Aufschlu� biologischer Matrices f�r die Bestimmung sehr niedriger Spurenelementgehalte bei begrenzter Einwaage mit Salpeters�ure unter Druck in einem Teflongef��

Neuroprotective effect of Moringa oleifera leaf extract on aluminium-induced temporal cortical degeneration

Aluminium (Al), one of the metals implicated in neurodegeneration easily gain access to the nervous system through its presence in many manufactured foods, medicines and drinking water, and causes neurotoxicity utilizing the reactive oxygen specie pathway. The need to curtail these effects on the nervous system motivated the use of the plant Moringa oleifera (MO). This study thus, investigated the neuroprotective effects of MO leaf extract on aluminium-induced temporal cortical degeneration in rats. 24 male albino Wistar rats were grouped (n = 6) into control (1 ml/kg distilled water), l00 mg/kg aluminium chloride (AlCl₃), 300 mg/kg MO, and 100 mg/kg AlCl₃ and 300 mg/kg MO groups. The administration lasted for 28 days and the rats were sacrificed on day 29 by perfusion-fixation after blood was obtained for serum Al estimation. The brain tissues were then routinely processed for some histological and immunnolabelling studies. There was no significant difference in serum Al in the test groups. Histological results showed atrophied and karyorrhetic cells with loss of Nissl substance in the temporal cortex of the AlCl₃ group, while no adverse effect was observed in the cytoarchitecture of the temporal cortex and Nissl substance of the MO group. However, groups which were administered AlCl₃ simultaneously with MO extract showed less degenerative features in the cyto-architecture of the temporal cortex with normal Nissl substance staining. There was increased neuron specific enolase (NSE) and glial fibrillary acidic protein (GFAP) expressions in the AlCl₃ group, while the MO group also showed increased NSE but decreased GFAP expression. However, the group which were administered AlCl₃ simultaneously with MO extract showed less expression of NSE and GFAP. In conclusion, MO protects against Al-induced neurotoxicity of the temporal cortex of rats.

Cadmium Accumulation in Periphyton from an Abandoned Mining District in the Buffalo National River, Arkansas

The Rush Mining District along the Buffalo River in Arkansas has a significant history of zinc and lead mining operations. The tails and spoils of these operations deposit heavy amounts of raw ore into streams. One element commonly found in the earth's crust that becomes a minor constituent of the deposition is cadmium. Periphyton samples from Rush Creek and Clabber Creek, two creeks within the Rush Mining District were measured for cadmium as well as two creeks with no history of mining, Spring Creek and Water Creek. Periphyton samples from Rush and Clabber Creek contained mean cadmium concentrations of 436.6 ± 67.3 and 93.38 ± 8.67 µg/kg, respectively. Spring Creek and Water Creek had a mean cadmium concentration of 40.49 ± 3.40 and 41.78 ± 3.99 µg/kg within periphyton. The results indicate increased metal concentrations in algal communities from mined areas. As periphyton is the base of the aquatic food chain, it acts as a conduit for movement of cadmium in the food web.

Qualitative and quantitative spectro-chemical analysis of dates using UV-pulsed laser induced breakdown spectroscopy and inductively coupled plasma mass spectrometry

Laser Induced Breakdown Spectroscopy (LIBS) is demonstrated for the spectral analysis of nutritional and toxic elements present in several varieties of date fruit samples available in the Saudi Arabia market. The method analyzes the optical emission of a test sample when subjected to pulsed laser ablation. In this demonstration, our primary focus is on calcium (Ca) and magnesium (Mg), as nutritional elements, and on chromium (Cr), as a toxic element. The local thermodynamic equilibrium (LTE) condition was confirmed prior to the elemental characterization of date samples to ensure accuracy of the LIBS analysis. This was achieved by measuring parameters associated with the plasma, such as the electron temperature and the electron number density. These plasma parameters aid interpretation of processes such as ionization, dissociation, and excitation occurring in the plasma plume formed by ablating the date palm sample. The minimum detection limit was established from calibration curves that involved plotting the LIBS signal intensity as a function of standard date samples with known concentrations. The concentration of Ca and Mg detected in different varieties of date samples was between 187 and 515 and 35-196mgL(-1) respectively, while Cr concentration measured between 1.72 and 7.76mgL(-1). In order to optimize our LIBS system, we have studied how the LIBS signal intensity depends on the incident laser energy and the delay time. In order to validate our LIBS analysis results, standard techniques such as inductively coupled plasma mass spectrometry (ICP-MS) were also applied on an identical (duplicate) date samples as those used for the LIBS analysis. The LIBS results exhibit remarkable agreement with those obtained from the ICP-MS analysis. In addition, the finger print wavelengths of other elements present in date samples were also identified and are reported here, which has not been previously reported, to the best of our knowledge.

