Printer Friendly

Relationships between extractable Ni, Co, and other metals and some microbiological characteristics of different ultramafic soils from New Caledonia.


New Caledonian ultramafic soils are derived from metalliferous rocks containing, in particular, nickel, magnesium, iron, manganese, cobalt, and chromium (Brooks 1987). These soils are also characterised by a low content of different elements, such as nitrogen, calcium, and phosphorus. The high concentration of metal ions, by creating a phytotoxic medium, has resulted in an adapted vegetation with a high degree of endemism (Jaffre 1980) and a microflora in which actinomycetes (bacteria) and Moniliaceae (fungi) predominate (Amir and Pineau 1998a).

The microbial ecology of these extreme soils is very poorly studied. This is especially true for the relationships between metals and soil microbiological properties (Schlegel et al. 1991; Quantin et al. 2001). In general, the metals can be released by 3 major mechanisms. The first is a chemical process, which involves soil acidity (Brooks 1987; Gasser et al. 1995; Werner 1996). The second mechanism is microbial attack of metalliferous rocks (Lebedeva et al. 1978; Berthelin et al. 1995). The third mechanism involves the action of plants through their secretions in the rhizosphere (Brooks 1987; Ghiorse 1988).

This study aimed to investigate the relationships between the microflora and the release of certain metals (measured by their extractability), especially Ni and Co which are not known to be released by microbiological processes in soil. Metal toxicity and fungal tolerance to metals have been studied for a few soils from New Caledonia (Amir and Pineau 1998b), but these limited results needed to be further investigated and generalised, especially for Ni. We therefore analysed 40 ultramafic soil samples, focusing on the relationships between content of extractable metals, pH, and some biological characteristics such as total microbial activity, microbial density, and percentage of microorganisms tolerant to Ni. A laboratory experiment was performed to verify some statistical conclusions.

Material and methods


Forty samples of ultramafic soils were collected from 4 different regions in the south of New Caledonia. Thirty-eight were sampled from scrublands with serpentine vegetation: 12 were from hypermagnesian brown soils (HBS) from Plum; 12 were from ferralitic oxidic soils (FOS) or lateritic topsoils from Ouenarou; and 14 were from colluvial ferralitic oxidic soils (CFOS) from 'Plaine des Lacs'. Two samples were from the Tontouta mine in an exploited area without topsoil, a lateritic subsoil and a saprolite. In each group of soils, the distance between 2 sampling points was 20-30 m. Soil samples were collected from the top 20-cm homogeneous soil layer, which was mixed before sampling, some on bare patches and others under plants, in contact with roots.

For the laboratory incubation experiment, two different soils were used. The first was a saprolite taken from Tontouta mine. A few clusters of Costularia sp. (Cyperaceae) had colonised some places. The saprolite was collected between costularia tufts, where the biological metal release process was assumed to be recent and active. The second was an HBS, from Plum, taken in a bare patch.

All samples were sieved through 2-mm mesh before use. The physico-chemical characteristics were determined by the chemistry laboratory of 'Institut de Recherche pour le Developpement' (IRD) of Noumea, using conventional methods. The DTPA-extractable concentrations of metals were determined by atomic absorption spectrometry (AAS). Some of these characteristics are given in Table 1 for the main types of soils studied.

Soil bacterial and fungal community count

Determination of the microbial densities was performed by the dilution-plate count technique. Only organotrophic and aerobic microflora were studied. Soil (10 g) was suspended in water (90 mL) and shaken vigorously for 15 min. Appropriate series of 10-fold dilutions were prepared and aliquots of 1 mL were placed in Petri dishes before the medium was added.

For bacteria a peptone/yeast extract medium was used (peptone 5 g, yeast extract 3 g, dextrose 5 g, agar 20 g). The pH was adjusted to 7. Filter-sterilised actidione (50 mg/L) was added after sterilisation of the medium by autoclaving. For fungi, the composition of the medium was malt extract (10 g), dextrose (5 g), agar (20 g), and demineralised water (1000 mL); 100 mg/L of filter-sterilised streptomycin was added after autoclaving.

Colonies were counted after 4-6 days of incubation at 25[degrees]C. Three replicates were prepared for each dilution and for each microbial group.

Soil microbial activity as measured by FDA hydrolysis

The FDA (fluorescein diacetate) enzymatic method is frequently used for comparative studies of total soil activity (Schnurer and Rosswall 1982; Zelles et al. 1987). The FDA test was performed as follows. A well-mixed soil sample (1 g) was added to phosphate buffer (10 mL), pH 7.6 (K[H.sub.2]P[O.sub.4] 0.1 N; [Na.sub.2]HP[O.sub.4] 0.1 N) and homogenised by strong stirring (vortex) for 1 min before incubation for 20 min at 23[degrees]C. To the soil suspension was added 1 mL FDA (fluorescein diacetate 200 mg, acetone 60 mL, distilled water up to 100 mL). The tubes were incubated again for 40 min at 23[degrees]C in the dark. The reaction was stopped by adding 1 mL of Hg[Cl.sub.2] solution (0.4 mg/mL in distilled water). The soil was removed from the solution by centrifugation at 2500G for 10 min and filtered through Whatman paper GF/F. The absorbance was measured with a spectrophotometer at 490 nm. The amount of fluorescein in the filtrate was determined from a standard curve (490 nm) and FDA hydrolysis activity was expressed as nmol of fluorescein/h.g soil (dry weight). Three replicates were prepared for each soil sample.

