HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Weil, R. R. Search for Related Content PubMed Articles by Weil, R. R. Agricola Articles by Weil, R. R.

SOIL FERTILITY

Soil and Plant Influences on Crop Response to Two African Phosphate Rocks

Department of Natural Resource Sci. and Landscape Arch., Univ. of Maryland, College Park, MD 20742-4452 USA

rw17{at}umail.umd.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION Material and methods Results and discussion Conclusions REFERENCES

 
Affordable technologies are needed to allow smallholder farmers to effectively use the phosphate rocks (PRs) found in many African countries. A pot study was conducted in Tanzania using two PRs (Panda and Minjingu) and two soils (an Alfisol and an Andisol) to assess responses of several types of crops to these PRs and to determine whether changes in crop responses to PR with time are due to crop sequence or merely contact time with soil. The Panda PR had no effect on growth or tissue P content in maize (Zea mays L.), bean (Phaseolus vulgaris L.), and pigeon pea [Cajanus cajan (L.) Millsp.], but it nearly tripled these parameters for cabbage (Brassica oleracea L.) on the Alfisol. Freshly applied Minjingu PR only slightly stimulated maize and pigeon pea, but nearly tripled cabbage yield in both soils. Previous crop had a greater effect than previously applied PR on second crop maize. Yields and P content of maize were always lowest following cabbage and highest following pigeon pea. Minjingu PR, but not Panda PR, had residual benefits on maize. Severe Mn toxicity occurred in all crops on the unamended Andisol. The calcareous Minjingu PR, but not the Panda PR, increased yields dramatically on the Andisol, partly by raising the soil pH in water enough (from 4.6 to 5.6) to alleviate Mn toxicity. Future P fertility work in Africa should pay adequate attention to the effects of crop sequences and soil biological properties.

Abbreviations: PR, phosphate rock • TSP, triple superphosphate • AM, arbuscular mycorrhizae

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Material and methods Results and discussion Conclusions REFERENCES

 
THE densely settled humid and subhumid regions in East Africa are suffering high rates of soil fertility depletion (Smaling et al., 1993). These areas are dominated by Alfisols, Oxisols, and Andisols—the productivity of which is commonly limited by deficiency of P (Weil et al., 1991; Jama et al., 1997). Because the majority of African farmers cannot afford to adequately fertilize their cropland with imported manufactured fertilizers, there is much interest in developing indigenous sources of nutrients, especially phosphate rock (PR) in sub-Sahara Africa (McClellan, 1989; Buresh et al., 1997).

Substantial deposits of PR in many sub-Sahara African countries are too low-grade to justify commercial production of refined fertilizer products, but could be mined and ground using labor intensive methods with a minimum of capital expenditure to produce inexpensive sources of potentially useful nutrients (Sheldon, 1982). However, since many of the PR deposits in sub-Sahara Africa are both low grade and low reactivity (Kurtanjek and Tandy, 1984), untreated PR often does not produce a plant growth response, even on acid soils (Lüken and Blümel, 1984; Ngatunga et al., 1989).

Tanzania has two principal deposits of PR that could potentially be used to boost agricultural production and enhance food security in Tanzania and neighboring countries (Harris, 1961). In northeast Tanzania, along the shores of Lake Manyara, a sedimentary PR occurs at Minjingu Hill. This material has been mined sporadically with and without various degrees of beneficiation since the early 1960s. After removing clay impurities, the Minjingu PR has a total P content of 115 to 160 g kg^-1 and a high content of carbonates. It is considered a reactive PR because of its high solubility in citric acid and moderate solubility in neutral ammonium citrate (Table 1) . Minjingu's location relative to Tanzanias transport infrastructure makes it exportable to Kenya and Uganda as well as available for use in the northern part of Tanzania.

An intriguing phenomenon reported in many longer-term experiments is a delay in crop response to PR such that the first crop grown after PR application gives little or no response, but the second and third crops respond well (Anderson, 1965, 1970; Bromfield et al., 1981; Gichuru and Sanchez, 1988; Kimbi, 1991; Haque and Lupwayi, 1998). For example, Bromfield et al. (1981) used Minjingu PR at 50 kg P ha^-1 and found a nonsignificant 29% increase in maize (Zea mays L.) yield compared with the control in the first harvest after PR application, but there was a significant 73% increase in the third harvest.

