Rev. Fac. Agron. (Maracay) 17:67-95. 1991.

Use of natural and modified phosphate rocks on annual, perennial and forestry crops in Brazil

Alfredo Scheid Lopes¹; Wenceslau José Goedert²; Luiz Roberto Guimarâes Guilherme³

COMPENDIO

La gran mayoría de los suelos brasileños, principalmente aquellos ubicados en la región del cerrado, presentan niveles extremadamente bajos de fósforo total y aprovechable para los cultivos. Este problema es magnificado por la alta capacidad de fijación de fósforo de estos suelos lo cual generalmente esta asociado con el tipo y porcentaje de arcilla. Como resultado altas dosis de fertilizantes fosfatados se hacen necesarias a fin de obtener rendimientos adecuados cuando estos suelos son incorporados en la producción. La necesidad de altas dosis de fertilizantes, el costo de esta práctica y el hecho de que la ocurrencia de depósitos de rocas fosfóricas en Brasil no es abundante, justifica todos los esfuerzos para investigar una más alta eficiencia agronómica de los fertilizantes fosforados. En este sentido se condujeron experimentos con cultivos anuales, perennes y forestales con los siguientes resultados: a) en los cultivos anuales la respuesta a los fertilizantes fosforados es proporcional al porcentaje de P₂O₅ soluble en relación al P₂O₅ total; b) la roca fosfórica brasilera tiene baja solubilidad y baja eficiencia en cultivos anuales; c) rocas parcialmente aciduladas pueden ser adecuadas para pastos y cultivos permanentes; d) los termofosfatos representan una excelente opción agronómica para la mayoría de los suelos ácidos del cerrado brasileño.

Palabras clave: Brasil, rocas fosfóricas, eficiencia agronómica, Coffea arabica, Eucalyptus, Brachiaria, Zea mays, Andropogon, abonos fosfatados

ABSTRACT

Key words: Brazil, phosphate rocks, agronomic efficiency, Coffea arabica, Eucalyptus, Brachiaria, Zea mays, Andropogon, phosphate fertilizer.

INTRODUCTION

The great majority of Brazilian soils, mainly those located in the "cerrado" region, present extremely low levels of total and available phosphorus for cultivated crops. This problem is enhanced by the high phosphorus "fixation" capacity of these soils which is in general associated with type and percentage of clay components. As a result, high rates of phosphate fertilizers are necessary to be applied in order to achieve adequate crop production when these soils are incorporated in the production (Goedert et al., 1986 b).

The necessity of high rates of phosphate fertilizers, the costs of this practice, and the fact that the ocurrence of natural phosphate rocks deposits in Brazil is not abundant, justify all efforts to search a higher agronomic efficiency of phosphate fertilizers. These efforts are not only agronomically sound but above all strategically necessary.

The search for a higher recuperation of phosphorus from Brazilian rock phosphate deposits involves several steps including mining, industrialization, transportation, storage, and finally the agronomic management techniques to achieve higher absorption efficiency in relation to the rate of applied phosphorus.

The agronomic science has already identified a series of factors that suggests several measures to increase agronomic efficiency of phosphate fertilizers:

1. reduction in phosphorus fixation capacity through adequate liming;
2. combination of most adequate rates and methods of application, according to the crops;
3. rational utilization of sources available in the region;
4. utilization of more efficient plant species in terms of utilization of applied phosphorus;
5. utilization of more efficient crop production systems including crop rotation, green manure, etc.

A serious limitation to evaluate agronomic efficiency of phosphate fertilizers in Brazil is the lack of long term field experiments, that would allow to evaluate long term residual effects. Another limitation is that economic evaluation together with agronomic evaluation are not common to be conducted at the same time.

In this paper the following concept relative to agronomic efficiency will be used, according to Goedert et al. (1987):

AEI % = Agronomic Efficiency Index

The objective of this paper is to discuss the agronomic efficiency of phosphate fertilizers for annual crops, pasture, perennial crops and reforestation in Brazil, with special emphasis to the Brazilian "cerrado" region based upon soil, climate, and crop characteristics and upon types of fertilizers produced and commercialized in this region.

The "cerrado" region

General characteristics. The area under "cerrado" vegetation in Central Brazil occupies 2 million square kilometers or 23% of the country. According to Cochrane et al. (1985), Oxisols cover more than 50% of the area; the remainder consists mostly of Ultisols, Entisols and Inceptisols. Red-Yellow Latosols and Dark Red Latosols, most of them included in the suborder Ustox, according to U.S. Soil Taxonomy, are preferred for intensive agriculture (Goedert, 1987). Most of these soils are loamy or clayey, but approximately 20% are sandy, deserving special management practices for continuous cropping. The clay fraction is dominated by kaolinite, gibbsite and iron oxides (Hughes, 1980). The predominance of these low- activity clays is responsible for most of the soil behavior, especially in terms of ion exchange, water-holding capacity, and phosphorus "fixation" characteristics (Goedert, 1987).

Most of these soils are very acid and have low availability of primary and secondary macronutrients and also present deficiency of zinc, boron and copper. They have high aluminum saturation and phosphorus "fixation" capacity.

Phosphorus is the most deficient plant nutrient in these soils. A survey conducted by Lopes (1975) and summaryzed by Lopes and Cox (1977) indicated that 92% of the topsoil samples contained less than 2 µg P ml^-1, as extracted by Mehlich-1 procedure. The very low available phosphorus is probably one of the most critical limitations for crop growth in these soils. In addition, correction of this problem is more difficult than the normal ones due to their extremely high phosphorus "fixation capacity" (Leal and Velloso, 1973). The amount of phosphorus required to obtain 0.1µg P ml^-1 in equilibrium solution has been reported to vary from 110 to 405 µg P ml^-1 (Le Mare, 1982; Smyth and Sanchez, 1982; Le Mare and Goedert, 1984).

Besides these chemical problems there are other production limitations: a) commom dry season (5 to 6 months - april to september); b) dry spells during the rainy season ("veranicos"), generally associated with high evapotranspiration rates; c) low water-holding capacity; d) limited rooting depth of many crops as a function of aluminum toxicity and/or calcium deficiency in subsurface layers.

These points emphasize the need for appropriate management technology to increase the probability of success when incorporating these soils to the crop production process, mainly under rain-dependent conditions.