Constructed wetlands as green tools for management of boron mine wastewater

Constructed wetlands are of increasing interest worldwide given that they represent an eco-technological solution to many environmental problems such as wastewater treatment. Turkey possesses approximately 70% of the world's total boron (B) reserves, and B contamination occurs in both natural and cultivated sites throughout Turkey, particularly in the north-west of the country. This study analyzes B removal and plant uptake of B in pilot plots of subsurface horizontal-flow constructed wetlands. Constructed wetlands were vegetated with Typha latifolia (referred to as CW1) and Phragmites australis (referred to as CW2) to treat wastewater from a borax reserve in Turkey--the largest of its type in the world and were assessed under field conditions. The B concentrations of water inflows to the systems were determined to be 10.2, 28.2, 84.6, 232.3, 716.4, and 2019.1 mg l(-1). The T. latifolia in the CW1 treatment group absorbed a total of 1300 mg kg(-1) B, whereas P. australis absorbed 839 mg kg(-1). As a result, CW1 had an average removal efficiency of 40.7%, while that of CW2 was 27.2%. Our results suggest that constructed wetlands are an effective, economic and eco-friendly solution to treating B mine wastewater and controlling the adverse environmental effects of B mining.

The phytoremediation ability of a polyculture constructed wetland to treat boron from mine effluent

This study focuses on describing the ability of a small-scale, subsurface-flow-polyculture-constructed wetland (PCW) to treat boron (B) mine effluent from the world's largest borax mine (Kırka, Turkey) under field conditions. This application is among the first effluent treatment methods of this type in both Turkey and the world. This study represents an important resource on how subsurface-flow-constructed wetlands could be used to treat B mine effluents in the field conditions. To this end, an experimental wetland was vegetated with common reed (Phragmites australis) and cattails (Typha latifolia), and mine effluent was moved through the wetland. The results of the present study show that B concentrations of the mine effluent decreased from 187 to 123 mg l(-1) (32% removal rate) on average. The T. latifolia individuals absorbed a total of 250 mg kg(-1) whereas P. australis in the PCW absorbed a total of 38 mg kg(-1) B during the research period.

Adsorption-desorption study of BSA conjugated silver nanoparticles (Ag/BSA NPs) on collagen immobilized substrates

There has been a growing interest in the use of protein conjugated nanoparticles for applications in biomedical, sensing, and advanced imaging. The objective of this study was to understand the interaction of protein conjugated silver nanoparticles (Ag/BSA NPs) with biological substrate (collagen layer). The adsorption behavior of synthesized Ag/BSA NPs on collagen immobilized silanized surface was followed by UV-vis spectroscopy by initially studying the formation of collagen layer and subsequent adsorption of Ag/BSA NPs to the immobilized layer. Surface plasmon resonance (SPR) data provided the real time profile of adsorption of Ag/BSA NPs from solution onto collagen immobilized and control substrates as well as desorption of nanoparticles from the substrates. The retention of NPs to substrate is sensitive to chemistry of the underlying substrate and on the external environment. UV-vis and atomic absorption spectrometric analysis of Ag/BSA NPs desorption performed under different pH conditions showed more NPs retained at physiological pH than the acidic and basic conditions. Nanoparticles retention on collagen immobilized substrate at physiological pH could influence properties of biological interest such as circulation lifetime and biodistribution of nanoparticles in the body.

Assessing the potential for cadmium phytoremediation with Calamagrostis epigejos: a pot experiment