Understanding such a delayed response to PR would be important because taking this delay into account might lead to more efficient use of PR in cropping systems by planting a high-P demanding, staple crop later in the rotation after applying the PR. Also, since poor farmers look for an immediate return on their investment, this delayed action of PR could discourage them from using even relatively reactive PRs. It has generally been assumed that the increased effectiveness of PR with time in acid soils is due to a slow reaction by which soil acids dissolve a certain fraction of the P from the PR over a period of several months to years (Sanchez and Salinas, 1981; Rajan et al., 1996). However, observations from some pot studies (Mughogho and Weil, unpublished data, 1993) have indicated that the PR benefits a second crop much more than the first crop, even if the first crop was grown for only about 1 mo. An alternative hypothesis is that improved effectiveness over time is related to the action of the particular plants growing in the soils, and not simply to the time of soil–PR contact.

Most studies of crop response to direct application of PR in Africa have used maize as the test crop (e.g., Bromfield et al., 1981; Mnkeni et al., 1991; Buresh et al., 1997). However, research has shown that some plants are better able than others to use PR as their source of P (Bekele et al., 1983; Flach et al., 1987; Floor and Kimambo, 1989; van Diest, 1991; Haynes, 1992). Further, arbuscular mycorrhizae (AM) infection is well known to greatly improve P uptake by many plants in low P soils (Islam et al., 1980; Flach et al., 1987). Crop strategies for P uptake may be grouped into four types.

First, many monocots have extensive fibrous root systems, generally with AM infection, that allow these plants' roots to have very intimate contact with PR particles in the soil, and may give them some advantage in PR use (Menon et al., 1995). Second, despite a less extensive root system, annual legumes have also been shown to use PR efficiently (Aguilars and van Diest, 1981; Bekele et al., 1983). Because of their N₂ fixation ability, legumes generally use such low amounts of nitrate that the plants take up an excess of cations over anions. This imbalance leads to an excretion of H⁺ ions, which acidify the rhizosphere and release soluble P from Ca phosphates such as found in PR. Third, pigeon pea [Cajanus cajan (L.) Millsp.] has been shown to excrete piscidic acid, which specifically complexes with Fe to greatly increase the availability of Fe-bound P (Ae et al., 1990). Fourth, it is well known that crops in the Cruciferae family, such as rape (Brassica napus L.) and cabbage (Brassica oleracea L.) do not form mycorrizhae, but apparently can enhance P availability from calcium phosphates by the excretion of citric and malic acids, the formation of extensive fine root hairs, and the uptake of high amounts of Ca (van Diest, 1991; Hoffland, 1992).

This paper reports on a glasshouse pot study conducted in Tanzania mainly to investigate the delayed response phenomenon and to test the hypothesis that one crop may influence the effectiveness of PR for a following crop. Secondarily, this study was designed to provide information on the responses to Panda and Minjingu PRs by the four types of crops mentioned in the previous paragraph.

	Material and methods

TOP ABSTRACT INTRODUCTION Material and methods Results and discussion Conclusions REFERENCES

 
Maize was used as the test crop in a pot study because this is the main staple food crop in the region where the Panda and Minjingu PRs are found and because it is the crop that has usually shown a delayed response to PR. Maize was planted after PR had been in moist soil for 58 d, with either no crop (bare soil), or one of four crops that are important in the southern highlands of Tanzania and have differing mechanisms of P utilization in soil: (i) open-pollinated maize (`Kalimina-SR')—a crop with fine, fibrous, AM-dependent roots; (ii) bean (`Kabamina')—a legume that is AM-dependent and acidifies its rhizosphere; (iii) pigeon pea (unnamed, large seeded local land race)—a crop with special ability to solubilize iron phosphates; or (iv) cabbage (`Glory of Enkhuizen')—a crop that does not form AM.

Soils and Crops Used
The study was conducted in a glasshouse at the Ministry of Agriculture Research and Training Institute at Uyole, Tanzania (1750 m above sea level) using two soils (Table 2) . The Ihanda soil is representative of the Kandiustalfs formed from hillwash from mafic rocks, which are commonly used for smallholder food crops in areas with approximately 1200 mm annual rainfall. The Sasanda soil is representative of the soils formed in volcanic ash and pumice in the higher rainfall (1500 mm yr^-1) and elevation (>2000 m) areas of the southern highlands, which are commonly used for coffee (Coffea arabica L.) and smallholder food crops. Since both soils also had been used for several years in field trials of maize response to P, the optimum levels of P application were known for each (J. Kamasho, personal communication, 1994).