Percentage of microorganisms tolerant to Ni

The proportion of Ni-tolerant microorganisms in the soils was determined by plate counts using the same media as for the microbial population counts. Tolerance to Ni was estimated by adding to the medium 50 [micro]g/mL of [Ni.sup.2+] (acetate) for bacteria and 150 [micro]g/mL of the same compound for fungi, which are more resistant to metals (Amir and Pineau 1998b). These concentrations were chosen for the statistical analysis among other tested doses. The metal was sterilised by dry heating to 150[degrees]C for 1 h, then mixed into the medium, which was poured into the Petri dishes after soil suspensions had been included. The plates were incubated at 28[degrees]C for 7 days and the number of CFU was determined. Three replicate plates were used for each treatment. The percentages of tolerance to Ni of bacteria and fungi were then calculated by comparison with total counts.

Laboratory experiment with incubated soil

In order to confirm results obtained with statistical analysis, an experiment with soils incubated in different conditions was performed. Small jars (300 mL) were sterilised empty by dry heating (150[degrees]C for 1 h). Each jar received 150 g of saprolite or HMB soil. Three different treatments of the soils were performed: (1) steam sterilisation in autoclave (120[degrees]C for 1 h), in order to eliminate the native microflora; (2) reinoculation of autoclaved samples by adding 1 g of non-treated soil; (3) carbon supply--a mixture of cellulose (0.66%), pectin (0.66%), and starch (0.66%) added to soil as powder. This complex amendment (total of 2%) was designed to keep slowly degradable carbon sources available for several months, accompaning the slow metal release process.

Combinations of some of these treatments were also prepared. There were 3 replicates for each treatment. The soil was humidified to 80% of its field capacity and shaken for 30 min in a rotative shaker; then the jars were closed and sealed with parafilm before being incubated in dark at 28[degrees]C. The soil humidity was checked each 2 months and reajusted if necessary.

After 9 months, 3 types of analysis were carried out. DTPA-extractable Ni and Co concentrations were determined. Global soil microbial activity was measured by FDA method (see before).

Statistical method of analysis of results

The statistical method used consists of investigating the various types of correlations between the parameters considered (extractable metals, pH, microbial activity, microbial density, tolerance to Ni, etc.). This method is regarded as sound and reliable, particularly when dealing with a large number of samples, as is the case here (Schwartz 1983). The links between the main parameters (or variables) to be explained (extractable metals and tolerance to Ni) and the explanatory variables (pH, microbial activity, microbial density, etc.) were analysed through 3 types of correlations:

(1) Total correlations, which indicate the importance of the link as determined directly from comparison of the values of the variables, considered in pairs.

(2) Partial correlations, which enable links to be detected between 2 parameters when a third variable is made constant. This approach makes it possible to correct distortions in the links between 2 variables that may occur as a result of wide variations of another contributing factor (Dagnelie 1986). They are useful for checking whether the correlation between 2 variables is indirect, expressing their common link with a third variable. The correlation between these 2 variables would then disappear when the third is made constant.

(3) Multiple correlations, which enable complex links to be detected between a number of other explanatory variables and one variable to be explained, the latter being theoretically regarded as resulting from the former (Dagnelie 1986). Multiple r shows the overall relationship between one variable to be explained, and the sum of the explanatory variables. Multiple P indicates the risk that the multiple correlation obtained is due to chance. Partial [r.sup.2] shows the contribution of each of these variables to the sum; Partial P indicates the corresponding risk.

In order to improve the linearity of the dot series, the values of some variables were converted to the logarithmic scale. The statistical analyses were performed by computer with the program STATITCF, version 5 (1991).


General comments on the soils investigated

The 40 soil samples investigated can be all considered as ultramafic or serpentinic soils, since they are all derived from an ultramafic metalliferous parent rock and therefore present a high metal content. As shown in Table 1, the metal contents were high. However, only a small fraction of these metals can possibly affect the microflora directly, i.e. the available fraction. They also contained very low amount of calcium and were deficient in phosphorus.

Apart from these few common characteristics, the 40 soil samples showed a wide diversity, both in their abiotic characteristics (Table 2) and in their biotic ones (Table 3). Their pH ranged from 4.84 to 7.38, and organic carbon content from 0.01 to 80.8 mg/g. The soils with high carbon content were those that were sampled under plant litter where fine roots were abundant. The content of metals extractable by DTPA also varied widely from one soil to another: >3000 [micro]g/g in certain hypermagnesian brown soils (HBS), <20 [micro]g/g in some ferralitic oxidic soils (FOS).

Regarding biotic characteristics, the total density of aerobic organotrophic microbial populations ranged from [10.sup.4] CFU/g soil to 550 x [10.sup.4] CFU/g soil. In some soils, Ni-tolerant microorganisms predominated, in others their percentage was low. Total microbial activity, measured by the FDA enzymatic method, ranged from 11 to 272 nmol fluorescein/h.g soil.

The diversity inside each of the 3 soil groups was as high as between the groups. However, it appeared that the rooty soils were generally more rich in carbon and extractable metals than the non-rooty soils and had a higher activity and more microorganisms. All these characteristics were also higher in the HBS group than in FOS and CFOS. In addition, bacteria and fungi appeared more tolerant to Ni in HBS. FOS contained generally the lowest concentrations of extractable metals and microbial density and activity were low.

In view of the large number of significant correlations observed (Table 4), the statistical analysis was focused on 3 types of variables of particular relevance for this study: the extractable metal content, the microbial activity, and density and the proportion of Ni-tolerant microorganisms.

Correlations between extractable metal contents and the other variables

Table 4 shows that extractable metal contents were significantly correlated with other variables and that these correlations were comparable for all 5 metals investigated. In order to avoid overloading and repetitions, the major trends have been summarised in a simplified table, in which the 5 metals are grouped together to form a single factor designated 'extractable metals' (we checked that partial and multiple correlations with the variables were also similar for all 5 metals). It must be emphasised that there were high positive total correlations (P < 0.01) between extractable metal content and the following characteristics: pH, organic carbon, microbial activity (FDA), microbial density, and tolerance of microflora to Ni. Table 5 thus summarises the significant total, partial, and multiple correlations obtained with the variables which have an explanatory value.