A pot experiment was conducted for three vegetation periods on a sandy soil (pH 7.5) to study the uptake and distribution of Cd in plant tissues of Calamagrostis epigejos (L.) Roth. Cadmium was applied as CdCl2 (a total of 11 solution of 0, 20. 100, and 200 mg Cd l(-1)). HNO3- and water-extractable concentrations of Cd in 2- and 20-cm soil depths were correlated with the applied Cd showing that Cd was very mobile in the soil. The uptake of Cd from soil by Calamagrostis epigejos was directly related to the total soil Cd content and to the water-soluble pool of Cd. The concentrations of Cd in plant tissues (roots, rhizomes, leaves) and litter increased with increased applied Cd. Most of the Cd that was taken up was accumulated in roots (range from 1.88+/-0.42 to 40.96+/-16.71 mg kg(-1) dry mass), followed by rhizomes (0.52+/-0.13 to 25.70+/-6.35 mg kg(-1)) and leaves (0.30+/-0.06 to 9.20+/-1.93 mg kg(-1)). Cd concentrations of the litter were about twofold greater than the concentrations in the leaves (0.67+/-0.07 to 18.98+/-7.00 mg kg(-1)). The bioaccumulation factor (leaf/soil concentration ratio) increased significantly from 0.70+/-0.10 (control) to 1.1+/-0.17 (100 mg Cd l(-1)), but decreased again at the highest Cd level (200 mg Cd l(-1)) toward 0.74+/-0.34, which was not significantly different from the control. The low transfer of Cd from soil to above-ground organs at higher soil Cd concentrations indicates an exclusion mechanism. The leaf/root Cd concentration ratio (translocation factor) shows no significant relationship to increasing soil contamination. Only 4-7% of the total plant Cd was accumulated in the above-ground tissues. The phytoextraction potential (total Cd removed from soil) within three growing seasons ranged from 0.11 to 0.25% of the total soil Cd. Total output in above-ground living and dead plant material of C. epigejos would be approximately 20 g ha(-1) a(-1) for the lowest contamination level (+20 mg Cd per pot) and approximately 275 g ha(-1) a(-1) for the highest contamination level (+200 mg Cd per pot). This is within the range where an application for phytoextraction of Cd has been suggested by other authors. However, we conclude that the practical use of C. epigejos for phytoremediation is not mainly in the field of phytoextraction, but phytostabilization. C. epigejos has the capability to structurally stabilize the soil and reduce Cd contamination spread due to erosion. The uptake of the available Cd pool and accumulation in below-ground biomass may further prevent leaching into ground water.

Molybdenum metabolism: stable isotope studies in infancy

The essential trace element molybdenum (Mo) is bound to and required for the function of molybdoenzymes, e.g. sulfite and xanthine oxidase. Dietary recommendations for early infancy are based on limited knowledge about its metabolism. 100Mo was used as an extrinsic tag to study the absorption and kinetics of excretion in infancy. 10 infants with a gestational age of 35 (30-39) weeks, a birth weight of 2.0 (0.9-2.3) kg and a post-natal age of 20 (10-54) days were studied. They received 25 microg 100Mo/kg with a feed of human milk or formula. Fractional urinary and fecal collections were conducted preceding the 100Mo intake and for 48-72 hours afterwards. The materials were analyzed by atomic absorption spectroscopy and inductively coupled plasma mass spectrometry. The median absorption of 100Mo intake was 97.5 (96.3 to 99.1) %. The retention of nutritive Mo intake and 100Mo in the study period was 11.2 (3.8-15.7) microg Mo/kg, equivalent to 35.7 (12.7-55.6) %. The Mo concentration increased to a peak value in urine within 8 (6-13) hours and in feces within 24 (7-48.5) hours. In addition, increases of copper in feces and urine were observed in 8 of 9 infants studied. Mo given orally is well resorbed in premature infants, and predominantly excreted in the urine. Dietary recommendations should prevent excessive intakes in infancy.

Molybdenum supplementation in phenylketonuria diets: adequate in early infancy?

BACKGROUND

Molybdenum concentrations in formulas exceed those in human milk by far. Infants with phenylketonuria require semisynthetic phenylalanine-restricted diets. Because these diets are presently supplemented with molybdenum, a study was conducted to determine whether retention and plasma concentration in the recipients are equivalent to those of healthy breast-fed infants.

METHODS

Balance and plasma studies were conducted in healthy breast-fed infants (n = 17) and in patients with phenylketonuria (n = 4) at the age of 4 weeks, and the plasma investigations were repeated at the ages of 4 and 12 months. The samples were analyzed by atomic absorption spectroscopy (balance studies) and high-resolution inductively coupled plasma mass spectrometry (plasma).

RESULTS

Molybdenum intake and retention in all infants with phenylketonuria were more than 18 times those of breast-fed infants. The plasma concentrations reflected these differences. A median of 0.04 microg/l was assessed in breast-fed infants at 4 weeks and less than 0.02 microg/l at 4 months of age. Comparative results of infants with phenylketonuria were 2.9 microg/l and 2.5 microg/l, respectively. There were no significant differences between the groups at 12 months of age.