However a series of positive points of this area should also be mentioned. Adamoli et al. (1986) estimate that about 73% of the "cerrado" area range from 300 to 900 meters elevation. The average annual temperature ranges from 22°C in the south to 27°C in the north, generally associated with changes in elevation. The average monthly temperature is almost constant throughout the year. Daily solar radiation input is high over the whole area, around 475 to 500 langleys (Adamoli et al., 1986). Temperature and the energy characteristics, therefore, are favorable for growth of almost any commercial crop during the whole year (Goedert, 1987). Total annual rainfall varies from 900 to 2,000mm distributed by distinct wet dry seasons.

A combination of gentle slopes (commonly less than 3%) with very good soil aggregate stability and deepness/well drained soil conditions favors intensive agricultural mechanization. These conditions indicate that approximately 50% of the region is covered by arable lands (Goedert, 1989).

In spite of previously mentioned problems, a burst in agricultural development has taken place in this area during the last fifteen years, mainly involving grain crops, sugarcane, improved pasture and coffee. The average yield of some of these crops exceed the national one and in total they contribute with more than 20 million metric tons of grains, and also reasonable percentage of the coffee and sugar cane Brazil's total agricultural production.

It is important to stress that until the 1970's, this region was considered marginal for crop production. However, the management technology developed for these soils and already practiced by the farmers have completely changed this picture. The estimated potencial for the "cerrado" region production is 190 million t of grains (150 dryland and 40 irrigated), 9 million t of beef cattle and 300 million m³ of forests (Goedert, 1989).

Rock phosphate deposits. The rock phosphate is the most important source of phosphorus and its compounds in the world. Estimated worlwide reserves indicate a percentage participation of only 2.6% for Brazil (Table 1). It should be stressed that from 1975 to 1985, mineral prospection indicated 240% increase in the Brazilian rock phosphate deposits. Besides that, another potencial areas for rock phosphate occurrence, allow a prospective increase in the known reserves in the country (Giulietti et al. 1987).

Concerning phosphate production, the percent increase in Brazil from 1975 to 1986, was the hightest (1207%) compared to others major producers countries (Morocco 52%; Tunísia 65%; USA 9%; China 188%; Jordan 362% and USSR 37%) (FAO, 1988). This observed growth was a result of discovering new deposits and mainly of the development of adapted technology to process magmatic origin phosphate rocks, as is the case of most Brazilian deposits. These magmatic origin deposits, besides the fact that present lower in situ P₂O₅ concentration, have also much more complex mineralogical aspects (iron, aluminum and magnesium impurities), than marine sedimentary deposits (Lobo, 1986).

Table 1. World phosphate rock deposits. (Phosphate Rock Statistics, IFA, cited by Berger et al., 1987).

Country	Deposits
-	million t	%
Morroco	40000	27.3
USA	35600	24.3
China	15000	10.3
USSR	7125 4.8	-
Brazil	3840	2.6
Others	44742	30.7

World Total 146,307 100.0

Figure 1 shows a distribution of phosphate deposits in Brazil, with emphasis on the "cerrado" region. It should be stressed that roughly 87% of country's capability to produce rock phosphate is located in this area (ANDA, 1987).

Fig. 1. Location of phosphate rock doposits in Brazil; 87% of the production capacity is located in the "cerrado" region.
Source: Adapted from Goedert (1989).

Fig. 2. "Ideal" utilization of brazilian phosphate rock deposit.
Source: Giulietti et al. 1987.

Table 2. Distribution of phosphorus consumption in Brazil (%) (Lobo 1986).

Phosphate fertilizers	-	-	98
Wet route	-	90	-
mmonium phosphates	28	-	-
Complex fertilizer	18	-	-
Double and Triple superphosphates	18	-	-
Single superphosphates	24	-	-
Partially acidulated phosphates	2	-	-
Thermic route	-	2	-
Thermophosphate	-	2	-
Direct soil application	-	6	-
Other uses	-	-	2
Total	-	-	100

Agronomic efficiency of phosphate fertilizers in the "cerrado" region

Agronomic efficiency for annual crops. One of the first studies concerning evaluation of agronomic efficiency of phosphate fertilizers in Brazil was carried out by Alvarez et al. (1965) with sugar cane. The results of this study, summarized in Figure 3, indicated that the natural rock phosphates under test (Araxá, Olinda and Alvorada) presented considerable lower yields when compared to triple superphosphate and magnesium thermophosphate.During the last decade, with the discovery and exploitation of new phosphate rock deposits located in the "cerrado" region a great number of research work was developed to evaluate these new phosphate sources for direct application or solubilization process concerning annual and perennial crops. One of these research programs was initiated in 1975 at the Cerrado Research Center (EMBRAPA), at Planaltina-DF and the results are summarized in Table 3. One of the main conclusions of this work was the low agronomic efficiency index (AEI) of the Brazilian natural rock phosphates tested for annual crops. Even considering five years under annual crops the AEI was below 50% when compared with Gafsa rock phosphate, magnesium thermophosphate and triple superphosphate. However for andropogon grass, cultivated for three years after the annual crops, the AEI was considerably higher.

Table 3. Agronomic efficiency of phosphates in a Dark Red Latosol, clayey, calculated from P absorption data during five years with annual crops followed by three years with Andropogon pasture. Rate of P₂O₅ broadcasted in the first year = 200 kg ha^-1 (Goedert and Lobato, 1984).

Phosphate	Agronomic Efficiency Index (AEI)
-	Annual crops	Andropogon	Total
-	%
Patos	45	81	59
Araxá	27	69	41
Abaeté	21	86	43
Catalâo	8	36	17
Gafsa	93	110	104
Thermophosphate	92	142	113
Triple Superphosphate	100	100	100

Figure 3. Response of sugar cane to sources and rates of phosphate fertilizers.
Source: Alvarez et al. 1965.

More recent studies as summarized by Goedert and Dias Filho (1986), have confirmed that Brazilian rock phosphates present low AEI for direct application in annual crops, mainly because these products are not capable to supply phosphorus at the adequate velocity and intensity required for these crops.

This behavior is often complicated because most of the Brazilian soils are extremely acid, with high aluminum toxicity and adequate rates of lime is the first management practice to achieve adequate yields of non-tolerant crops. Several papers have shown even more reduction of AEI for Brazilian rock phosphates after adequate liming (Braga, 1969; Tanaka, 1978; EMBRAPA, 1980; Silva et al., 1983).

Some technics aiming to increase the solubility of the Brazilian natural rock phosphates such as theirs granulation with TSP had no positive effect in terms of increasing phosphorus availability from the natural rock phosphates (Nachtigall et al., 1989). Similar results were obtained with soybeans and wheat by using gafsa phosphate (Fioreze et al., 1990). On the order hand melted mixtures of rock phosphates and iron slag have been studied as a potencial alternative of phosphate fertilizer. Greenhouse experiments developed by Firme et al. (1990) demonstrated that mistures with 40% of Araxá rock phosphate can be considered phosphate fertilizers and also soil acidity amenders.

