MATERIALS AND METHODS

Two field trials were carried out at the Agricultural Experimental Station Farm, Fac. of Agriculture, Alexandria Univ., Alexandria, Egypt in 2007 and 2008 summer seasons. The most famous Egyptian sweet potato local cultivar ‘Abees’, a purple skin with sweet orange-flesh was used. Preliminary to each experiment, soil samples of 0-30 cm layer were taken at random from various locations of the study area before planting for analysis (Page et al., 1982). Soil analysis results of the study area are shown in Table 1.

Phosphorus fertilizer sources and levels: Two foremost sources of P fertilizers; chemical calcium superphosphate and natural rock phosphate were used at the equal full dose of 45 kg P₂O₅ feddan^-1 (one fed. = 4200 m² = 2.4 ha) which is considered as a recommended dose for sweet potato production (Hassan et al., 2005a).

Table 1:	Pre-planting soil analysis results of the study area
The physical and chemical properties analyses were conducted at the Soil and Water Science Department, Faculty of Agriculture, Alexandria University

(a) Calcium superphosphate form (15.5% P₂O₅) was applied at 4 levels; 0, 72.5, 145 and 290 kg which represent 00, 25, 50 and 100% of the recommended dose (45 kg P₂O₅ fed.^-1, 100% dose), (b) Rock phosphate form (22.5% P₂O₅) was added at 4 levels; 00, 50, 100 and 200 kg which represent 00, 25, 50 and 100% of the same recommended dose.

All amounts of different P sources were applied at the time of final land preparation.

Vesicular arbuscular mycorrhiza inoculation source: The symbiotic VAM-fungi species (Glomus mosseae) was supplied by the Soil and Water Science Dep., Fac. of Agriculture, Alexandria Univ., Egypt. The VAM-fungi inoculums (50 g) were added to the root absorption zone of sweet potato plants, two weeks from planting (Dawa et al., 2007). Percentage of mycorrhizal colonization was estimated by gridline intersect method (Giovannetti and Mosse, 1980).

Experimental design: Experiment was executed in split-split-plots system in a Randomized Complete Blocks Design (RCBD) with three replications. Each replicate included 16 treatments which represent all possible combinations. Two P fertilizer sources; superphosphate and rock phosphate were set as the main plots. The sub-plots were considered four levels of super-and rock-phosphate (0, 25, 50 and 100% P₂O₅). While, the two VAM-fungi inoculation treatments (with and without) were allocated as the sub-sub-plots. Individual plots consisted of three rows of 4 m long and 0.75 m apart. Healthy vine cuttings (20 cm length) of ‘Abees’ cv. were set up in the field on 5 and 12 June of 2007 and 2008, respectively. They were planted on one side of the row at 25 cm apart. All experimental units received (100 kg fed.^-1) of each N as ammonium sulphate (20.5% N) and K as potassium sulphate (48.5% K₂O). N fertilizer was equally side-dressed to the soil in three diverse intervals after 3, 7 and 10 weeks from planting. The doses of K fertilizer were equally split and applied after 3 and 7 weeks from planting. Extra agricultural practices often used in sweet potato production; irrigation and pest control were adopted using the Egyptian Ministry of Agriculture recommendations.

Measured parameters
Vegetative growth traits: Five plants randomly picked up from the sub-sub-plots, two weeks before harvesting (within 100 days from transplanting), to measure the following traits: vine length plant^-1 (cm), number of branches plant^-1, number of leaves plant^-1, vine fresh weight plant^-1 (kg) and leaves dry weight plant^-1 (%).

Tuber roots yield and its components: At harvesting stage 120 days after planting, all tuber roots plots^-1 were collected and the following measurements were recorded: fresh weight of root (average five uniform roots, g), total root yield plant^-1 (kg), total root yield (ton fed^-1), marketable root yield (ton fed.^-1) and unmarketable root yield (ton fed.^-1). Unmarketable yield is represented by thinner (less than 3 cm) or thicker (more than 10 cm) roots in diameter as well as cut and injured roots.

Organic composition of tuber roots: A random sample of five uniform roots from the sub-sub-plots were carefully washed with distilled water, then weight and analyzed for root organic composition analyses. Total carotenoides as β-carotene (mg 100 g^-1 fresh weight) was measured (Witham et al., 1971). Total sugars, starch and carbohydrates were determined following the standard methods of association of official analytical chemists (AOAC, 1995). Total Soluble Solid (TSS) was estimated using a portable refractometer.