CONCLUSIONS

The phenylketonuria diets investigated showed excessive retention and plasma concentrations of the essential trace element molybdenum in early infancy. In view of these findings, the present practice of molybdenum fortification should be revised.

Effect of perinatal lead exposure on rat behaviour in open-field and two-way avoidance tasks

In view of conflicting results in literature concerning lead exposure associated with behavioural alterations, this study investigated behaviour in the open-field and shuttle avoidance, for as well as tissue lead burdens of pre- and postnatally lead-exposed rats. Rats were exposed to the metal from conception to weaning by giving the dams 0.5, 2.0 or 4.0 mM lead acetate in drinking water. This regimen did not affect body weight gain of dams or offspring development and had no effect on cerebral weights nor on haematological parameters of 23-day-old rats. In 1-day-old rats, lead accumulated in the blood but not in the brain, whereas both in 23-day-old rats and in dams lead accumulated in blood, kidney and cerebral cortex. In the open-field, lead-exposed groups showed higher locomotor activity in the test session as compared to controls and did not show any decrease in rearing responses in the test, indicating less habituation. Lead-treated rats subjected to a shuttle avoidance task showed no significant increase in avoidance responses between sessions as compared to control, indicating less retention. Moreover, only the control group presented a significant reduction of the footshock escape latency along testing session, suggesting a lead effect on footshock escape acquisition. In the shuttle box, intertrial crossing responses were not affected by lead treatment. The behavioural alterations occurred in animals with blood lead levels in the range 11-50.6 micrograms/dl.

Influences of gold on zinc, copper and metallothionein kinetics in liver and kidney of the rat

1. Single, very low dose of gold (500 micrograms/kg body weight) was given intramuscularly to male Sprague-Dawley rats (mean body weight: 200 g) as aurothiokeratinate (Auro-Detoxin(R) dissolved in 0.9% NaCl. 2. Blood, liver and kidney were samples at autopsy 0.5, 1, 2, 3, 6 and 12 h after the gold injection (six animals per time). 12 rats were treated with 0.9% saline only as controls. The zinc (Zn), copper (Cu), gold and metallothionein (MT) concentrations were determined in homogenates and cytosols of liver and kidney using atomic absorption spectrometry and gel chromatography, respectively. 3. The main changes in metal and MT concentrations occurred in the kidney, where an increase of gold was found 0.5 h after the injection, followed by an increase in Cu and MT concentrations after 6 h. Zn homeostasis did not change. 4. The results suggest that gold by itself induces an increase of MT-like peptides in the kidney cytosol, which was accompanied by an increase in the concentration of Cu mainly bound to these MT-like peptides.

Pancreatic contrast enhancement with Mn-bis pyridoxal ethylene diamine diacetic acid and the influence of a hormonal stimulation of the pancreas in pigs

RATIONALE AND OBJECTIVES

Pancreatic and hepatic contrast enhancement after different modes of administration of Mn-bis pyridoxal ethylene diamine diacetic acid (Mn-DPDP) and the influence of the hormones secretin and cholecystokinin on the uptake of this new contrast agent were studied by using magnetic resonance (MR) imaging and atomic absorption spectroscopy (AAS) examinations carried out on 13 young domestic pigs.

METHODS

Mn-DPDP was administered either as bolus injection or 30-min infusion. The course of contrast enhancement was noted. MR imaging was carried out with or without pancreatic stimulation. The Mn content in pancreatic and hepatic tissue was analyzed by using AAS.

RESULTS

Mn-DPDP caused a significant increase (66 +/- 40.1%) in pancreatic signal intensity, an increase in hepatic signal intensity (78 +/- 23.0%), and a significantly higher Mn content on tissue samples of both organs. The results showed a wide individual range. With infusion of Mn-DPDP, pancreatic enhancement was slower, with a peak at 82 +/- 34 min compared with 54 +/- 51 min following bolus injection. The Mn content of both hepatic and pancreatic tissue determined by AAS seemed to be higher after infusion. An incomplete washout of Mn after 24 hr was demonstrated. Hormonal stimulation of the pancreas further increased pancreatic signal intensity but did not influence hepatic contrast enhancement.

CONCLUSION

Infusion is the method of choice for administering Mn-DPDP because it causes an increase in hepatic signal intensity but does not change pancreatic enhancement. Although pancreatic hormonal stimulation intensifies enhancement, the increase is not statistically significant, and this practice will not contribute to day-to-day clinical practice.