Taking into account the mineralogical and chemical characteristics and the agronomic behavior of Brazilian natural rock phosphates, great effort has been developed searching for better alternatives aiming to improve the low AEI of the simple ground product. Basically four alternatives have been deeply studied:

1. Traditional soluble phosphates.
2. Phosphates partially acidulated with sulphuric acid.
3. Phosphates solubilized by other chemical process.
4. Thermophosphates.

Most of the phosphate fertilizers produced in Brazil are the superphosphates (single and triple) and ammonium phosphates (MAP and DAP). These water soluble phosphorus sources are considered excellent in terms of agronomic efficiency but present some limitations concerning industrial aspects: a) relatively complex technology; b) necessity of sulphur; and c) necessity of concentrated phosphates with low levels of contaminants (R203). The last points are extremely important in Brazil, since more than 80% of sulphur consumption is imported, and approximately 40% of the extracted P₂O₅ of the mines is lost during ore concentration and impurities elimination (Lobo and Silva, 1984).

Production of partially acidulated phosphate at industrial scale is more recent in Brazil (around 10 years) and it is a result of the need to reduce sulphur consumption and to use rock phosphates with a high degree of impurities, mainly iron and aluminum oxides. The technology for industrial production of partially acidulated phosphates is simpler than to produce superphosphates.

Research to evaluate the agronomic efficiency of these sources is relatively recent. A literature review conducted by Goedert et al. (1986a) indicated that with 50% acidulation, the AEI was approximately 60% in relation to the AEI of triple superphosphate.

A broad research program initiated in 1983 involved EMBRAPA (Empresa Brasileira de Pesquisa Agropecuária) and PETROFERTIL (Petrobrás Fertilizantes S.A.). Petrofértil produced five levels of acidulation in six diverse rock phosphates (Anitápolis, SC; Araxá, MG; Catalâo, GO; Olinda, PE; Patos de Minas, MG; and Tapira, MG). Initially, these thirty products were evaluated under greenhouse conditions by using corn as the test crop (Goedert and Sousa, 1986). The main conclusion of this study was that the agronomic efficiency of the several phosphate sources was directly related to the percentage of water-soluble, citric acid or ammonium citrate extractable phosphorus.

The relationship between AEI and proportion of total phosphorus extracted by citric acid is depicted in Figure 4 (Goedert and Sousa, 1986). It can be noted that the relationship is linear and that the gradient is close to the unit or, in another words, the AEI is similar to the proportion of phosphorus extracted by citric acid, irrespective to the origin of the phosphate rock. Similar results were obtained in relation to ammonium citrate extractions. This behavior has been confirmed by recent studies with annual crops (Nachtigall et al., 1989; Goedert et al., 1990).

Figure 4. Relationship between Agronomic Efficiency Index (AEI%) and several degrees of phosphate acidulation. Corn dry matter data in greenhouse. Citric acid P extraction (2%; 1/100 ratio).Source: Goedert and Sousa, 1986.

Part of these products have been evaluated under field conditions with sevaral crops. The results, in general, confirm greenhouse data: the agronomic efficiency was proportional to the level of acidulation of the rock phosphate and directly related to the percentage of total phosphorus that is soluble in ammonium citrate. Results of two field experiments involving soybeans and corn are showed in Figures 5 and 6. In both locations the AEI for partially accidulated phosphates was considerably lower than that obtained for triple superphosphate. It should also be stressed that the efficiency for the partially acidulated phosphate did not increase with time, suggesting that the insoluble phosphorus fraction remained so and was not absorbed by the crop roots.

Figure 5. Soybean yield in a Dark Red Latosol under "cerrado" as a function of annually applied phosphate fertilizers (TSP and partially acidulated Araxá phosphate) at seeding furrow. Source: Goedert et al., 1986a.

Figure 6. Corn response to rates of phosphate fertilizer (TSP and partially acidulated phosphate), applied at seed furrow. Source: Raij and Feitosa, 1986.

In this sense, it is agreed that the economic viability of partially acidulated rock phosphate for annual crops should be considered as a function of the price of citric acid or ammonium citrate soluble P₂O₅ (Goedert et al., 1986a).

Besides sulphuric acid, other acid materials have been evaluated to solubilize Brazilian rock phosphates. This research program is justified because try to search for better and complete utilization of this natural resource. The following acid materials are under study: phosphoric acid, nitric acid, urea nitrate, urea phosphate, and ammonium bi-sulphate.

Partial acidulation of several rock phosphates (Araxá, Catalâo, and Patos de Minas) with phosphoric acid were studied by Barreto (1977), Franco (1977) and Ferreira and Kaminski (1979). The results indicated that the efficiency of these products were also proportional to the quantity of phosphorus solubilized by the acid.

The production of nitrophosphates can be of interest in countries with limitation of sulphur reserves, as Brazil. These products however present high degree of instability and hygroscopicity. According to Malavolta (1981), nitrophosphates do not present advantages, compared to ammonium phosphates, in agronomic aspects. Goedert and Souza (1986) concluded that the Patos rock phosphate acidulated with nitric acid presented similar AEI to acidulation with sulphuric acid at the same acidulation level (molar ratio H⁺/P₂O₅).

Mixtures of rock phosphate with urea nitrate or urea phosphate have been of low efficiency in phosphate solubilization.

Ammonium bi-sulphate, a mixture of ammonium sulphate (AS) and sulphuric acid (SA), has also been evaluated to solubilize rock phosphates. Several products obtained by this process received the name RNC (Not Conventional Route) and involve several molar ratios H⁺/P₂O₅ and AS/SA. The results of these studies indicated the potencial of this route to solubilize the phosphate rocks with high impurity percentage (iron and aluminum oxides), as the Patos deposit, as showed in Figure 7.

Figure 7. Summary of agronomic efficiency of several phosphate fertilizers for annual crops. Source: Goedert et al., 1986a.

The phospho-nitro-sulpho-calcium is an alternative product in the ammonium phosphate production route, having the same ratio N/P₂O₅ of the ammonium diphosphate (DAP). This product, which is less water-soluble than ammonium phosphate, has shown similar efficiency to triple superphosphate.