Soils and Phosphorus Amendments
For each soil, Ap horizon material was collected from an unfertilized maize field during the dry season. The soil was sieved to pass either a 12-mm screen (Ihanda) or a 3-mm screen (Sasanda) to homogenize the soil while preserving the natural structural aggregates. Four L of screened, air-dry soil (3.9 kg for Ihanda and 2.9 kg for Sasanda) was placed in each pot and mixed with the P amendment appropriate for the particular treatment 1 d before planting seeds. For Treatments 20, 21, and 22 the amendments were mixed into the upper 7.5 cm of soil 58 d later, just after the first crop was harvested. From previous field and pot studies (J.A. Kamasho, personal communication, 1994) it was determined that the optimum level of P application as TSP was 100 mg P kg^-1 of Ihanda soil and 400 mg P kg^-1 of Sasanda soil. Therefore all P amendments were applied to supply these rates of total P (390 mg P pot^-1 of Ihanda and 1160 mg P pot^-1 of Sasanda).

Initial Crops
On Day 0 the soils were moistened and pregerminated seeds (four per pot, later thinned to two plants per pot) were sown for pots with initial bean, maize, or pigeon pea crops. Two seedlings were transplanted (after rinsing the roots) into each pot meant to grow cabbage as the initial crop. Pots were hand weeded weekly and plants checked for insect pests three times per week. On Day 17 chlorfenvinphos [2-chloro-1-(2,4-dichlorophenyl) vinyl] was sprayed on all plants to control white fly.

A P-free nutrient solution was added to all pots to supply the following amounts of nutrients per pot: 65 mg N as ammonium sulfate, 100 mg K as KCl, 25 mg Mg as Mg(NO₃)₂, 25 mg Ca as CaSO₄, 5 mg Cu as CuSO₄, and 5 mg Zn as ZnSO₄. An additional 100 mg N was applied as NH₄NO₃ on Day 28 for pots with nonlegume crops and on Days 65 and 85 for the second crop maize in all pots. Just before planting the first crop all pots were brought to -10 kPa water potential (0.35 L L^-1 for Ihanda and 0.42 for Sasanda) with distilled water (8.5 µS m^-1). Thereafter, this water content was reestablished every second day by bringing each pot up to its original weight with distilled water. Air temperature in the shade ranged from 20 to 35°C. Soil temperature at 5 cm depth ranged from 17 to 30°C. No supplementation of natural sunlight was used. On Day 58, all aboveground plant parts were harvested by cutting off the shoot 0.5 cm above the soil line. Fresh and dry (48 h at 65°C) weights of the tissue were recorded.

Second Crop
The day after harvesting the first crop, fresh P amendments were mixed into the upper 7.5 cm of soil for Treatments 20, 21, and 22. On Day 62, four pregerminated maize seeds were planted in each pot. This second crop maize served as the test crop for the effects of previous crop, initial soil amendments, and fresh soil amendments.

On Day 102 (40 d after sowing), all maize plants were harvested. Shoots were harvested by cutting of 0.5 cm above the soil line and analyzed for fresh weight, dry weight, total P, and total Mn. Soil was sampled from selected treatments. Roots from Treatments 11 to 15 and 4 were harvested by soaking the root–soil mass in water and then washing over a 2-mm screen. The clean roots were then blotted dry, weighed, dried, and reweighed.

Soil and Plant Analyses
The dried plant tissue was ground in a Wiley mill (<1 mm) and digested in perchloric acid. The digest was analyzed for total P by the Murphy–Riley colorimetric method and for total Mn (in Treatments 2–5, 7–10, and 12–15 for initial crops and in Treatments 1, 8, 9, 12–15, and 22 for second crop maize) by atomic absorption spectrometry. Samples of soil were obtained from selected treatments immediately after harvest of the first and second crops by removing two 1 cm by 8 cm cores. The soil cores were air-dried, sieved to <2 mm, and analyzed for Bray 1 extractable P, electrical conductivity of a 2:1 water/soil slurry (EC), and pH in a 2:1 water/soil slurry (pH_w) and in a 2:1 slurry using 1 M KCl solution (pH_KCl).