Cadmium, lead, selenium, and zinc in semen of occupationally unexposed men

Concentrations of cadmium, lead, selenium, and zinc were determined in semen and seminal plasma of 22 volunteers by atomic absorption spectrometry. Additionally conventional semen parameters and, by means of computer videomicrography, motion parameters of spermatozoa were evaluated. Concentrations of Cd, Pb, and Zn determined in semen were not significantly different from those measured in seminal plasma. However, selenium levels were significantly higher in semen (53.8 +/- 22.9 micrograms l-1) than in seminal plasma (40.4 +/- 15.5 micrograms l-1, P < 0.01). The investigated semen samples on average contained low levels of Cd (0.4 +/- 0.23 micrograms l-1) and Pb (9.8 +/- 6.5 micrograms l-1). Studies on the intra-individual variability revealed the following average coefficients of variation (%) for element concentrations: Pb (70), Cd (53), Se (27), and Zn (23); and for semen parameters: total sperm count (46), sperm concentration (37), motility (22), ejaculate volume (21), linearity (19), linear velocity (11), curvilinear velocity (10), and percentage of normally formed sperm (9). Significant positive correlations were detected between semen selenium levels and sperm concentration (r = 0.51, P < 0.05), and percentage of normally formed sperm (r = 0.46, P < 0.05), respectively. Sperm motility (r = 0.53, P < 0.02), linear (r = 0.76, P < 0.001) and curvilinear velocity (r = 0.64, P < 0.002) were significantly correlated with semen cadmium levels.

Measurement of iron bioavailability by means of stable 54Fe and mass spectrometry

Stable isotopes represent a useful tool for the assessment of biokinetic parameters in in vivo studies on humans. A procedure is described to evaluate the bioavailability of iron in pharmaceutical preparations by means of 54Fe as a tracer and mass spectrometry for the determination of time dependent changes in the isotope ratio of 54Fe/56Fe in red blood cells. Iron tablets with an increased portion of 54Fe were administered to iron deficient subjects and red cell iron utilization was used as a measure of iron bioavailability. Iron utilization was derived from changes in the 54Fe/56Fe ratio as evaluated by means of fast atom bombardment-mass spectrometry (FAB-MS) on processed blood samples. A good intraindividual reproducibility was observed for blood samples drawn at various times after application of the trial drug. Figures for bioavailability and its interindividual variations were in the range expected from comparable studies on similar iron preparations using radioiron as tracers. The results obtained show that quantitative data of bioavailability from pharmaceutical iron preparations may be obtained without radiation exposure of the subjects investigated.

Longitudinal zinc balances in breast-fed and formula-fed infants

Longitudinal zinc (Zn) balance studies were performed under domestic conditions in term breast-fed (n = 10), term formula-fed (n = 5; Zn concentration in the formula: 4 mg/l) and preterm formula-fed (n = 3) infants during the first 17 weeks of life. Samples of milk, urine and feces were analyzed by atomic absorption spectrometry. The median daily Zn intake in breast-fed infants decreased from 0.592 (0.457-0.829) mg Zn/kg body weight to 0.151 (0.095-0.304) mg Zn/kg body weight in the first 17 weeks of life; comparable values for bottle-fed term infants were 0.58 (0.511-0.701) and 0.674 (0.529-0.731) mg Zn/kg body weight. The median percent retention of Zn intake was 27 (-60 to 81.4)% in breast-fed infants and 21.5 (-42 to 64)% in formula-fed infants. In view of the urinary and fecal Zn losses measured, a daily intake of 0.3-0.5 mg Zn/kg body weight is considered to be sufficient to ensure a Zn retention equivalent to breast-fed infants. This requires a Zn concentration of 2-3 mg/l of Zn depending on milk volume intake.

Assessment of iron availability using stable 54Fe

This paper describes a method of quantitatively assaying the bioavailability of orally administered iron in order to promote haemoglobin synthesis in iron deficiency anaemia. The non-radioactive tracer substance 54Fe was employed. An experimental iron deficiency model was tested in 18 healthy male volunteers. The trial design made it possible to assess intestinal absorption and efficacy of iron substitution. The iron deficiency was experimentally induced by weekly phlebotomy. Two commercially available iron preparations with different rates of iron release were investigated at a dosage of 150 and 160 mg Fe2+ daily, respectively. In the first seven days of treatment, both preparations were administered in 54Fe-labelled form. Afterwards, iron substitution was given with the commercially available preparations. Measurements were made of erythrocyte utilization of 54Fe and plasma iron tolerance curves at the beginning of the periods in which the 54Fe-labelled product and the commercially available preparation were administered, and of haemoglobin and serum ferritin concentration curves over three months. The mean utilization of the iron administered was virtually identical for the two preparations (23 and 22%, respectively). Likewise, there was no difference with respect to the average daily increase in haemoglobin concentration in the blood (1.5 g 1-1). There was also no significant difference with respect to serum ferritin concentration curves. In contrast, the two preparations differed markedly with respect to the plasma iron tolerance curves. This suggests that evaluation of plasma iron tolerance curves alone is not suitable for comparative assessment of the therapeutic value of orally administered iron preparations.