Statistical analysis: The reached data were analyzed statistically using a computer program (Co-Stat Software, 2004). The comparisons among means of the treatments were achieved by way of Duncan’s multiple range tests at 0.05 probability level (Steel and Torrie, 1980).

RESULTS AND DISCUSSION

Vegetative growth traits
Effects of phosphorus sources: Addition of P fertilizer as a superphosphate for sweet potato plants gained the most vigor plant growth traits; tallest vine length, largest number of branches plant^-1 and largest number of leaves plant^-1, if compared with a rock phosphate (Table 2). This might be due to that the superphosphate is more availability and solubility for plant absorption, increasing P near roots zone and hence better uptake by the roots (Shaheen et al., 2012). These results agree with Shaheen et al. (2007, 2012), who reported that onion plants which received its P requirements as chemical superphosphate were detected the vigor plant growth if compared with that plants supplied by the natural rock phosphate. On the contrary, vine fresh weight plant^-1, in both seasons and leaves dry weight, in the second season were increased by the application of rock phosphate source. This result is in conformity with El-Banna and Abd El-Salam (2000), who reported in potato plants that using rock phosphate fertilizer increased significantly both foliage fresh and dry weights plant^-1. They explained this increase in potato foliage fresh and dry weights was derived from the contents of rock phosphate source from CaO, F₂O₃, SiO4, SO₄, AlO₃ and MgO, which help in increase metabolites such as plant growth-promoting substances.

Effects of phosphorus levels: Data in Table 2 show that sweet potato plants which received three P fertilizer levels (25, 50 and 100% P₂O₅) were improved compared with the control treatment. Increasing P level from 50 to 100% P₂O₅ was associated with marked increase in most vegetative growth traits. These increases may be due to the beneficial effect of P on the activation of photosynthesis and metabolic processes of organic compounds in plants, thus, encourage plant growth (Gardner et al., 1985).

Table 2:	Vegetative growth traits of sweet potato plants ‘Abees’ cv. as affected by P sources, P levels and VAM-fungi inoculation treatments during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

These results are in accordance with those of Hassan et al., (2005a) and El-Sayed et al. (2011), who found that application of 45 kg P₂O₅ fed.^-1 (100% P₂O₅) to sweet potato plants was significantly superior in increasing vine length, number of branches plant^-1, leaf area plant^-1, canopy fresh and dry weights, compared with control treatment.

Effects of VAM-fungi inoculation treatments: Inoculated sweet potato plants with VAM-fungi produced significantly higher growth traits; number of leaves plant^-1, number of branches plant^-1, vine length plant^-1, vine fresh weight and leaves dry weight plant^-1 when compared with un-inoculated plants (Table 2). Similar results were proved by El-Morsy et al. (2002) and Hassan et al. (2005a), who reported that inoculated sweet potato plants with VAM-fungi markedly, increased vine growth plant^-1, fresh and dry weights compared with un-inoculated plants. Mycorrhizal inoculation improves plant growth by facilitating mineral nutrition and progressing water relation which led to larger plant size (Auge, 2001).

Interaction effects between P sources and P levels: Medium or higher levels of superphosphate (50 or 100% P₂O₅) was associated with significant increases in vine length plant^-1, number of branches plant^-1, number of leaves plant^-1 and leaves dry weight. However, using rock phosphate at the highest level (100% P₂O₅) gave the highest significant value for vine fresh weight plant^-1 in the second season (Table 3). This could probably be explained based on the available P content in the study soil area was low (Table 1), which might have led to higher response to increased supply of the nutrient. The obtained results are in harmony with those of Shaheen et al. (2007), who indicated that the best plant growth of onion plants was attained with the plants that received the highest P rate (48 units of P₂O₅) as superphosphate form.

Interaction effects between P sources and VAM-fungi inoculation: The integrated effect between superphosphate and VAM-fungi generated the highest significant values for vine length plant^-1, number of branches plant^-1, number of leaves plant^-1 and leaves dry weight (Table 3).