As an exploratory measure, roots of the first crop for Treatments 2, 3, and 17 for both soils in one block were collected from the soil cores before drying and cleared and stained for detection of AM fungal infection by the method of Koske and Gemma (1989). Mycorrhizal infection of the prepared roots was quantified by determining the total and infected root lengths in each sample by the grid intersection method under 50 to 100x magnification. The results were then expressed as percentage of total root length that was infected. For the second maize crop, roots in Treatments 11 to 15 and 4 were washed and blotted, and then a small subsample of fine roots (<2% of the total root mass) was removed and prepared for mycorrhizal analysis as described above.

	Results and discussion

TOP ABSTRACT INTRODUCTION Material and methods Results and discussion Conclusions REFERENCES

 
Initial Cropping Period
Levels of Bray 1 P were reduced by cropping from their low initial levels (Table 2) in both Ihanda and Sasanda soils (Table 5) , with those in the Sasanda soil as low as 0.05 mg kg^-1. Amendment with Minjingu PR maintained the Bray 1 P levels near the original levels, but soil amended with Panda PR was not significantly higher in Bray 1 P than the unamended soils.

View this table: [in this window] [in a new window]   Table 5 Influence of P amendments on Bray 1 extractable P and soil pH for two Tanzania Southern Highlands soils after a 58-d initial crop growing period.

View larger version (34K): [in this window] [in a new window]   Fig. 1 Effect of P source on shoot dry matter yield of four crops grown for 58 d in Ihanda and Sasanda soils immediately after P application. Different letters within a graph indicate statistically significant differences at P = 0.05 by Fishers LSD. For the unreplicated TSP treatment, least squares means are shown

Manganese Toxicity
Toxicity of Mn was first suspected in a preliminary unreplicated pot trial using four soils. The pots were accidentally waterlogged for 5 d, after which bean plants growing in the Sasanda soil and another soil of volcanic origin became stunted and showed mottled interveinal and marginal chlorosis on the younger trifoliate leaves, and small (<0.5 mm) black to purple necrotic spots on the unifoliate leaves. Once the soil was allowed to dry down, the bean plants appeared to partially recover. This behavior is consistent with the occurrence of Mn toxicity (Weil et al., 1997; Flor and Thung, 1989), which has been reported for bean on volcanic soils (Andisols) in South America. The Sasanda soil is an Andisol very high in total Mn (Table 2).

Scientists and extension workers in Tanzania have reported difficulties growing bean crops on Sasanda and similar soils formed in volcanic materials (M. Madata, personal communication, 1994). In the area where the Sasanda soil was obtained for the present study, discussions with farmers revealed that sweet potato [Ipomoea batatas (L.) Lam.] when planted on 1-m high soil ridges was the only consistently successful crop. Since sweet potato has a very high tolerance for Mn and tall ridges would assure that the soil did not become waterlogged during wet periods, these field observations are consistent with the existence of a Mn toxicity problem on the Sasanda soil. Furthermore, the pH_w of the Sasanda soil was 5.3 (Table 2), which is sufficiently low to promote excessive availability of Mn, but high enough that toxicity of Al is unlikely to be a problem (Weil et al., 1997). The high level of organic C in the Sasanda soil could also be expected to enhance toxicity of Mn, but not of Al, by promoting reduced conditions when the soil is wet.

Symptoms of Mn toxicity occurred in some treatments in the pot study, although never as severely as in the waterlogged preliminary pots. Symptoms suggestive of Mn toxicity included leaf cupping and stunting of cabbage and mottled chlorosis on younger leaves of bean and pigeon pea. The close relationship between the Mn concentration of initial crop tissue and the pH of the soil is illustrated in Fig. 2a for bean and cabbage. The relationship is shown for pH_KCl, but not for the pH_w, because the latter is subject to fluctuation with soluble salts such as accumulated from unused fertilizer in some of the pots. For soils with pH_KCl below 4.5, Mn concentrations in bean and cabbage tissue were >500 mg kg^-1, a level reported to be toxic (Flor and Thung, 1989). None of the pots with Ihanda soil had pH_KCl values this low, but pots with unamended or Panda PR amended Sasanda soil consistently had pH_KCl values below 4.5 and tissue Mn concentrations as high as 1100 mg kg^-1. Sasanda soil without Minjingu PR also produced up to 400 mg kg^-1 Mn in maize tissue and up to 510 mg kg^-1 in pigeon pea tissue (data not shown).