The distribution of cadmium within the human prostate

Five normal prostates from autopsies of humans aged 61-76 yr were divided into 2 x 24 topographically well-defined pairs of slices that were analyzed for cadmium (Cd) and examined histologically. The corresponding kidney cortex concentrations were also determined. The concentrations in ng Cd/g wet wt were in the range of 50-500 in the prostates and 8,000-39,000 in the kidneys with good mutual correlation. Large variations in Cd concentrations within the prostates were found. The concentrations were highest at the base (near the bladder) and lowest at the apex of the gland. Furthermore, large variations within horizontal layers were found, and this variation was not correlated to the histological amount of stroma or glands.

Trace element excess in PKU diets?

Knowledge of trace element requirements of infants with phenylketonuria (PKU) fed a semisynthetic diet is limited. Three infants with PKU detected early were studied longitudinally in classical balance studies for 72 h, under domestic conditions, at the ages of 2, 5, 8, 12 and 16 weeks. Iron, copper and manganese concentrations in the diet and faeces were determined by atomic absorption spectroscopy. The median concentrations in the diet (4.8 mg Fe/L, 1.7 mg Cu/L, 0.43 mg Mn/L) exceed those in human milk. This is mainly due to supplementation of the amino acid preparation used. The increased intake led to a significantly higher daily retention of Cu and Mn from the PKU-diet fed, with a median of 0.17 mg Cu/kg and 6.4 micrograms Mn/kg body weight; the median retention of Fe was 0.24 mg Fe/kg. Our results confirmed the doubts about the suitability of the present trace element supplementation in formula for infants with PKU during the first four months of life.

Longitudinal manganese and copper balances in young infants and preterm infants fed on breast-milk and adapted cow's milk formulas

1. Mn and Cu intake and retention in twenty full-term infants and six preterm infants were studied on the basis of 72 h balances. The age of the infants was 2-16 weeks and the gestational age of the preterm infants (triplets) 34 and 36 weeks. Three nutrition schemes were pursued: breast-fed, formula-fed with unsupplemented adapted formula and formula-fed with trace element supplementation. 2. The mean Mn concentration of all breast-milk samples (n 2339) was 6.2 micrograms/l. The two formulas had similar Mn concentrations (77 and 99 micrograms/l) but had different Fe, Cu (121 and 619 micrograms/l), Zn and I contents. The mean Cu concentration in mother's milk was 833 micrograms/l. 3. The following mean daily Mn intakes and retentions (micrograms/kg) respectively were measured: breast-fed full-term 1.06 (SD 0.43) and 0.43 (SD 0.65), formula-fed full-term 14.2 (SD 3.1) and 2.8 (SD 4.8), formula-fed preterm 15.0 (SD 2.2) and 0.06 (SD 5.87). The results for Cu were 114.5 (SD 22.3) and 88.0 (SD 46.5) micrograms/kg in breast-fed, 19.8 (SD 4.2) and 4.6 (-11.5-9.6) in the unsupplemented formula-fed and 106.4 (SD 18.9) and 55.5 (SD 20.3) in the supplemented formula full-term infant group. No significant influence of the trace element contents of the formulas on the relative retention of Mn or Cu was found. 4. Young preterm infants, and to some degree young full-term infants, often had negative Mn balances caused by a high faecal excretion. The formulas with a Mn concentration below 100 micrograms/l gave a sufficient supply of Mn. Preterm infants fed on the unsupplemented formula had a marginal Cu supply and their first balances were negative (-3.8 (SD 1.8) micrograms/kg). 5. In accordance with the estimated safe and adequate daily dietary intakes (recommended dietary allowances), formula-fed infants receive much more Mn than breast-fed infants and their absolute retention is higher. 6. Cu from breast-milk had a significantly better biological availability than that from cow's milk formula. If retentions similar to those in breast-fed infants are intended, we conclude, therefore, that cow's milk formula should be fortified with Cu up to a level of at least 600 micrograms/l.