Research involving thermophosphates in Brazil started twenty years ago with sugar cane (Alvarez et al., 1965). Meanwhile this product does not present water-soluble phosphorus, when applied as powder (75% smaller than 0.15 mm sieve), the AEI for annual crops has been equivalent or even superior to superphosphates (Table 3 and Figure 7). Another aspects concerning these products are the additional effects in increase soil pH, high percentage of calcium, magnesium, silicon and also some micronutrients that can explain the observed AEI above 100%.

Under the agronomic point of view, magnesium thermophosphates are excellent phosphorus sources for acid soils of the "cerrado" region of Brazil. The main limitations are the high comparative price in terms of P₂O₅, the low percentage of total P₂O₅ and the necessity to be applied as powder, as observed by Kaminski et al. (1988).

Agronomic efficiency for pasture species. A literature review concerning the use of natural phosphate rocks for direct application in pasture can be summarized as follows, according to Sanzonowicz and Goedert (1986):

1. the efficiency of the phosphates is dependent upon soil type, mainly in relation to texture and acidity level;
2. the level of AEI is dependent upon the rate, being very low for low rates of applied P₂O₅, leading to problems of stand stablishment;
3. the great majority of field experiments is of short duration, not allowing residual effects evaluation;
4. the average values for AEI, calculated from several greenhouse and field experiments, was around 30%, confirming the fact that Brazilian rock phosphates are slow and of low solubilization in these soils.

Aditional insight in this subject can be obtained by data presented in Figure 8 and Table 4 (Sanzonowicz et al., 1987). It was observed a positive response of Brachiaria decumbens dry matter yield up to the higher P₂O₅ rate, irrespective to the phosphate fertilizers used. The efficiency of tested sources concerning total dry matter yield was as follows: simple superphosphate = magnesium thermophosphate>Gafsa phosphate > Araxá phosphate. At the lowest rate of applied P₂O₅ (87 kg ha^-1), Araxá phosphate presented extremely low AEI (1%) as compared to Gafsa phosphate (66%), and simple superphosphate (100%), mainly due to the low initial solubilization and poor stand. However at higher P₂O₅ rates, 343 and 1374 kg ha^-1, the AEI achieved 61% and 88% for the Brazilian rock phosphates, respectively. These points are extremely important to make any final economic evaluation of the use of this low solubility rock phosphate in "cerrado" region of Brazil.

Figure 8. Cumulative relative yield of Brachiaria decumbens during ten years, as a function of sources and rates of phosphate fertilizer applied in the first year in a Dark Red Latosol under "cerrado". Source: Sanzonowicz et al., 1987.

Other experiments involving Brachiaria humidicola and Andropogon gayanus conducted in these soils, have shown good results upon application of these natural rock phosphates, as indicated in Table 3. A recent field experiment with three phosphorus sources - triple superphosphate, thermal phosphate and phosphate rock - on a pasture association of Andropogon gayanus and Stylosanthes capitata showed best results with thermal phosphate, specially during first and second year (Couto et al., 1989).

zAs a conclusion, the viability for direct application of low solubility Brazilian rock phosphates for pasture production in the "cerrado" region is dependent upon a number of agronomic and economic factors. In the majority of the cases, the use of this product is a viable practice; if one considers the following aspects:

a) Pasture species must have high phosphorus absorption capacity and tolerance to high acidic conditions. Most of the cultivated Brachiaria and Andropogon species are included in this category.

b) These low solubility rock phosphates present higher efficiency in acid, clayey and high phosphorus fixing capacity soils.

c) The efficiency is dependent upon the rate of phosphorus applied. In clayey soils with very low natural phosphorus availability, rates below 80 kg P₂O₅ ha^-1 do not seem viable.

d) To increase the solubilization velocity, the product must be finely ground (85% smaller than 0.075 mm sieve), broadcasted and mixed with a great soil volume.

e) Considering the slow initial formation of pasture stand when only these rock phosphates are used, there are advantages in terms of a initially small complementary use of water-soluble sources.

f) Meanwhile the economic viability is specific for each situation, the price per unit of total P₂O₅ of the rock phosphates should be less than half of the price per unit of total P₂O₅ of a soluble source (p.e.superphosphate).

Concerning the use of partially acidulated rock phosphate for pasture species, their economic viability, as described for annual crops, should be considered as a function of the price of citric acid soluble P₂O₅ (Faria and Albuquerque, 1989).

Agronomic efficiency for perennial crops and reforestation. Studies concerning agronomic evaluation of phosphorus sources for perennial crops (fruit crops, coffee, reforestation, etc.), have been less in number than for annual crops and pastures.

The main focus has been to compare low solubility Brazilian rock phosphates with more soluble sources. In general, these studies mainly refer to initial growth and development parameters, existing very little data available on yield and product quality.

Braga et al. (1977) evaluated five phosphorus sources and fixed rates of P₂O₅/400 g soil recipient to produce seedlings of Eucalyptus camaldulensis. Data presented in Table 5 indicated the possibility of replacement of soluble sources by the Araxá rock phosphate, and the low efficiency of Patos phosphate, concerning height of eucaliptus seedlings, after 115 days of application of the sources. In another study involving Eucalyptus saligna, 2.5 g of Araxá phosphate per seedling recipient, the seedlings reached 29 cm after 90 days, as compared to 15.1, 15.7, 21.2 and 15.4 cm for the same rate of Patos, Catalâo, Tapira and Tufito rock phosphates, respectively. However, field evaluation of height after 790 days for the same species, showed that the 250 g single superphosphate rate per plant were slightly better than 500 g of these natural rock phosphates per plant (Braga and Rocha, 1979).

Table 4. Cumulative dry matter yield of Brachiaria decumbens during ten years (twenty three harvests), average for three lime levels, as response to sources and rates of phosphate fertlizers applied in the first year. Dark Red Latosol, clayey, under "cerrado" (Sanzonowicz et al., 1987).

Phosphate	Rate of P2O5	Cumulative Yield	AgronomicEfficiency Index (AEI)
-	kg ha-1	t ha-1	%
Test plot (-P)	0	17.7	-
Araxá phosphate	87	17.8	1
Araxá phosphate	343	47.6	61
Araxá phosphate	1 343	92.4	88
Gafsa phosphate	87	24.7	66
Gafsa phosphate	343	63.3	93
Gafsa phosphate	1 343	99.9	97
Thermophosphate	87	25.6	75
Thermophosphate	343	64.8	97
Thermophosphate	1343	109.4	108
Single superphosphate	87	28.3	100
Single superphosphate	343	66.5	100
Single superphosphate	1 343	102.8	100

Fig. 9. Cummulative dry matter yield of Brachiaria humidicola during seven years, as a function of rates of phosphate fertilizer (SSP nad Pastos phosphates). Fertilizers were applied two years before pasture formation in a red-yellow Latosol, clayey, under "cerrado".
Source: Sousa et al. 1987.