View larger version (31K): [in this window] [in a new window]   Fig. 2 The relationship between soil pHKCl at harvest and tissue Mn concentration for (a) initial crop bean and cabbage and (b) second crop maize in two soils receiving one of two phosphate rocks (PR) or no added P

Second Cropping Period
A very striking effect was the much lower mean yield for the second crop maize in the Sasanda (2.4 g pot^-1) than in the Ihanda soil (5.5 g pot^-1). In the Ihanda soil (where Mn toxicity was not a problem) a previous crop of bean, maize, or pigeon pea markedly increased the second maize shoot dry matter compared with that of the no P added, no previous crop control. This was true regardless of whether Panda PR, Minjingu PR, or no P amendment had been previously applied (Table 3). However, where cabbage was the previous crop (with or without a previous P amendment), maize shoot dry matter was not different from the no-P added bare soil control.

In the Sasanda soil, none of the previous crops when combined with either Panda PR or no-P amendment produced a significant increase in maize dry matter compared with the no P added, bare soil control. This lack of response to a previous bean, maize, or pigeon pea crop was likely due to the inability of the Panda PR to alleviate Mn toxicity, as discussed above. Partially because of its ability to ameliorate Mn toxicity, Minjingu PR was dramatically more effective than Panda PR in the Sasanda soil (Table 3). Minjingu PR applied to Sasanda soil in combination with a non-cabbage crop in the initial cropping period gave second crop maize dry matter yields as high as those obtained in the Ihanda soil. When the Minjingu PR was applied in combination with either cabbage or no crop during the initial cropping period, the second crop maize yield was comparable with the no-P control. The combination of Minjingu PR with a previous crop of pigeon pea produced as much dry matter as did freshly applied TSP, and significantly more dry matter than any other PR treatment on the Ihanda soil.

As with the initial crop, the uptake of Mn by the second maize crop was related to the soil pH_KCl measured at harvest (Fig. 2b). Tissue Mn levels >225 mg kg^-1 were associated with very low maize dry matter production (Fig. 3) , a result that agrees well with the 200 mg kg^-1 level reported as toxic for young maize plants by Reuter and Robinson (1986). All treatments on the Ihanda soil had pH_KCl values above 4.6 and maize Mn tissue levels below 100 mg kg^-1, however, on Sasanda soil without Minjingu PR the pH_KCl was consistently below 4.5 and the Mn in the maize tissue was in the toxic range (Fig. 2b).

View larger version (34K): [in this window] [in a new window]   Fig. 3 The relationship between shoot dry matter yield and tissue Mn concentration for second crop maize in two soils previously treated with no added P, or with one of two phosphate rocks (PR)

The large difference in maize growth between the two soils was also apparent when P amendments were applied to the second maize crop (Table 4). Neither Panda nor Minjingu PR, when freshly applied to the second crop maize following a 58-d bare fallow, had any effect on maize dry matter or P content (Table 4) compared with no P amendment. Freshly applied TSP produced about three times the dry matter and nearly four times the P content as did either of the freshly applied PRs.

Maize responses to a previous crop in conjunction with application of Minjingu PR are further elucidated by data on root dry matter and shoot/root ratio for second crop maize in the selected treatments for which roots were collected (Fig. 4) . The shoot/root ratio for maize was low where no P was applied to pigeon pea first crop—the only no P treatment for which root weights were measured. Root growth where Minjingu PR was incubated in bare soil for 58 d was not significantly different from the no P treatment. But a previous bean, pigeon pea, or maize crop in combination with Minjingu PR significantly increased both root dry matter and shoot/root ratio of the second crop maize compared with pigeon pea with no P added. Thus, the residual effect of Minjingu PR on maize following pigeon pea, bean, or maize was to enhance shoot growth more than root growth. The shoot/root ratio and root dry weight for maize with residual Minjingu PR following cabbage did not differ from corresponding values obtained with bare soil during the initial period.