The partial correlations generally confirmed the total correlations, except in the case of organic carbon, whose relationship to extractable metal content was no longer significant when FDA was regarded as constant.

For the multiple correlations, the extractable Ni content, and also the total extractable metal content, can be better explained by the additive effects of the FDA activity and the Ni tolerance of the microflora (the multiple correlation coefficient is 0.679 instead of 0.624 for Ni, and 0.733 instead of 0.638 for extractable metals).

Fig. 1 shows the linear regressions between extractable Ni content on the one hand and FDA and microbial density on the other hand, thus clearly revealing the positive correlations between these characteristics. Fig. 2 shows the same regressions for Co.


Correlations between the percentage of Ni-tolerant microorganisms and the other variables

The percentage of Ni-tolerant microorganisms (Table 3) was low for some samples (>10% for saprolite) and >100% for others (bacteria from HBS7; fungi from HBS4), which means that the number of CFU in the Ni-added medium was higher than in the control medium without Ni. Table 6 shows this percentage of Ni-tolerant microorganisms to be significantly correlated only with extractable Ni content and total extractable metal content. For these variables, the partial correlations with Ni tolerance remained high when other factors were held constant. No multiple correlation was detected for Ni tolerance.

Laboratory experiment with incubated soil

The extractable concentrations of Ni and Co in soil microcosms incubated 9 months at 28[degrees]C varied with treatment (Table 7). In the non-treated but incubated saprolite (incubated control), available Ni was significantly higher than in the non-incubated soil (general control), but extractable Co was not affected. The autoclaved soil, where the native microflora were mainly eliminated, contained nearly the same concentrations of Ni and Co as the general control, but when the autoclaved soil was re-inoculated by a pinch of the same soil, the amounts of extractable Ni and Co were higher (at the same level as in the incubated control). Slowly degradable organic supply enhanced greatly the extractable contents of Ni and Co in the non-autoclaved saprolite.

The amount of metals released varied considerably. The concentration of Co showed an increase of 9% (compared with the non-incubated control) for the incubated non-treated saprolite, and 77% for the carbon-amended treatment. The increase varied from 4% to 20% for Ni.

In the HBS, the effects on extractable Ni and Co of incubation without treatment, or incubation with organic carbon supply, were similar to those noticed for the saprolite.

The soil microcosms that had the highest contents of extractable Ni and Co also showed the highest levels of global microbial activity, except when the native microflora was eliminated (autoclaved samples).


Although all 40 ultramafic samples of soil studied were characterised by a relatively high total metal content, they displayed a wide diversity in terms of the characteristics considered. This diversity is one of the essential requisites for discovering links between these characteristics by statistical methods. The high variance in the values of each characteristic makes it possible to detect a complex pattern of relationships between them.

It must be noted, first of all, that the links shown by correlation analysis do not necessarily indicate causal relationships. In some cases the correlation may be explained by the fact that 2 characteristics are themselves determined by a third factor. The clues provided by the partial correlations, together with comparisons with data obtained in other contexts, allow coherent hypotheses, but laboratory experiments are necessary to complete the statistical approch.

It is well known that microbial activity and microbial density in soil depend in greatly on carbon sources (Paul and Clark 1989). That is why HBS samples, containing more organic carbon, were richer in microorganisms and more active than the other samples. The same reason applies to the high values of microbial density and activity in the rooty soils, in comparison with the non-rooty ones.

Metal release in the soil

Only a few authors have published data on the various contents of extractable metals in New Caledonian ultramafic soils (Jaffre and Rigault 1991; L'Huillier 1994; Becquer et al. 1995; Becquer et al. 2002). These contents are observed to vary widely according to the soils. What causes this variability: soil pH, microflora, or plants?

Although these 3 mechanisms are more or less related (both microorganisms and plants can affect the soil pH; plants stimulate microbial activity), their relative importance requires further consideration. If the influence of pH on the extractable metal contents (metal release induced by acidity) in New Caledonian ultramafic soils was high, one would expect to observe a negative correlation between these 2 characteristics, whereas, on the contrary, we recorded a significant positive correlation between pH and different extractable metals. This positive link is probably the result of the high weight of the HBS in the statistical analysis. Indeed, these soils are very rich in extractable metals and are actually characterised by higher pH values than the other soils. Nevertheless, this last observation, and the absence of a negative correlation, suggest that overall acidity of soil does not play an important role in the release of metals in New Caledonian ultramafic soils. But it remains probable, as suggested by Berthelin et al. (1995), that, in any given soil, the local acids secreted by microorganisms or plant roots can contribute to metal release.

Regarding the biological mechanism, it is noted that the soils with the highest extractable metal content are also the most active and the richest in microorganisms (high positive correlation, Figs 1 and 2), which suggests that it is microbial activity that releases the metals, especially Ni and Co, in the soil. The high positive correlations between the various extractable metals on the one hand, and microbial activity and density on the other, do seem to reflect a direct relationship, for they remain high in terms of partial correlations (when other variables are held constant).