Table 3:	Vegetative growth traits of sweet potato plants ‘Abees’ cv. as affected by integrated effects of P sources xP levels, P sources xVAM-fungi inoculation and P levels xVAM-fungi inoculation treatments during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

The enhancement of growth traits resulted in sweet potato plants that inoculated with VAM-fungi may be refer to that the bio-phosphate fertilizer which had a microorganism, its mode of action with chemical P is more effective which in sequence resulted in increase the absorption of nutrimental elements from the soil (Shaheen et al., 2012). The combination between rock phosphate and VAM-fungi revealed significant value for vine fresh weight trait.

Interaction effects between P levels and VAM-fungi inoculation: Inoculation of sweet potato plants with VAM-fungi alone in the absence of P application level gave the highest values of vine length plant^-1, number of leaves plant^-1 and vine fresh weight plant^-1 traits (Table 3). These results may be explain on the basis that mycorhizal symbiosis facilitates acquisition of many nutrients, mainly P which plays a decisive role in improving plant development in sensitive stage such as transplanting period (Bolandnazar, 2009). However, the combination between VAM-fungi inoculation and medium or higher P levels (50 or 100% P₂O₅) produced the highest values of branches number plant^-1 and leaves dry weight (Table 3). The obtained results could be explained based on the results of Negeve and Roncadori (1985), who mentioned that phosphate fertilizer was more effective in stimulating growth of sweet potato plants in the presence of VAM-fungi.

Sweet potato root yield and its components
Effects of phosphorus sources: Application of superphosphate was accompanied by significant increases in average root fresh weight, root yield plant^-1, root yield fed.^-1 and marketable root yield fed.^-1 in comparison with the rock phosphate (Table 4). Application of superphosphate showed significant increase in total root yield plant^-1 (10.63 and 7.69%) as well as total root yield fed.^-1 (34.62 and 25.37%) and marketable root yield fed.^-1 (24.59 and 26.49%) over than rock phosphate, in the first and second seasons, respectively. It could be concluded that, superphosphate source achieved the heaviest root weight and the best root yield of plants. This might be attributed to the better plant growth resulted from superphosphate than rock phosphate (Table 2). These findings can supported by the results of Shaheen et al. (2007, 2012), who found that the lowest onion bulbs weight and bulbs yield was associated with the addition of rock phosphate, while superphosphate form gained the heaviest bulbs weight and tonnage of bulbs yield. On the other hand, unmarketable root yield under natural rock phosphate (1.588 and 1.480 ton fed.^-1) was lower than unmarketable yield with chemical superphosphate (2.633 and 1.782 ton fed.^-1) (Table 4). This may happened due to the development of outsized or jumbo roots under mineral superphosphate.

Efects of phosphorus levels: Increasing P level from 25 to 100% P₂O₅ reflected significant increases in average root fresh weight, root weight plant^-1, total and marketable root yield, while unmarketable yield was significantly decreased comparing with control treatment. There were insignificant differences among the three P levels used on unmarketable root yield trait (Table 4). In general, application of the highest P level was statistically responsible for giving the highest increases for all roots yield and its component traits. This means that P fertilizer impacts on the productivity of sweet potato crop. These results agree with those obtained by El-Sayed et al. (2011). They indicated that root yield and its components of sweet potato were increased by increasing P-rate from 15 to 45 kg P₂O₅ fed^-1. The increases in both total and marketable yield of sweet potato resulting from mineral fertilization might be credited to its favorable effects on the vegetative growth traits (Table 2), which in turn increased root fresh weight, root weight plant^-1 and consequently, root yield fed.^-1.

Table 4:	Root yield and its components of sweet potato plants ‘Abees’ cv. as affected by P sources, P levels and VAM-fungi inoculation treatments during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

The role of P fertilizer on sweet potato production attracted the attention of many researchers e.g., El-Morsy et al. (2002), Hassan et al. (2005a) and El-Sayed et al. (2011). These authors reported that fertilization of sweet potato plants by P caused significant increases in the total and marketable root yield as well as decreased the unmarketable root yield.

Efects of VAM-fungi inoculation: Inoculation of sweet potato plants with VAM-fungi reflected corresponding and significant increases in all roots yield and its component traits than un-inoculated treatment. It is also reduced unmarketable root yield than un-inoculated one (Table 4). The positive effect of VAM-fungi inoculation on sweet potato root yield and its components may be attributed to that, plants inoculated with VAM-fungi (Glomus mosseae) are potentially more effective on nutrient and water acquisition (George et al., 1992). The extra-radical phase of the VAM fungus acts in effect, as an extension of the root system for the uptake of mineral nutrients, especially immobile nutrients like P, Cu and Zn which are transported back to the intra-radical structures where they are released by the fungus for uptake by the plant root cells. Thus, increase root and shoot biomass and improve growth and yield (Douds et al., 2005).