View larger version (28K): [in this window] [in a new window]   Fig. 4 Influence of previous cropping history (bean, maize, pigeon pea, or bare soil) on second crop maize root dry matter (g pot-1) and shoot/root ratio. All treatments, except pigeon pea with no P added (Pig. Pea, P0) had Minjingu phosphate rock applied at the beginning of the initial cropping period. Error bars indicate Fisher's LSD (P < 0.05) for the respective variable

For the second crop maize, pots amended with Minjingu PR were analyzed for AM fungi infection on roots. Infection was still significantly lower in the Sasanda (27% of root length) than in the Ihanda soil (37% of root length). Because there was no significant soil x crop interaction, AM infection data were averaged across both soils to compare treatments (Fig. 5) . A previous crop of cabbage depressed second crop maize AM infection. Residual Minjingu PR tended to enhance AM infection in the second crop maize (as compared with the no P treatment following pigeon pea). One could speculate that although AM fungi helped the maize take up P from the Minjingu PR, the P availability from this PR was low enough to not inhibit AM infection. Thus, the limited evidence available suggests the effects of previous crop and PR application may have been partially due to differences in AM infection, but that AM infection by itself cannot explain all the effects.

View larger version (55K): [in this window] [in a new window]   Fig. 5 Arbuscular mycorrhizal infection (infected root length as a % of total root length examined) on second crop maize as influenced by previous crop in Sasanda soil. All treatments, except pigeon pea without P amendment (Pig. Pea, P0) had Minjingu phosphate rock applied at the beginning of the initial cropping period. Error bar indicates Fisher's LSD (P < 0.05)

	Conclusions

TOP ABSTRACT INTRODUCTION Material and methods Results and discussion Conclusions REFERENCES

This is the first work to show a dramatic response of cabbage to the otherwise ineffective Panda PR, indicating that cabbage (and probably other Brassica cash crops grown in the region) may be key to obtaining an economic benefit from certain local PR deposits. This result should be followed up by using Brassica crops to test Panda and other unreactive, Fe-rich, igneous PRs in both pots and the field.

The present study provides clear evidence that one crop treated with unacidulated PRs may substantially influence the residual effectiveness of the PR for following crops. Where Mn toxicity was not a problem, second crop maize had significantly greater dry matter and P content when bean, pigeon pea, or maize was previously grown in the soil, with or without application of PR. Although inadequate to fully explain the effect of the previous crop, a role for mycorrhizae is suggested by the limited data on AM root infection, and by the fact that cabbage, a nonhost plant for AM fungi, consistently failed to have a positive effect comparable to that of the other crops. Certainly, with the adequate fertility applied to all plots and with previous crop maize showing the same effect as the legumes, the previous crop effect is unlikely to be associated with enhanced availability of N or other nutrients (except P).

The discovery that Sasanda soil produces Mn toxicity in four crops is of great importance to the management of farming systems on this and similar soils. This study should therefore be followed up by field and pot studies investigating the occurrence of Mn toxicity in similar acid Andisols from high rainfall areas of East Africa. Use of carbonate-rich, reactive PRs such as Minjingu may have special value in combining P supply with Mn toxicity alleviation. Also, plant breeders could make Mn tolerance a goal in developing varieties of bean and other Mn sensitive crops. This work has already begun in western Tanzania.

Finally, the results reported here suggest that future P fertility work in Africa should take a whole cropping system approach. Adequate attention should be paid to the effects of crop sequences and soil chemical and biological properties.

	ACKNOWLEDGMENTS

 
This work was carried out while the author was at the Tanzania Ministry of Agriculture Research and Training Center in Uyole as a Senior African Regional Research Fellow under the Fulbright-Hayes Program in 1994. He wishes to express his appreciation for the support provided by both of these institutions. Joseph Kamasho generously shared his experience with Panda PR and assisted in obtaining the soils for the study. Roden Mwamila and Rashid Kapenzi provided able assistance with the laboratory analyses. The author also thanks Roland Buresh of the International Rice Research Institute in the Philippines, Michael Adetunji of Abeokuta University of Agriculture in Nigeria, and three anonymous reviewers for their insightful suggestions during the preparation of the manuscript.