The results of the laboratory experiment clearly support this conclusion; a saprolite and an HBS, incubated 9 months in conditions of humidity and temperature favourable to microorganisms, were found to present higher contents of Ni and Co, without there being any notable change in the pH. When the native microflora were mainly eliminated by heat treatment (autoclave), the extractable content of Ni and Co remained at the same level as the non-incubated control. In contrast, when the autoclaved microcosms were re-inoculated with a pinch of the same soil, the values were clearly enhanced. This indicates that Ni and Co release, in the 2 tested soils, corresponds to biological processes due probably to specialised microorganisms that disappeared in the autoclaved soil. Indeed, in this case, the microorganisms that re-colonised the soil reached the same level of activity as in the non-steamed soils, but could not induce the metal release. The enrichment of soil with a mixture of slowly degradable carbon compounds induced the highest microbial activities and also the highest levels of extractable Ni and Co, which suggests that the release of these metals is due to heterotrophic microorganisms.

A number of studies have stressed the role of microorganisms in the release of Fe (Munier-Lamy and Berthelin 1987; Francis and Dodge 1988; Backes et al. 1993), Mn (Broomfield 1956; Ghiorse 1988; Berthelin et al. 1995), and Mg (Lebedeva et al. 1978; Berthelin 1988). Few studies have concerned Ni and reported only indirect effects related to acid production (Mehta et al. 1999; Valix et al. 2001). However, a recent study has pointed out the possible role of bacterial reduction in the release of very small amounts of Ni and Co in New caledonian ferralsols (Quantin et al. 2001). In the same way, the microbial transformation of Co is not well studied. Some authors have suggested the bacterial enzymatic oxidation of Co in seawater (Lee and Tebao 1994; Moffett and Ho 1996).

Ni tolerance of microorganisms

The higher abundance of microorganisms in the soils having the highest extractable metal contents seems paradoxical, since the metal concentrations in these soils are clearly toxic to microorganisms (Amir and Pineau 1998b). This apparent contradiction disappears when the level of tolerance of the microflora to these metals is taken into account, for a positive correlation was observed between the extractable metal contents in the soils and the Ni tolerance of their microflora, suggesting microbial tolerance to metals may increase as metals are released by microbial activity. The multiple correlation obtained also bears out this hypothesis, since the effects of FDA activity and of tolerance to Ni compound to produce an even higher correlation with the extractable metal content. In these soils, therefore, release of metals constitutes a selective pressure that causes the microflora to evolve, by natural selection, higher metal tolerance, enabling an adequate degree of activity to be maintained in an increasingly toxic environment. However, if one compares the microbial activity and density of HBS with soils having no metal toxicity but a comparable organic matter content, the former are found to be markedly less active and poorer in microflora (H. Amir unpublished data), which would suggest that metal toxicity nevertheless places a limit on this activity, or that adjustment to the metals has a high energy cost.

The percentages of microorganisms tolerant to Ni vary considerably with the soils considered but are generally higher than in non-serpentinic soils (Amir and Pineau 1998b). This higher abundance of metal-tolerant microorganisms in metal-contaminated soils or soils that are naturally rich in metals has already been reported by other authors (Jordon and Lechevalier 1975; Shetty et al. 1994), but only for one or a few soils. Investigation of a large number of soils here revealed a gradient of Ni-tolerance percentage that increases with the extractable contents of these two metals. This suggests that the phenomenon of adaptation of the microflora to metal toxicity is general in New Caledonian ultramafic soils, and that the higher the quantity of toxic metals in the soil, the more marked the process is. Some tolerance values exceed 100%, which, with the technique used, indicates that a number of isolates require the presence of metals at normally toxic concentrations for development. Some of these isolates have been purified. Their better growth in presence of 10-50 [micro]g/mL of Ni has been confirmed. Most of them appeared to have a slow mycelial growth (H. Amir unpublished data) and are now under study.

As our investigation was performed on a large number of different soils, it has the advantage of bringing to light the general phenomena affecting certain complex characteristics of ultramafic soils, such as the extractable metal content and the percentage of metal-tolerant microorganisms. Two points can be stated by way of conclusion. The release of different metals, in New Caledonian ultramafic soils, is mainly related to microbial activity; this was confirmed especially for Ni and Co, which have not previously been reported as directly influenced by microflora in soil. The phenomenon of adaptation of the microflora to metal toxicity is widely spread in these soils; the higher the quantity of toxic metals in the soil, the more marked the process.
Table 1. Some physico-chemical characteristics of samples of the main
types of the studied soils

Characteristics Hypermagnesian Ferralitic Ferrallitic Saprolite
 brown soil oxidic oxidic
 soil colluvial

Clay (%) 29.7 9.1 11.6 15.5
Silt (%) 39.8 13.8 31.9 33.1
Sand (%) 20.4 80.7 58.0 51.7
Organic carbon (%) 3.62 1.94 2.60 0.16
Total nitrogen (%) 0.25 0.079 0.009 0.01
C/N 14.47 24.5 293.5 16.0
pH ([H.sub.2]O) 6.9 4.4 5.6 6.2
pH (KCl) 5.9 4.8 5.8 /
Ca (%) 0.49 0.02 0.06 0.64
K (%) 0.03 Traces Traces 0.00
Na (%) 0.05 Traces Traces 0.02
Mg (%) 8.89 1.68 0.49 27.88
P ([micro]g/g) 123 153 56 /
Fe (%) 25.89 62.16 74.3 17.03
Al (%) 3.97 7.43 3.97 0.78
Mn(%) 0.57 0.21 0.35 0.31
Ni (%) 0.72 0.32 1.12 1.93
Cr (%) 1.90 12.45 3.07 0.28
Co (%) 0.07 0.03 0.07 0.03

Table 2. Physico-chemical characteristics of 40 ultramafic soil
samples from New Caledonia