A more recent study proved the possibility of rock phosphate use in reforestation, mainly when associated with N-P-K (10-28-6) + B + Zn in soils under "cerrado" of Minas Gerais State. Data from Table 6, adapted from Rezende et al. (1982), indicate that the Patos phosphate at Bom Despacho was not so efficient to increase the yield of Eucalyptus grandis, as Araxá phosphate at Carbonita. However, broadcast of 2 t ha^-1 of Patos phosphate, complemented with 150 g NPK + B + Zn per plant, led to 90% more yield than the mixed fertilizer alone. At Carbonita broadcast of 2 t ha^-1 of Araxá phosphate gave yield similar to the mixed fertilizer, but the yield was doubled when this rate of Araxá phosphate was associated with 75g mixed fertilizer per plant. The possibility of using rock phosphate in reforestation was also demonstrated by Dantas et al. (1988).

Recent partial conclusions concerning the use of low solubility brasilian rock phosphates in reforestation with eucalyptus are reported in a series of fields trials conducted at seven locations of Minas Gerais State (UFV, 1985). These data suggest that less soluble sources of phosphorus, particularly Araxá phosphate, can substitute more soluble sources, if the eucalyptus growth period is, at least, three years. The yields, however, can be improved by combination of broadcast low solubility rock phosphates, with more soluble sources at planting sites. In the case of sandy soils more soluble phosphorus sources are more efficient.

Table 5. Average height of Eucalyptus camaldulensis seedlings, 115 days after application of different phosphate sources and rates. (Braga et al., 1977).

Rate applied	Araxá phosphate	Single super phosphate	Triple super phosphate	DAP	Patos "Tufito" phosphate
g pot-1	cm
0.35	1.95	1.80	4.10	2.23	2.55
0.70	7.13	3.83	3.93	3.90	5.05
1.05	5.55	4.00	7.63	4.83	4.28
1.40	7.60	4.00	8.38	6.03	4.03
ACI	1.10	1.00	1.53	0.66	0.38

ACI: Availability Coefficient Index

Table 6. Eucalyptus grandis yield (at 26 and 24 months), as function of several fertilizers treatments at two location of Minas Gerais State (Bom Despacho and Carbonita). (Adapted from Rezende et al., 1982).

Treatment	Bom Despacho (Patos phosphate)	Carbonita (Araxá phosphate)
-	m3 ha-1
A) 2t ha-1PR+150g/plant NPK+B+Zn	86.8a	92.3ab
B) 2t ha-1PR+ 75g/plant NPK+B+Zn	68.8b	100.8a
C) 2t ha-1PR	13.2d	49.3cd
D) 1t ha-1PR+150g/plant NPK+B+Zn	52.2bc	62.2bc
E) 1t ha-1PR	17.2d	29.0de
F) 150g/plant NPK+B+Zn	45.7c	49.7cd
G) Test plot (no fertilizers)	8.4d	2.2e
CV (%)	14.7	25.6

Average values followed by same letters are not different at the 5% level.

Meanwhile coffee plants are considered not much responsive to phosphorus, there are sufficient data of remarkable positive effects of phosphate fertilizers use during initial stages of this crop in the soils under "cerrado" vegetation (Hara et al, 1974; Souza and Caixeta, 1974). In this last study, five years coffee production data showed linear response up to 120g P₂O₅/plant/year as simple superphosphate. Also Pereira et al. (1979) observed increases in coffee yield ranging from 50 to 148%, due to 30g P₂O₅/plant/year in soils from the Zona da Mata of Minas Gerais State. A more recent study indicated that phosphorus is the most important plant nutrient for coffee during the first six months in a Dark Red Latosol of Minas Gerais (Santinato et al., 1980).

Coffee production response or agronomic efficiency of rates of phosphorus are extremely dependent upon soil type, as depicted in Table 7 according to Guimarâes (1986). Rates 1 and 2 in this Table refer to 78 and 156g P₂O₅/plant, half and half by weight as simple superphosphate and Araxá phosphate, respectively.

In terms of comparison among sources of phosphorus for coffee crops, the great majority of field experiments has demonstrated the possibility of low solubility Brazilian rock phosphates to substitute water-soluble phosphorus sources. In this case, AEI has been similar when using the double of total P₂O₅ as rock phosphate in relation to simple superphosphate and quadruple of total P₂O₅ as rock phosphate in relation to triple superphosphate (Silva et al., 1983; Santinato et al., 1983; Viana et al., 1985; Santinato et al., 1985).

Concerning fruits crops, mainly citrus, Sousa (1976) concluded that phosphorus was the most important nutrient related to a series of parameters evaluated during the first three years in these soils. Maximum aerial plant development was achieved with 420g total P₂O₅/plant.

For both fruit crops and coffee plants, avalilable research data suggest that combination of natural rock phosphates with more soluble phosphorus fertilizers is a good approach for plant bed preparation. Souza et al. (1978) compared 17 different "Pera Rio" citrus bed fertilization in soils under "cerrado", at Alfenas, south of Minas Gerais. Two harvest data gave the following yields increase in relation to the test plot (no phosphorus): 900 g simple superphosphate (169%); 2670 g Araxá phosphate (223%); 640g bone meal (165%); 20 liter cow manure (290%); the best treatment was a combination of 1330 g Araxá phosphate + 400g single superphosphate + 20 liter cow manure (308%).The official fertilizer and lime recomendation for coffee and fruit crops bed preparation in Minas Gerais and Goias States in Brazil suggests that one half of P₂O₅ should be as Brazilian rock phosphates and the other portion as more soluble sources (CFSEMG, 1978; CFSEMG, 1989; CFSG, 1988). For maintenance fertilization, since the previous and continuous lime use reduces the agronomic efficiency of Brazilian rock phosphates, phosphorus fertilizers should be in more soluble forms.

Table 7. Responses of Catuaí coffee to an increase in phosphorus rates in two Oxisols under "cerrado" at Minas Gerais State. Measurements at 16 months old plants. (Adapted from Guimarâes 1986).