Received for publication December 13, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION Material and methods Results and discussion Conclusions REFERENCES

 

• Ae N., Arihara J., Okada K., Yoshihara T., Johansen C. Phosphorus uptake by pigeon pea and its role in cropping systems of Indian subcontinent. Science (Washington, DC) 1990;248:477-480.[Abstract/Free Full Text]
• Aguilars A., van Diest A. Rock phosphate mobilization induced by the alkaline uptake pattern of legumes utilizing symbiotically fixed nitrogen. Plant Soil 1981;61:27-42.
• Anderson G.D. Estimation of the potential productivity of some pastures in the Mbulu Highlands of Tanganyika by application of fertilizers. East Afr. Agric. For. J. 1965;30:206-210.
• Anderson G.D. Fertility studies on a sandy loam in semi-arid Tanzania: II. Effects of phosphorus, potassium and lime on yields of groundnuts. Exp. Agric. 1970;6:213-222.
• Bekele T., Cino B.J., Ehlert P.A.I., Der Maas A.A., van Diest A. An evaluation of plant-borne factors promoting the solubilization of alkaline rock phosphates. Plant Soil 1983;75:361-378.
• Bromfield A.R., Hancock L.R., Debenham D.F. Effects of ground phosphate and elemental S on yield and P uptake of maize in western Kenya. Exp. Agric. 1981;17:383-387.
• Buresh R.J., Smithson P.C., Hellums D.T. Building soil phosphorus capital in Africa. In: Buresh R.J., et al. , ed. Replenishing soil fertility in Africa. Madison, WI: SSSA, 1997:111-149 SSSA Spec. Publ. 51..
• Chesworth W., Semoka J.M.R., van Straaten P., Mnkeni P.N.S., Kamasho J.A., Mchihiyo E.P. Tanzania–Canada agroecology project: Report on completion of the first phase. Guelph, ON, Canada: Univ. of Guelph, 1988.
• Flach E.N., Quak W., van Diest A. A comparison of the rock phosphate-mobilizing capacities of various crop species. Trop. Agric. 1987;64:347-352.
• Floor, J., and E.I. Kimambo. 1989. Considerations on the use of Minjingu rock phosphate as a direct phosphate fertilizer. p. 23–33. In F.B.S. Kaihura, et al. (ed.) Proc. of the Phosphate Meeting, Mkonge Hotel, Tanga, Tanzania. 1–3 Aug. 1988. Misc. Publ. M9. National Soil Service Agric. Res. Institute-Mlingano, Tanga, Tanzania.
• Flor C.A., Thung M.T. Nutritional disorders. In: Schwartz H., Pastor-Corrales M.A., eds. Beans production problems in the tropics. Cali, Colombia: CIAT, 1989:69-73.
• Gichuru M.P., Sanchez P.A. Phosphate rock in tilled and no-till low-input systems in the humid tropics. Agron. J. 1988;80:943-947.[Abstract/Free Full Text]
• Haque I., Lupwayi N.Z. Agronomic effectiveness and residual effects of Egyptian and Togo phosphate rocks on clover production. J. Plant Nutr. 1998;21:2013-2033.
• Harris, J.F. 1961. Summary of the geology of Tanganyika. p. 113–115. Memoir 1. Part 4. Economic geology. Government Printer, Dar Es Salaam, Tanzania.
• Haynes R.J. Relative ability of a range of crop species to use phosphate rock and monocalcium phosphate when grown in soil. J. Sci. Food Agric. 1992;74:1-7.
• Hoffland E. Quantitative evaluation of the role of organic acid exudation in the mobilization of rock phosphate by rape. Plant Soil 1992;140:279-289.
• Islam R., Ayanaba A., Sanders F.E. Response of cowpea (Vigna unguiculata) to inoculation with VA-mycorrhizal fungi and to rock phosphate fertilization in some unsterilized Nigerian soils. Plant Soil 1980;54:107-117.
• Jama B., Swinkels R.A., Buresh R.J. Agronomic and economic evaluation of organic and inorganic sources of phosphorus in western Kenya. Agron. J. 1997;89:597-604.[Abstract/Free Full Text]
• Kamasho J.A., Mayona C.M., Rees D.J. The use of organic and geologic materials for soil fertility maintenance in the Southern Highlands of Tanzania. In: Ekpere J.A., et al. , ed. Proc. of the Conf. on Agricultural Research, Training and Technology Transfer in the Southern Highlands of Tanzania: Past Achievements and Future Prospects, Mbeya, Tanzania. 5–9 Oct. 1992. Mbeya, Tanzania: Uyole Agric. Centre, 1992:265-279.