Soils Organic pH([H.sub.2]O)
 carbon Mg

HBS1 (A) 0.96 7.38 1363
HBS2 (B) 1.81 6.94 2235
HBS3 (A) 1.45 5.20 1438
HBS4 (B) 2.78 6.89 1518
HBS5 (A) 1.09 6.93 2015
HBS6 (B) 2.82 6.56 1900
HBS7 (A) 2.23 6.90 2217
HBS8 (B) 7.41 6.82 2441
HBS9 (A) 2.40 7.18 1949
HBS10 (B) 3.07 6.86 1507
HBS11 (A) 6.93 6.59 1642
HBS12 (B) 8.08 6.57 1684
Average (AC) 2.51 6.69 1770.6
Average (BD) 4.32 6.77 1880.8

FOS1 (A) 0.48 6.57 4
FOS2 (B) 2.70 5.93 41
FOS3 (A) 0.46 5.18 2
FOS4 (B) 2.79 4.98 63
FOS5 (A) 0.64 5.52 2
FOS6 (B) 2.34 5.76 20
FOS7 (A) 0.62 6.02 2
FOS8 (B) 3.18 5.89 51
FOS9 (A) 1.39 4.85 3
FOS10 (B) 3.12 5.54 54
FOS11 (A) 0.95 4.99 3
FOS12 (B) 1.86 5.26 17
Average (AC) 0.76 5.52 2.6
Average (BD) 2.66 5.56 41.0

CFOS1 (A) 1.03 5.11 3
CFOS2 (B) 2.75 4.91 26
CFOS3 (A) 0.65 5.70 174
CFOS4 (B) 1.71 5.20 14
CFOS5 (A) 0.24 5.99 22
CFOS6 (B) 3.40 5.23 79
CFOS7 (A) 3.34 5.36 125
CFOS8 (B) 4.64 6.02 161
CFOS9 (A) 2.19 6.52 38
CFOS10 (B) 3.16 5.23 134
CFOS11 (A) 0.71 5.28 3
CFOS12 (B) 3.19 5.45 61
CFOS13 (A) 0.11 5.25 4
CFOS14 (B) 2.20 5.86 62
Average (AC) 1.18 5.60 52.7
Average (BD) 3.00 5.41 76.7

Lateritic 0.001 5.51 2
Saprolite 0.052 6.75 380

Soils Extractable metals ([micro]g/g)
 Mn Fe Ni Co

HBS1 (A) 221 38 83 29
HBS2 (B) 339 64 175 44
HBS3 (A) 370 54 146 38
HBS4 (B) 383 74 179 32
HBS5 (A) 175 66 107 23
HBS6 (B) 227 70 187 34
HBS7 (A) 119 72 125 14
HBS8 (B) 336 95 305 25
HBS9 (A) 339 57 161 37
HBS10 (B) 429 59 333 67
HBS11 (A) 287 74 265 49
HBS12 (B) 228 82 225 31
Average (AC) 251.8 60.1 147.8 31.6
Average (BD) 323.6 74.0 234.0 38.8

FOS1 (A) 2 6 0.1 0.1
FOS2 (B) 14 59 2 0.1
FOS3 (A) 5 8 0.1 0.1
FOS4 (B) 29 22 8 1
FOS5 (A) 4 9 0.1 0.1
FOS6 (B) 6 15 3 0.1
FOS7 (A) 7 15 0.1 0.1
FOS8 (B) 53 27 15 5
FOS9 (A) 2 11 0.1 0.1
FOS10 (B) 83 31 8 3
FOS11 (A) 10 11 0.1 1
FOS12 (B) 70 31 2 2
Average (AC) 5.0 10.0 0.1 0.2
Average (BD) 42.5 30.8 6.3 1.8

CFOS1 (A) 99 15 1 10
CFOS2 (B) 536 89 11 34
CFOS3 (A) 286 19 21 28
CFOS4 (B) 417 88 8 27
CFOS5 (A) 23 9 0.1 6
CFOS6 (B) 540 76 87 68
CFOS7 (A) 368 51 55 41
CFOS8 (B) 555 29 45 29
CFOS9 (A) 397 50 24 97
CFOS10 (B) 664 62 60 192
CFOS11 (A) 11 15 0.1 2
CFOS12 (B) 345 71 20 34
CFOS13 (A) 59 18 0.1 13
CFOS14 (B) 362 45 27 48
Average (AC) 177.5 25.2 14.4 28.1
Average (BD) 488.4 65.7 36.8 61.7

Lateritic 15 23 0.1 2
Saprolite 5 23 57 1

(A) Soils collected in bare patches with only few roots.
(B) Corresponding rooty soils, collected under plants and under
litter. (C) Average of the non-rooty soils. (D) Average of the
rooty soils.

Table 3. Studied microbiological characteristics of 40 ultramafic
soil samples from New Caledonia

Soils Microbial Fungal Microbial
 density (A) microflora (A) activity
 (x [10.sup.4]) (x [10.sup.2]) (FDA) (B)

HBS1 (D) 35.80 95.58 89.10
HBS2 (E) 389.60 106.70 272.20
HBS3 (D) 95.70 106.70 98.80
HBS4 (E) 343.20 68.72 196.50
HBS5 (D) 61.00 91.84 99.30
HBS6 (E) 348.20 115.58 230.30
HBS7 (D) 13.80 71.52 85.50
HBS8 (E) 550.00 123.97 246.60
HBS9 (D) 142.60 129.02 114.70
HBS10 (E) 382.80 152.93 211.50
HBS11 (D) 220.00 72.97 169.50
HBS12 (E) 519.50 247.15 206.60
Average (DF) 94.80 94.60 109.48
Average (EG) 422.21 295.89 227.28

