (wet weight) of perch in the first 2 weeks of July in the control treatment (Holopedium at or near ambient lake densities) and only 3–7% of total biomass ingested in the experimental treatment (Holopedium density selectively reduced). Predation on Holopedium decreased dramatically after the 2nd week of July in the control treatment after which Chaoborus, chironomids, and Sida became dominant prey items (by weight) of juvenile perch. These findings suggest that growth and survivorship of age 0+ perch in Precambrian Shield lakes may be coupled to Holopedium abundance. Thus, utilization of Holopedium by young-of-the-year yellow perch may affect recruitment of this species since overwintering survivorship, range of accessible prey sizes or species, and vulnerability of juvenile perch to predation by larger fish depend on body size, which is reduced when Holopedium is excluded from the diet.     

Seasonal changes in composition and biomass of epiphytic algae on the macrophyte Ranunculus penicillatus in a chalk stream,with estimates of production,and observations on the epiphytes of Cladophora glomerata

The epiphytic algae on surfaces of the macrophyte Ranunculus penicillatus (Dumort.) Bab. var. calcareus (R. W. Butcher) C. D. K. Cook and on Cladophora glomerata (L.) Kütz growing in the River Itchen at Otterbourne near Southampton were studied between February 1984 and June 1985. The river at this site has a mean flow rate of 0.33 m s^–1, and is about 16 m wide and on average 20 cm deep, with a discharge ranging through the year between 0.34 and 2.46 m³ s^–1. The pH of the river varies little around 8.2, with a mean alkalinity of 236 mg HCO _inf3 ^sup1 l^–1, because of its origin from chalk springs. Ranunculus grows throughout the year, with peaks of biomass in spring and autumn. It forms a very large surface for attachment of epiphytes, and covers on average 40% of the stream bed. Numbers of epiphyte cells removed from Ranunculus ranged through the year between 52 × 10³ and 271 × 10³ cells mm^–2 stream floor, with maximum numbers in April, and a secondary peak in October. This pattern partly reflects fluctuations in the biomass of Ranunculus; the number of cells per unit area of plant surface showed a broader spring peak and lower fluctuations in other seasons. Diatoms formed 65 to 98% of these epiphyte cells, with chlorophytes reaching their peak (10%) in summer and cyanophytes (25%) in autumn. Estimates of biomass of these epiphytic forms, derived from measurements of chlorophyll c, indicate a range between 30 and 100 g dry weight m^–2 of weed bed. Colonisation studies showed that the algae grow and reproduce throughout the year, with a mean generation time of about 5 days, suggesting an annual production of about 3 kg dry weight m^–2 of weed bed, which makes epiphytic algae the principal primary producers in the stream. The numbers and biomass of epiphytic algae on Cladophora are considerably less.The species of epiphytic algae found on Ranunculus were generally similar to those growing on Cladophora, and to epilithic algae on pebbles of the stream bed, but different species were dominant on the different substrata. Algal cells in the water column were all derived from benthic habitats, although their relative abundance was very different.