• Kimbi G.G. Evaluation of Minjingu PR products as sources of phosphorus for maize and sorghum in some Morogorro and Hai soils. Morogoro, Tanzania: M.S. thesis. Sokoine Univ. of Agric, 1991.
• Koske R.E., Gemma J.N. A modified procedure for staining roots to detect VA mycorrhizas. Mycol. Res. 1989;92:486-488.[Web of Science]
• Kurtanjek, M.P., and B.C. Tandy. 1984. The potential development of phosphate rock resources in tropical Africa. p. 364–376. In Fertilizer 83 Int. Conf. Proc. of British Sulphur Corp., London, UK. 13–16 Nov. 1983. British Sulphur Project Development, London, UK.
• Lüken, H., and G. Blümel. 1984. Phosphate rock as a direct application fertilizer. Geologisches Jahrbuch, Rehe F Bodenkunde, Heft 17. E. Schweizerbart'sche Verlagsbuchhandlung, Joahannesstrasse 3 A, D-7000 Stuttgart 1, Germany.
• McClellan, G.H. 1989. Use of east and southeast African indigenous agrominerals for fertilizer production. p. 1–4. In S.J. Van Kauwenbergh, et al. (ed.) Proc. of East and Southeast African Geology Workshop on Fertilizer Minerals, Lusaka, Zambia. 8–10 Dec. 1987. Int. Fertilizer Dev. Center (IFDC), Muscle Shoals, AL.
• Menon R.G., Chien S.H., Hellums D.T. Iron and aluminum oxide influence on agronomic effectiveness of modified phosphate rocks. Soil Sci. Soc. Am. J. 1995;59:1762-1767.[Abstract/Free Full Text]
• Mnkeni P.N.S., Semoka J.M.R., Buganga J.B.B.S. Agronomic effectiveness of Minjingu phosphate rock as a source of phosphorus for maize in four soils of Morogoro district, Tanzania. Zimbabwe J. Agric. Res. 1991;30:27-37.
• National Soil Survey Center. Soil survey laboratory methods manual. Soil Survey Invest. Rep. 42. Lincoln, NE: USDA, Natural Resour. Conserv. Service, 1996.
• Ngatunga, E.N.L., K.L. Haule, and J. Deckers. 1989. A comparison of the effects of application of Minjingu PR and triple super phosphate on the yield of sorghum on acid sandy soils of Naliendele in South East Tanzania. p. 31–35. In F.B.S. Kaihura, et al. (ed.) Proc. of the Phosphate Meeting, Mkonge Hotel, Tanga, Tanzania. 1–3 Aug. 1988. Misc. Publ. M9. National Soil Service Agric. Res. Institute-Mlingano, Tanga, Tanzania.
• Norvell W.A. Comparison of chelating agents as extractants for metals in diverse soil materials. Soil Sci. Soc. Am. J. 1984;48:1285-1291.[Abstract/Free Full Text]
• Rajan S.S., Watkinson J.H., Sinclair A.G. Phosphate rocks for direct application to soils. Adv. Agron. 1996;57:77-159.
• Reuter D.J., Robinson J.B. Plant analysis: an interpretation manual. Melbourne, Australia: Inkata Press, 1986.
• Sanchez P.A., Salinas J.G. Low-input technology for managing Oxisols and Ultisols in tropical America. Adv. Agron. 1981;34:280-406.
• Sheldon R.P. Phosphate rock. Sci. Am. 1982;246:45-51.
• Smaling E.M.A., Stoorvogel J.J., Windmeijer P.N. Calculating soil nutrient balances in Africa at different scales: II. District scale. Fert. Res. 1993;35:237-250.
• van Diest A. Various forms of root action influencing the availability of soil and fertilizer phosphorus. In: McMichael B.L., Persson H., eds. Plant roots and their environment. Amsterdam: Elsevier Science Publ, 1991:164-170.
• van Straaten P., Semoka J.M.R., Kamasho J.A.M., Mchihiyo E.P., Mnkeni P.N.S., Chesworth W. Report on the status of the Tanzania–Canada agrogeology project, phase II (1988–1991). Guelph, ON, Canada: Univ. of Guelph, 1992.
• Weil R., Mukurumbira L., Butai P. An approach to improving site-specific fertilizer recommendations in peasant agriculture. Trop. Agric. 1991;68:186-190.
• Weil R.R., Foy C.D., Coradetti C.A. Influence of soil moisture regimes on subsequent soil manganese availability and toxicity in two cotton genotypes. Agron. J. 1997;89:1-8.[Abstract/Free Full Text]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Weil, R. R. Search for Related Content PubMed Articles by Weil, R. R. Agricola Articles by Weil, R. R.