FOS1 (D) 5.58 20.91 47.00
FOS2 (E) 11.25 68.03 153.80
FOS3 (D) 3.60 41.26 28.70
FOS4 (E) 11.70 35.87 117.80
FOS5 (D) 8.10 29.37 31.60
FOS6 (E) 10.40 34.47 82.00
FOS7 (D) 1.30 18.18 50.20
FOS8 (E) 20.40 62.18 80.70
FOS9 (D) 4.10 20.91 60.30
FOS10 (E) 47.90 54.60 108.50
FOS11 (D) 13.20 40.45 92.60
FOS12 (E) 36.90 50.91 124.40
Average (DF) 5.98 28.51 51.73
Average (EG) 23.09 51.01 111.20

CFOS1 (D) 7.10 37.34 56.30
CFOS2 (E) 55.80 47.47 172.80
CFOS3 (D) 8.50 17.29 97.00
CFOS4 (E) 47.70 37.71 154.00
CFOS5 (D) 1.40 20.09 31.20
CFOS6 (E) 17.30 36.23 167.70
CFOS7 (D) 57.20 42.95 137.10
CFOS8 (E) 98.90 56.83 181.80
CFOS9 (D) 24.90 43.38 146.20
CFOS10 (E) 59.70 41.68 193.50
CFOS11 (D) 4.00 19.11 65.30
CFOS12 (E) 44.50 46.06 176.50
CFOS13 (D) 4.50 23.34 108.80
CFOS14 (E) 29.00 40.04 209.30
Average (DF) 15.37 29.07 91.70
Average (EG) 50.41 43.71 179.37

Lateritic 15.50 37.34 11.10
Saprolite 39.10 30.88 30.90

Soils Bacterial Fungal tolerance
 tolerance to Ni (%) (C)
 to Ni (%) (C)

HBS1 (D) 79.60 46.60
HBS2 (E) 74.60 91.10
HBS3 (D) 46.50 56.60
HBS4 (E) 51.10 145.30
HBS5 (D) 71.30 44.90
HBS6 (E) 40.40 58.80
HBS7 (D) 157.60 86.70
HBS8 (E) 48.20 46.40
HBS9 (D) 53.90 36.60
HBS10 (E) 33.40 46.90
HBS11 (D) 39.70 64.50
HBS12 (E) 50.40 6.80
Average (DF) 74.76 55.98
Average (EG) 49.68 65.88

FOS1 (D) 45.80 29.10
FOS2 (E) 45.90 22.40
FOS3 (D) 42.30 13.50
FOS4 (E) 48.70 13.70
FOS5 (D) 37.90 36.50
FOS6 (E) 23.30 14.70
FOS7 (D) 40.40 48.20
FOS8 (E) 35.50 39.20
FOS9 (D) 38.40 9.40
FOS10 (E) 82.40 40.40
FOS11 (D) 36.30 36.20
FOS12 (E) 20.20 32.40
Average (DF) 40.18 28.81
Average (EG) 42.66 27.13

CFOS1 (D) 35.50 52.40
CFOS2 (E) 18.70 40.60
CFOS3 (D) 31.50 23.80
CFOS4 (E) 31.60 46.80
CFOS5 (D) 21.40 48.50
CFOS6 (E) 45.50 67.60
CFOS7 (D) 37.90 53.80
CFOS8 (E) 39.20 28.30
CFOS9 (D) 68.10 31.20
CFOS10 (E) 48.30 23.90
CFOS11 (D) 32.50 43.00
CFOS12 (E) 61.30 41.50
CFOS13 (D) 36.10 55.10
CFOS14 (E) 45.10 20.80
Average (DF) 37.57 43.97
Average (EG) 41.38 38.50

Lateritic 14.80 6.50
Saprolite 43.20 39.90

(A) Total aerobic organotrophic, and fungal microflora
(CFU/g dry soil). (B) Fluoroscein diacetate activity
(nmoles fluorescein/h.g dry soil). (C) Bacteria and fungi
tolerant to Ni, expressed as % of the control (without Ni).
(D) Soils collected in bare patches with only few roots.
(E) Corresponding rooty soils, collected under plants and
under litter. (F) Average of the non-rooty soils.
(G) Average of the rooty soils.

Table 4. Correlation matrix between different physico-chemical and
microbiological characteristics of 40 ultramafic soil samples from
New Caledonia

OC, organic carbon; logMg--logCo, log extractable Mg,
Mn, Fe, NJ, Co; logmetals, log of the 5 extractable metals sum;
logmicrof., log total aerobic organotrophic micro-flora; logfungi,
log fungal microflora; FDA, fluoroscein diacetate activity
(global microbial activity); tol.Nibact., % of bacteria tolerant to
Ni; tol.Nifungi, % of total fungi tolerant to Ni. Significance levels:
for P--0.05, 0.3125; for P = 0.01, 0.4032; for P = 0.001, 0.5013.
Interesting significant correlations are underlined

 OC pH logMg logMn logFe logNi logCo

pH 0.236 1.000
logMg 0.535 0.706 1.000
logMn 0.507 0.235 0.662 1.000
logFe 0.614 0.346 0.739 0.828 1.000
logNi 0.616 0.553 0.932 0.795 0.842 1.000
logCo 0.414 0.271 0.648 0.948 0.745 0.746 1.000
logmetals 0.548 0.561 0.924 0.871 0.883 0.955 0.850
logmicrof 0.676 0.492 0.802 0.694 0.782 0.843 0.645
logfungi 0.613 0.558 0.781 0.555 0.697 0.744 0.485
FDA 0.679 0.272 0.609 0.746 0.773 0.703 0.652
tol.Nibact. 0.100 0.465 0.462 0.207 0.321 0.389 0.205
tol.Nifungi 0.041 0.365 0.426 0.375 0.377 0.375 0.384

 logmetals logmicrof. logfungi FDA tol.Nibact.

logmetals 1.000
logmicrof 0.830 1.000
logfungi 0.745 0.845 1.000
FDA 0.716 0.785 0.602 1.000
tol.Nibact. 0.401 0.156 0.330 0.119 1.000
tol.Nifungi 0.452 0.329 0.212 0.308 0.357