Interaction effects between P sources and P levels: The obtained results exhibited significant differences in total root yield plant^-1 and both total and marketable root yield fed.^-1 (Table 5). The interactions signify that good root yield can be obtained when superphosphate is applied at the highest P level (100% P₂O₅). This was probably because application of superphosphate increased P contents in rooting zone area and hence increased their availability for the growing plants (Sebastiani et al., 2007). The relative increases in root weight plant^-1, total root yield and marketable yield fed.^-1 were: 46.55 and 31.25%, 63.63 and 60.45% and 127.63 and 81.20% in comparison with control treatment in both seasons, respectively. Similar findings were detected by Shaheen et al. (2007), who concluded that, the heaviest onion bulbs weight and bulbs yield were correlated with that plants received superphosphate at the highest P rate (48 units of P₂O₅).

Table 5:	Root yield and its components of sweet potato plants ‘Abees’ cv. as affected by integrated effects of P sources xP levels, P sources xVAM-fungi inoculation and P levels xVAM-fungi inoculation treatments, during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

Interaction effects between P sources and VAM-fungi inoculation: Pertaining to root yield and yield related components of sweet potato, ‘Abees’ cv., adding of superphosphate + VAM-fungi inoculation treatment produced the highest values of root fresh weight, total root yield plant^-1, as well as total and marketable root yield (ton fed.^-1) (Table 5). Such favorable interaction effect results may be expected on the basis that different mycorrhizal fungi explore the soil volume with their hyphae penetration to various extents and this can increase efficiency of P assimilation due to these structures facilitate nutrient transport partly by increasing both the root surface area and the root-soil contact (Westphal et al., 2008).

Interaction effects between P levels and VAM-fungi inoculation: The highest P level combined with VAM-fungi inoculation treatment resulted in more improvement in root fresh weight, root yield plant^-1 and gave higher total and marketable root yield (ton fed.^-1) comparing with other treatments (Table 5). Such good integrated effect can support the findings of Yano and Takaki (2005), who stated that sweet potato plants were highly reliant on mycorrhizal symbiosis for improving shoot growth and root development. On the other side, the highest P level in the absence of VAM-fungi inoculation decreased unmarketable root yield (ton fed.^-1) (Table 5). This result is in disagreement with that of Hassan et al. (2005a), who found that unmarketable root yield significantly reduced at the same P level used (45 kg P₂O₅ fed.^-1) but with the presence of VAM-fungi inoculation treatment.

Chemical composition of sweet potato roots
Effects of P sources: There were significant differences regarding to total sugars, carbohydrates, TSS and total carotene contents, in both seasons as a result of applied superphosphate. However, it revealed higher content of starch, in the second season when compared with rock phosphate (Table 6). This could be due to more solubility and higher availability of nutrients resulted from a mineral Ca-superphosphate compared to rock phosphate or due to the limited solubility of rock phosphate (Shafeek et al., 2004).

Effects of P levels: Increasing the P level up to 50 or 100% P₂O₅ reflected significant increases in root chemical composition traits (total sugars, starch, carbohydrates, TSS and total carotene) of sweet potato compared with the lowest level of P fertilizer (25% P₂O₅) or control treatment (0% P₂O₅), in the two seasons with only one exception for total carotene, in the first season. Whereas, this trait showed linear increase with each increase in P level comparing with control treatment (Table 6). The obtained results are in agreement with those of El-Morsy et al. (2002), Hassan et al. (2005b), El-Sayed et al. (2011). These authors pointed out that an increase in the rate of applied P fertilizer from 15 to 60 kg P₂O₅ fed.^-1 caused an increase in total sugars, TSS, carbohydrates, starch and carotenoids contents in sweet potato root tissues.