Rates (NPK)	Machado (LE) Plant height Canopy diameter		S.S.Paraíso (LRd) Plant height Canopy diameter
-	cm	-	-	-
2-0-2	47.6	55.5	55.0	79.0
2-1-2	62.3	88.2	61.5	84.5
2-2-2	73.0	108.2	59.5	73.3

CONCLUSIONS

Based upon the data presented and discussed in this literature review the following conclusions can be drawn:

1. The agronomic efficiency of phosphate fertilizers produced in Brazil for annual crops is, in general, proportional to the percentage of soluble P₂O₅ in relation to total P₂O₅.
2. Brazilian rock phosphates in general, present low solubility and low agronomic efficiency for annual crops. Their use is justified only for acid tolerant pasture species formation and for certain perennial crops.
3. Even if the positive industrial aspects of partially acidulated rock phosphates are taken into account, these products will only be viable for annual crops if the costs for the farmer are established as a function of the soluble P₂O₅ fraction. Partially acidulated rock phosphates, however, can be adequate for pastures and perennial crops.
4. Thermophosphates represent an excellent agronomic option for most acid soils of the Brazilian "cerrado", justifying increases in production in Brazil.
5. The traditional soluble sources (superphosphates and ammonium phosphates) are highly efficient and justify their production. It is not justified the myth that being originally developed for climate and soils from the temperate regions are not adequate for the Brazilian conditions.
6. Long term field experiments (10 or more years) have shown cumulative absorption of applied soluble phosphorus in order of 60 to 70%. This indicates that phosphorus fixation problems are minimized with cropping time.
7. Considering the diversity of soils, climates and crops in the Brazilian agriculture, is absolutely valid to produce and commercialize several kinds of phosphate fertilizer in Brazil.
8. The agronomic efficiency of phosphate fertilizers produced and commercialized in Brazil is much more a function of adequate soil and crop management than change in industrial technology. In the specific situation of water soluble phosphorus fertilizers, the use of small rates of lime per hectare is one serious limiting factor to increase the agronomic efficiency of these products.