Table 5. Significant total, partial, and multiple correlations between
the concentration of extractable metals in 40 New Caledonian ultramafic
soil samples and some of their physico-chemical and microbiological

Significance levels: total correlations (d.f. = 2): for P = 0.05,
0.3125; for P = 0.01, 0.4032; for P = 0.001, 0.5013. Partial
correlations (d.f. = 3): for P = 0.05, 0.3165; for P = 0.01,
0.4082; for P = 0.001, 0.5072

Variables Total correlations with Partial correlations
 'extractable metals' (A) with 'extractable

Organic carbon 0.548 ** 0.120 when FDA
Fluorescein diacetate
 activity (FDA) 0.716 ** ** (B)
Microbial density (C) 0.830 ** **
Total fungal community 0.745 ** **
Tolerance to Ni (D) 0.674 ** **

Explanatory Multiple correlation with
variables 'extractable Ni' (E)

 Multiple r Multiple P (%)

1. FDA Level 1: 0.624 ** 0.00
2. Tolerance to Ni Level 2: 0.679 ** 0.00

Explanatory Multiple correlation with 'extractable
variables metals' (E)

 Multiple r Multiple P (%)

1. FDA 0.638 ** 0.00
2. Tolerance to Ni 0.733 ** 0.00

* P < 0.05; ** P < 0.01

(A) Extractable metals: sum of the concentrations of the 5 studied

(B) There are several values of partial correlations in relation to
the factor held constant; all are highly significant.

(C) Aerobic organotrophic microflora.

(D) Average of the % tolerance to Ni of bacteria and fungi.

(E) Shown here are the statistical parameters that could reflect the
complementarity of the links between some variables and the
characteristics 'extractable NJ' or 'extractable metals': multiple
correlation coefficient (multiple r) showing the overall relationship
between this characteristic and the sum of the explanatory variables;
multiple P (in %) indicating the risk that the multiple correlation
obtained was due to chance.

Table 6. Significant total and partial correlations between the
percentage of microorganisms tolerant to Ni in 40 New Caledonian
ultramafic soil samples and some of their characteristics

No multiple correlation was detected. Significance levels: total
correlations (d.f. = 2): for P = 0.05, 0.3125; for P = 0.01,
0.4032; for P = 0.001, 0.5013. Partial correlations (d.f. = 3):
for P = 0.05, 0.3165; for P = 0.01, 0.4082; for P = 0.001, 0.5072

Explanatory variables Total correlations with Partial correlations
 'tolerance to Ni' (A) with 'tolerance to Ni'

Extractable Ni 0.467 ** ** (B)
Extractable metals (C) 0.524 ** **

* P < 0.05; ** P < 0.01

(A) Average of % tolerance to Ni of bacteria and fungi.

(B) There are several values of partial correlations in relation to the
factor held constant; all are highly significant.

(C) Sum of the concentrations of the 5 studied metals.

Table 7. Influence of different treatments on Ni and Co extractable
contents, on microbial activity, and on propagule density of a
saprolite and hypermagnesian brown soil from New Caledonia, after 9
months incubation

For each soil, values followed by the same letter are not significantly
different at P = 0.05 (Newman-Keuls test)

Treatments pH (A) DTPA-extractable FDA (B) (nmol
 contents fluorescein/
 ([micro]g/g soil) h.g soil)

 Ni Co

Non incubated (C) 6.2 329b 22a 20.6a
Non treated (D) 6.3 342c 22a 32.4b
Autoclaved 6.3 325a 24a --
Autoclaved reinoculated (E) 6.3 344c 23a 39.9b
+ Organic carb. (F) 6.5 395d 39b 149.7c

 Hypermagnesian brown-soil

Non incubated (c) 6.9 52a 9a 54.3a
Non treated (D) 6.8 60b 12b 87.8b
Autoclaved 6.8 48a 9a --
+ Organic carb. (F) 7.0 71c 14c 271.2c

(A) pH ([H.sub.2]O), at the end of the experiment except when

(B) Fluorescein diacetate hydrolic activity (global microbial

(C) Analysis done before the experiment.

(D) Soil humidified and incubated without treatment.

(E) Autoclaved 1 h at 120[degrees]C, then reinoculated with a pinch
of non-treated saprolite.

(F) Organic carbon supply: cellulose (0.66%) + pectin (0.66%) + starch
(0.66%), total of 2% C.


This research was mainly supported by the South Province of New Caledonia.


Amir H, Pineau R (1998a) Influence of plants and cropping on microbiological characteristics of some New Caledonian ultramafic soils. Australian Journal of Soil Research 36, 457-471.

Amir H, Pineau R (1998b) Effects of metals on the germination and growth of fungal isolates from New Caledonian ultramafic soils. Soil Biology & Biochemistry 30, 2043-2054.

Backes CA, Pulford ID, Duncan HJ (1993) Seasonal variation of pyrite oxidation rates in colliery spoil. Soil Use and Management 9, 30-34.

Becquer T, Bourdon E, Petard J (1995) Disponibilite du nickel le long d'une toposequence de sol developpee sur roche ultramafique de Nouvelle-Caledonie. Comptes Rendu de l'Academie des Sciences 321, 585-592.

Becquer T, Rigault F, Jaffre T (2002) Nickel bioavailability assessed by ion exchange resin in the field. Communications in Soil Science and Plant Analysis 33, 439-450.