Effects of VAM-fungi inoculation: Sweet potato plants inoculated with VAM-fungi significantly resulted in higher root contents of total sugars, starch, carbohydrates and TSS, in both seasons, as well as total carotene, in the first season in comparison with un-inoculated treatment (Table 6). The favorable effect of VAM-fungi on organic composition of sweet potato may be pass on VAM-fungi increase the uptake of P element, P plays the prime role during the breakdown of carbohydrates and/or synthesis of polysaccharides and it is very effective in the synthesis of starch from glucose (Jakobsen and Rosendahl, 1990).

Table 6:	Chemical constituents of sweet potato roots ‘Abees’ cv. as affected by P sources, P levels and VAM-fungi inoculation treatments during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

These results agree with those of Hassan et al. (2005b), who found that inoculation of sweet potato plants with bio-phosphate fertilizer; VAM-fungi significantly increased total carbohydrates, total sugars and total carotene in root tissues than un-inoculated plants.

Interaction effects between P sources and P levels: The results in Table 7 reflected progressive effect for superphosphate source, particularly at the highest level of P for all chemical composition traits (e.g., total sugars, starch, carbohydrates, TSS and total carotene) in comparison with other treatments. This result may be explained on the basis that the necessity of available P as a plant nutrient is emphasized by the fact that it is an essential constituent of many organic compounds that are vital for metabolic processes (Purekar et al., 1992).

Interaction effects between P sources and VAM-fungi inoculation: Sweet potato plants tended to express their best performance with the highest values of total sugars, starch, carbohydrates and total carotene contents under application of superphosphate + inoculation with VAM-fungi treatment. However, it was noticed that the differences between superphosphate form+VAM-fungi inoculation and rock phosphate source+VAM-fungi inoculation treatments with respect to starch and TSS were not high enough to be significant (Table 7). The symbiotic association between VAM-fungi and plant roots allows the fungus to supply the growing plants with N, certain micronutrients including P and some hormones, which in turn enhances the uptake of nutrients and increases chemical composition (Hassan et al., 2005b).

Interaction effects between P levels and VAM-fungi inoculation: The results in Table 7 observed higher contents of most chemical composition of sweet potato roots at any level of P used with VAM-fungi inoculation. Meanwhile, the highest P level (100% P₂O₅) in conjunction with VAM-fungi inoculation treatment generated the highest values of starch and carbohydrates of root tissues. Similar treatment reflected higher mean value for total sugars in the first season (Table 7). On the other hand, the highest content of either TSS or total carotene was recorded for fertilizing plants treated with the highest P level + un-inoculated VAM-fungi.

Table 7:	Root chemical composition of sweet potato plants ‘Abees’ cv. as affected by integrated effects of P sources xP levels, P sources xVAM-fungi inoculation and P levels xVAM-fungi inoculation treatments, during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

Table 8:	Integrated effects among P sources, P levels and VAM-fungi inoculation treatments on root yield and quality traits of sweet potato ‘Abees’ cv. during 2007 and 2008 summer seasons
Values with the same letter(s) in the same column in each season are not significantly different using Duncan’s multiple range test at 0.05 probability level

These results are in line with those of Hassan et al. (2005b), who found that the highest content of total sugars and carbohydrates in roots were obtained as a result of interaction between VAM-fungi + the highest P rate (60 kg P₂O₅ fed.^-1).

Integrated effects among P sources, P levels and VAM-fungi inoculation: On the subject of the interaction among various P sources, P levels and VAM-fungi inoculation treatments on root production and quality of sweet potato ‘Abees’ cv., the obtained results showed that the highest significant values for both total and marketable root yield (13.566-16.600 ton fed.^-1) and (9.466-14.330 ton fed.^-1) in the two seasons, respectively as well as root quality traits (total sugars and carbohydrates) were obtained by the application of chemical superphosphate source at the highest P level with the presence of VAM-fungi inoculation treatment in comparison with all other treatments (Table 8). Hence, the obtained results can support the recent concept of Yeng et al. (2012), who reported that the use of organic fertilizer to supplement inorganic fertilizer use, as an integrated management strategy, is more importance for detecting several aims such as reducing soil mineral input cost, maximizing yield and sustaining sweet potato crop as well as other vegetable crops.

CONCLUSION

Integrated mineral superphosphate form under the recommended P level with VAM-fungi inoculation treatment improved sweet potato ‘Abees’ cultivar growth and development which led to increased root productivity and enhanced root quality more than either using inorganic fertilizer or bio-phosphate fertilizer (VAM-fungi) alone.