REFERENCES

1. Alvarez, R., H.V. Arruda, de and A.P.C. Wutke. 1965. Adubaçâo de cana-de-açucar. X - Experiência com diversos fosfatos (1959-60). Bragantia 24:1-8.
2. ANDA. Associaçâo Nacional para Difusâo de Adubos e Corretivos Agrícolas. 1987. Plano Nacional de Fertilizantes. ANDA, Sâo Paulo, SP. 235 p.
3. Adamoli, J., J. Macedo, L.G. Azevedo and J. Madeira Neto. 1986. Caracterizaçâo da regiâo dos cerrados. In: Goedert, W.J. (ed.) Solo dos cerrados: tecnologia e estratégia de manejo. Editora Nobel, Sâo Paulo, SP. pp.33-74.Barreto, A.C. 1977. Efeitos de acidulaçâo parcial, de misturas com superfosfato triplo e enxofre e da granulaçâo na eficiência de fosfatos naturais. Porto Alegre, Facultad de Agronomia, UFRGS, MSc Thesis. 63 p.
4. Barreto, A.C. 19977. Efeitos de acidulaçao parcial, de misturas com superfosfato triplo e enxofre e da granulação na eficência de fosfato naurais. Porto Alegre, Facultadade de Agronomia UFRGS, MSc Thesis. 63 pp.
5. Belger, R.B., C.E. da S. Braga, E.A. de Carvalho, G.B. de Moura, L.A. Takata, L.C. Alvarenga, E.C. Damasceno and P.C.W. Albuquerque. 1987. Evoluçâo da recuperaçâo de P₂O₅ na mineraçâo. In: IBRAFOS (ed.) Seminário Sobre Recuperaçâo de Fósforo. Sâo Paulo, SP. pp. 151-164.
6. Braga, J.M. 1969. Comparaçâo de fosfatos aplicados ao feijoeiro: 1. Trabalhos de casa de vegetaçâo. Revista Ceres 16:88-101.
7. Braga, J.M., Couto, L., Neves, M.J.B. and Brandi, R.M. 1977. Comportamento de mudas de Eucalytus spp em viveiro, em relaçâo a aplicaçâo de N, P e K e diferentes fontes de fósforo. Revista Arvore 1:135-148.
8. Braga, J.M. and Rocha, D. 1979. Estudos de adubos fosfatados na cultura do eucalipto em solos de cerrado de Minas Gerais. Universidade Federal de Viçosa, Boletim Técnico. Paper presented at the I Seminário de Investigaçôes Florestais. Belo Horizonte. 18 p.
9. CFSEMG. 1978. Comissâo de Fertilidade do Solo do Estado de Minas Gerais. Recomendaçôes para o uso de corretivos e fertilizantes em Minas Gerais - 3a aproximaçâo. EPAMIG, Belo Horizonte, MG. 80 p.
10. CFSEMG. 1989. Comissâo de Fertilidade do Solo do Estado de Minas Gerais. Recomendaçôes para o uso de corretivos e fertilizantes em Minas Gerais - 4a aproximaçâo. Lopes, A.S. and Guimarâes, P.T.G (eds.) CFSEMG, Lavras, MG. 159 p.
11. CFSG. 1988. Comissâo de Fertilidade do Solo de Goiás. Recomendaçôes para o uso de corretivos e fertilizantes para Goiás - 5a aproximaçâo. Inf.Tec. UFG/EMGOPA N° 1, Goiânia, GO. 101 p.
12. Cochrane, T.T., Sanchez, L.G., Rodrigues, G.C., Souza, D.M.G. and Freitas Jr., E.de.1985. Land in Tropical America. CIAT/EMBRAPA-CPAC, Cali, 3V.
13. Couto, W.; G. Leite and C. Sanzonowicz. 1989. The effectiveness of three phosphorus sources for pasture fertilization in a cerrado soil. Pesquisa Agropecuária Brasileira 24:423-430.
14. Dantas, S.V.; Barros, N.F.; Novaes, R.F.; Sediyana, C.S. and Neves, J.C.L. 1988. Avaliaçâo da eficiência de fosfatos em plantios de Eucaliptus grandis. In: Reuniâo Brasileira de Fertilidade do Solo, 18, Guarapari, Summary. pp. 33-34.
15. EMBRAPA. Empresa Brasileira de Pesquisa Agropecuária. 1980. Centro de Pesquisa Agropecuária dos Cerrados, Planaltina, DF. Relatório Técnico Anual, 1978-1979. 170 p.
16. FAO. 1988. FAO Yearbook - Fertilizer - 1987. FAO Statistics Series 83. FAO, Rome, Italy. 123 p.
17. Faria, C.M.B. de and Albuquerque, S.G. de. 1989. Fosfato parcialmente solubilizado em pastagem de capim-búfel no submédio Sâo Francisco. Revista Brasileira de Ciência do Solo 13:393-397.
18. Ferreira, T.N. and Kaminski, J. 1979. Eficiência agronômica dos fosfatos naturais de Patos de Minas e Gafsa, puros e modificados por acidulaçâo e calcinaçâo. Revista Brasileira de Ciência do Solo 3:158-162.
19. Fioreze, I.; Antoniazzi, N.; Ferreto, M. and Rossato, V. 1990. Efeito de misturas de superfosfato triplo com fosfato de gafsa como fonte de fósforo às culturas de soja e trigo. In: Reuniâo Brasileira de Fertilidade do Solo e Nutriçâo de Plantas, 19, Santa Maria, Summary. 90 p.
20. Firme, D.J.; Ribeiro, A.C.; Costa, L.M. da; Barros, N.F. de and Novais, R.F. de. 1990. Enriquecimento e fusâo de escória de siderurgia com fosfato natural: II. Testes biológicos. Revista Brasileira de Ciência do Solo 14:247-254.
21. Franco, M. 1977. Fosfatos parcialmente acidulados com H₃PO₄, HCl e H₂SO₄ na cultura do sorgo granífero (Sorghum bicolor L. Moench) em um solo de cerrado de Ituiutaba-MG. Viçosa, Faculdade de Agronomia, UFV, MSc Thesis. 75 p.
22. Giulietti, M., Weyne, G.R.S., Guardani, R. and Calmanovici, C.E. 1987. Rotas opcionais visando o aproveitamento de fosfatos brasileiros na produçâo de fertilizantes. In: IBRAFOS (ed.) Seminário Sobre Recuperaçâo de Fósforo. Sâo Paulo, SP. pp. 281- 322.
23. Goedert, W.J. 1987. Management of acid tropical soils in the savannas of South America. In: IBSRAM (International Board for Soil Research and Management). Management of acid tropical soils for sustainable agriculture: proceedings of an IBSRAM inaugural workshop, Bangkok, Thailand. pp. 107-127.
24. Goedert, W.J. 1989. Regiâo dos cerrados: potencial agrícola e política para o seu desenvolvimento. Pesquisa Agropecuária Brasileira 24:1-17.
25. Goedert, W.J. and F.A. Dias Filho. 1986. Relatório Bienal 1984/85. Convênio EMBRAPA/Petrofértil. Brasília, DF. 164 p.
26. Goedert, W.J. and Lobato, E. 1980. Eficiência agronômica de fosfatos em solo de cerrado. Pesquisa Agropecuária Brasileira 15: 311-318.
27. Goedert, W.J. and Lobato, E. 1984. Avaliaçâo agronômica de fosfatos em solo de cerrado. Revista Brasileira de Ciência do Solo 8:97-102.
28. Goedert, W.J., Rein, T.A. and Sousa, D.M.G. de. 1986a. Eficiência agronômica de fertilizantes fosfatados nâo tradicionais. Planaltina, EMBRAPA - CPAC. (EMBRAPA-CPAC. Documentos, 24). 21p.
29. Goedert, W.J., Rein, T.A. and Sousa, D.M.G. de. 1990. Eficiência agronômica de fosfatos naturais, fosfatos parcialmente acidulados e termofosfatos em solo de cerrado. Pesquisa Agropecuâria Brasileira 25:521-530.
30. Goedert, W.J. and D.M.G. de Sousa. 1986. Avaliaçâo preliminar da eficiência de fosfatos com acidulaçâo parcial. Revista Brasileira de Ciência do Solo 10:299-301.
31. Goedert, W.J., D.M.G. de Sousa, Lobato, E. and Rein, T.A. 1987. Avaliaçâo agronômica de fertilizantes fosfatados no cerrado brasileiro. Paper presented at the Simposium "Alternativas sobre el uso como fertilizantes de fosfatos provenientes de depósitos de América Latina Tropical y Subtropical", CIAT, Cali, Colombia. (in press).
32. Goedert, W.J., Sousa, D.M.G. de and Lobato, E. 1986b. Fósforo In: Goedert, W.J. (ed.) Solos dos cerrados, tecnología e estratégias de manejo. NOBEL/EMBRAPA-CPAC, Sâo Paulo/Brasília. pp. 129-166.
33. Guimarâes, P.T.G. 1986. Respostas do cafeeiro (Coffea arabica L. cv. Catuai) à adubaçâo mineral e orgânica em solos de baixa fertilidade do Sul de Minas Gerais. Piracicaba, ESALQ, USP. PhD Dissertation. 140 p.
34. Hara, T., P.T.G. Guimarâes, F.M. Freire, F.G.F.T.C. Bahia, J.G. de Carvalho, M.M. de Carvalho, S.P. de Souza, J.F. Mendes and J.M. Braga. 1974. Ensaios de adubaçâo mineral e orgânica para a formaçâo de cafezais em solos sob vegetaçâo de cerrado. In: Congresso Brasileiro de Pesquisas Cafeeiras, 2, Poços de Caldas. Summary. pp. 277-278.
35. Hughes, J.C. 1980. Crystalinity of kaolin minerals and their weathering sequence in some soils from Nigeria, Brazil and Colombia. Geoderma 24:317-325.
36. Kaminski, J.; Santos, D.R. and Oliveira, E.P. 1988. Eficiência de termofosfatos em funçâp de sua granulometria-Gramíneas forrageiras. In: Reuniâo Brasileira de Fertilidade do Solo, 18, Guarapari, Summmary. pp. 105-106.
37. Leal, J.R. and A.C.X. Velloso. 1973. Adsorçâo de fosfato em latossolo de cerrado. Pesquisa Agropecuária Brasileira, Serie Agron. 8:81-88.
38. Le Mare, P.H. 1982. Sorption of isotopically exchangeable and non-exchangeable phosphate by some soils of Colombia and Brazil, and comparisons with soils of southern Nigeria. J.Soil Sci. 33:691-707.
39. Le Mare, P.H. and W.J. Goedert. 1984. Effects of liming on sorption and desorption of phosphate by some acid soils of South America. Proc. 3rd Int. Congress of Phosphorus Compounds, World Phosphate Inst. OMPHOS, Casablanca, Morocco.
40. Lobo, M.G. 1986. Evoluçâo e perspectivas da industria nacional de fosfatos. In IBRAFOS (ed.) III Encontro Nacional da Rocha Fosfática. Brasília, DF.pp. 77-94.
41. Lobo, M.G. and R.M. Silva. 1984. Produçâo de fertilizantes fosfatados. In: Anais do Simpósio sobre Fertilizantes na Agricultura Brasileira. Espinosa, W. and Oliveira, A.J. de (eds.) EMBRAPA/DEP, Brasília. pp.73-102.
42. Lopes, A.S. 1975. A survey of the fertility status of soils under "cerrado" vegetation in Brazil. Department of Soil Science, North Carolina State University, Raleigh, NC, USA. MSc Thesis. 138 p.
43. Lopes, A.S. and F.R. Cox, 1977. A survey of the fertility status of surface soils under "cerrado" vegetation in Brazil. Soil. Sci. Am.J. 41:742-747.
44. Malavolta, E. 1981. Adubos e adubaçâo fisfatada. In: Manual de Química Agrícola - Adubos e Adubaçâo. 3a ed. Ceres. Sâo Paulo. pp.97-166.
45. Nachtigall, G.R.; Brauner, J.L.; Zolezzi, M. and Tapia, F. 1989. Mistura de superfosfato triplo com fosfato de Jacupiranga no mesmo grânulo na eficiência do fosfato natural. Revista Brasileira de Ciência do Solo 13:269-273.
46. Pereira, J.E., Bragança, J.B. and Paulino, A.J. 1979. Modo e frequência de aplicaçâo de fósforo na formaçâo de cafeeiro. In: Congresso Brasileiro de Pesquisas Cafeeiras, 7. Araxá, Summary. pp. 306-307.
47. Raij, B.V. and C.T. Feitosa. 1986. Avaliaçâo agronômica de fertilizantes fosfatados. In: Goedert, W.J. e Dias Filho, F.A. (eds.) Relatório Bienal 1984/85. Convênio EMBRAPA/Petrofértil. Brasília, DF. pp. 113-121.
48. Rezende, G.C. de, Barros, N.F. de, Moraies, T.S. de A., Mendes, C.J. and Suiter Filho, W. 1982. Aplicaçâo de fosfatos naturais em plantios de Eucalyptos grandis W. Hill ex Maiden. Revista Arvore 6:74-83.
49. Santinato, R., Silva, O.A. and Barros, U.V. 1983. Fosfato natural de Patos-MG, em doses crescentes, como fonte de fósforo no plantio do cafeeiro, In: Congresso Brasileiro de Pesquisas Cefeeiras 10, Poços de Caldas, Summary. pp. 144-147.
50. Santinato, R., Silva, O.A., Figueiredo, J.P. Calvalho, C.H.S. and Barros, U.W. 1980. Estudo do crescimento, composiçao química e deficiência do cafeeiro cultivado em solo de cerrado - LVE. In: Congresso Brasileiro de pesquisas Cafeeiras, 8, Campos do Jordâo, Summary. pp. 428-437.
51. Santinato, R., Silva, O.A., Figueiredo, J.P. and Santo, J.E. 1985. Estudo do fosfato natural de Patos, em doses crescentes como fonte de fósforo no plantio do cafeeiro. In: Congresso Brasileiro de Pesquisas Cafeeiras, 12, Caxambú, Summary.pp.97-98.
52. Sanzonowicz, C. and W.J. Goedert. 1986. Uso de fosfatos naturais em pastagens. EMBRAPA-CPAC. Planaltina (EMBRAPA-CPAC Circular Técnica, 21). 33 p.
53. Sanzonowicz, C., E. Lobato, and W.J. Goedert. 1987. Efeito residual de calagem e de fontes de fósforo numa pastagem establecida em solo de cerrado. Pesquisa Agropecuária Brasileira 22:233-243.
54. Silva, J.B.D. da, Lopez, A.S. and Carvalho, M.A. de 1983. Sequência e intervalos de calagem e aplicaçâo de fosfato de Patos para produçâo de soja em Latossolos sob cerrado. XIX Congresso Brasileiro de Ciência do Solo, Curitiba, PR (summary).
55. Silva, O.A., D. Santinato, R. and Carvalho, C.H.S. 1983. Fosfato natural de Araxá, em doses crescentes, como fonte de fósforo no plantio do cafeeiro. In: Congresso Brasileiro de Pesquisas Cafeeiras, 10, Poços de Caldas, Summary. pp. 137-140.
56. Smyth, T.J. and Sanchez, P.A. 1982. Phosphate rock dissolution and availability in cerrado soils as affected by phosphorus sorption capacity. Soil Sci. Soc. Am. J. 46:339-445.
57. Sousa, D.M.G. de, Sanzonowicz, C., Miranda, J.C.C. de and Smyth, T.J. 1987. Manejo de fósforos na cultura da soja e braquiaria em Latossolo Vermelho Amarelo (LV) de cerrado. In: Relatório Técnico Anual do Centro de Pesquisa Agropecuária dos Cerrados 1982-1985. EMBRAPA-CPAC. Planaltina, pp. 115-120.
58. Souza, S.P. de and Caixeta, J.V.M. 1974. Resposta do cafeeiro Mundo Novo á aplicaçâo de nitrogênio, fósforo, potássio e nutrientes. In: Congresso Brasileiro de Pesquisas Cafeeiras, 2, Poços de Caldas. Summary. pp. 276.
59. Souza, M.A. de, Chalfun, N.N.J. and Pádua, T. de 1978. Efeitos de adubaçâo nas covas de laranjeiras "Pera Natal" em solos de cerrado. Proj. Frutic., Relat. 74/77, Belo Horizontes, Summary. pp. 121-125.
60. Souza, M. de. 1976. Efeito do P, K, e Ca no crescimento da parte áerea de laranjeira "Pera Rio" (Citrus Sinensil L. Osbeck em Latossolo Vermelho Escuro Fase Cerrado. Piracicaba, ESALQ, USP. PhD thesis, 132 p.
61. Tanaka, R.T. 1978. Disponibilidade de fósforo do fosfato de Patos para a cultura do milho (Zea mays L.) em Latossolo Roxo Distrófico sob condiçôes de casa de vegetaçâo. Esc. Sup. Agric. Lavras, Lavras, MG. MSc. Thesis. 66 p.
62. UFV. Universidade Federal de Viçosa. 1985. Centro de Ciência Agrárias. Departamento de Solos. Relatório do Convênio FINEP/UFV/SOLOS. Nutriçâo, fertilizaçâo e manejo de solos sob reflorestamento e utilizaçâo de escórias de siderurgia como corretivo e fertilizante de solo. 168 p.
63. Viana, A.S., Correa, J.B., Fioravante, N, and Giovanni, H. 1985. Efeitos de fontes de P e matéria orgânica combinadas com gesso no plantio do cafeeiro (cv Catuí). In: Congresso Brasileiro de Pesquisas Caffeiras, 12, Caxambú, Summary. pp. 46-50.