Berthelin J (1988) Microbial weathering processes in natural environments. In 'Physical and chemical weathering in geochemical cycles'. (Eds A Lerman, M Meybeck) pp. 33-59. (Kluwer Academic Publishers: London)

Berthelin J, Munier-Lamy C, Leyval C (1995) Effect of microorganisms on mobility of heavy metals in soils. In 'Environmental impact of soil component interactions. Metals, other inorganics and microbial activities'. (Eds PM Huang, J Berthelin., JM Bollag, WB McGill, AL Page) pp. 3-17 (CRC Lewis: London)

Brooks RR (1987) The serpentine factor. In 'Serpentine and its vegetation. A multidisciplinary approach'. (Ed. RR Brooks) pp. 32-59. (Discorides Press: Portland, OR)

Broomfield SM (1956) Oxidation of manganese by soil micro-organisms. Australian Journal of Biological Science 9, 238-252.

Dagnelie P (1986) 'Analyse statistique a plusieurs variables.' (Presses Agronomiques de Gembloux: Gembloux, France)

Francis AJ, Dodge CJ (1988) Anaerobic microbial dissolution of transition and heavy metal oxides. Applied and Environmental Microbiology 54, 1009-1014.

Gasser UG, Dahlgren RA, Ludwig C, Lauchli AE (1995) Release kinetics of surface associated Mn and Ni in serpentinic soils: pH effects. Soil Science 16, 273-280.

Ghiorse WC (1988) The biology of manganese transforming microorganisms in soil. In 'Manganese in soils and plants'. (Eds D Graham, RJ Hannan, NC Uren) pp. 75-85. (Kluwer Academic Publishers: Dordrecht)

Jaffre T (1980) Etude ecologique du peuplement vegetal des sols derives de roches ultrabasiques en Nouvelle Caledonie. (ORSTOM Press: Paris)

Jaffre T, Rigault F (1991) 'Recherche sur les possibilites d'implantation vegetale sur sites miniers.' Convention Science de la Vie, No. 5. (ORSTOM Press: Noumea)

Jordon MJ, Lechevalier MP (1975) Effects of zinc smelter emissions on forest soil microflora. Canadian Journal of Microbiology 21, 1855-1865.

Lebedeva EV, Lyalikova NN, Bugelskii YY (1978) Participation of nitrifying bacteria in the weathering of serpentinized ultrabasic rocks. Mikrobiologiya 47, 1101-1107.

Lee Y, Tebao BM (1994) Cobalt (II) oxidation by the marine manganese (II)-oxidizing Bacillus sp. strain SG-1. Applied and Environmental Microbiology 60, 2949-2957.

L'Huillier L (1994) Biodisponibilite du nickel dans les sols ferrallitiques ferritiques de Nouvelle-Caledonie. Effets toxiques de Ni sur le developpement et la physiologie du mais. PhD thesis, University of Montpellier II.

Mehta KD, Pandey BD, Premchand (1999) Bio-assisted leaching of copper, nickel and cobalt from copper converted slag. Materials Transactions JIM 40, 214-221.

Moffett J, Ho J (1996) Oxidation of cobalt and manganese in seawater via a common microbially catalysed pathway. Geochimica Cosmochimica Acta 60, 3415-3424.

Munier-Lamy C, Berthelin J (1987) Formation of polyelectrolyte complexes with the major elements Fe and Al and the trace elements U and Cu during heterotrophic microbial leaching of rocks. Geomicrobiological Journal 5, 119-147.

Paul EA, Clark FE (1989) 'Soil microbiology and biochemistry.' (Academic Press: San Diego)

Quantin C, Becquer T, Rouiller JH, Berthelin J (2001) Oxide weathering and trace metal release by bacterial reduction in a New Caledonian ferralsol. Biogeochemistry 53, 323-340.

Schlegel HG, Cosson JP, Baker AJM (1991) Ni hyperaccumulating plants provide a niche for Ni-resistant bacteria. Botanical Acta 104, 18-25.

Schnurer J, Rosswall T (1982) Fluorescein diacetate hydrolysis as a measure of total microbial activity in soil and litter. Applied and Environmental Microbiology 43, 1256-1261.

Schwartz D (1983) 'Methodes statistiques a l'usage des medecins et des biologistes.' (Flammarion: Paris)

Shetty KG, Banks MK, Hetrick BA, Schwab AP (1994) Biological characterization of a southeast Kansas mining site. Water Air and Soil Pollution 78, 169-177.

Valix M, Usai F, Malik R (2001) Fungal bio-leaching of low grade laterite ores. Minerals Engineering 14, 197-203.

Werner W (1996) Biotransfer of heavy metals as a function of site-specific and crop-specific factors. In 'Plant research and development 43'. (Institute For Scientific Cooperation: Tubingen, Germany)

Zelles L, Scheunert I, Kreutzer K (1987) Bioactivity in limed soil of spruce forest. Biology and Fertility of Soils 3, 211-216.

Manuscript received 26 March 2002, accepted 30 October 2002

Laboratoire de Biologie et Physiologie Vegetales Appliquees, Universite de la Nouvelle-Caledonie, BP 4477, 98847, Noumea cedex, New Caledonia.
COPYRIGHT 2003 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:nickel, cobalt
Author:Amir, Hamid; Pineau, Rene
Publication:Australian Journal of Soil Research
Geographic Code:8NEWC
Date:Mar 1, 2003
Previous Article:Using quantity/intensity relationships to assess the potential for ammonium leaching in a Vertosol.
Next Article:Dispersibility of glacial loess in particle size analysis, USA.

Related Articles
Caledonia Announces a 953 Million Tonne Resource for Its Nama Project, and Future Development Plans
INCO announces completion of acquisition by Sumitomo Metal Mining Co., Ltd. and Mitsui & Co., Ltd. of a 21% interest in Goro Nickel S.A.
Inco announces decision to proceed with Goro Project.
Bison Confirms Nickel Province.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters