Soil organic carbon (SOC) is the principal factor contributing to enhanced soil fertility and also functions as the major carbon sink within terrestrial ecosystems. Applying fertilizer is a crucial agricultural practice that enhances SOC and promotes crop yields. Nevertheless, the response of SOC, active organic carbon fraction and hay yield to nitrogen and phosphorus application is still unclear. The objective of this study was to investigate the impact of nitrogen-phosphorus interactions on SOC, active organic carbon fractions and hay yield in alfalfa fields. A two-factor randomized group design was employed in this study, with two nitrogen levels of 0 kg·ha-1 (N0) and 120 kg·ha-1 (N1) and four phosphorus levels of 0 kg·ha-1 (P0), 50 kg·ha-1 (P1), 100 kg·ha-1 (P2) and 150 kg·ha-1 (P3). The results showed that the nitrogen and phosphorus treatments increased SOC, easily oxidized organic carbon (EOC), dissolved organic carbon (DOC), particulate organic carbon (POC), microbial biomass carbon (MBC) and hay yield in alfalfa fields, and increased with the duration of fertilizer application, reaching a maximum under N1P2 or N1P3 treatments. The increases in SOC, EOC, DOC, POC, MBC content and hay yield in the 0–60 cm soil layer of the alfalfa field were 9.11%-21.85%, 1.07%-25.01%, 6.94%-22.03%, 10.36%-44.15%, 26.46%-62.61% and 5.51%-23.25% for the nitrogen and phosphorus treatments, respectively. The vertical distribution of SOC, EOC, DOC and POC contents under all nitrogen and phosphorus treatments was highest in the 0–20 cm soil layer and tended to decrease with increasing depth of the soil layer. The MBC content was highest in the 10–30 cm soil layer. DOC/SOC, MBC/SOC (excluding N0P1 treatment) and POC/SOC were all higher in the 0–40 cm soil layer of the alfalfa field compared to the N0P0 treatment, indicating that the nitrogen and phosphorus treatments effectively improved soil fertility, while EOC/SOC and DOC/SOC were both lower in the 40–60 cm soil layer than in the N0P0 treatment, indicating that the nitrogen and phosphorus treatments improved soil carbon sequestration potential. The soil layer between 0-30 cm exhibited the highest sensitivity index for MBC, whereas the soil layer between 30-60 cm had the highest sensitivity index for POC. This suggests that the indication for changes in SOC due to nitrogen and phosphorus treatment shifted from MBC to POC as the soil depth increased. Meanwhile, except the 20–30 cm layer of soil in the N0P1 treatment and the 20–50 cm layer in the N1P0 treatment, all fertilizers enhanced the soil Carbon management index (CMI) to varying degrees. Structural equation modeling shows that nitrogen and phosphorus indirectly affect SOC content by changing the content of the active organic carbon fraction, and that SOC is primarily impacted by POC and MBC. The comprehensive assessment indicated that the N1P2 treatment was the optimal fertilizer application pattern. In summary, the nitrogen and phosphorus treatments improved soil fertility in the 0–40 cm soil layer and soil carbon sequestration potential in the 40–60 cm soil layer of alfalfa fields. In agroecosystems, a recommended application rate of 120 kg·ha-1 for nitrogen and 100 kg·ha-1 for phosphorus is the most effective in increasing SOC content, soil carbon pool potential and alfalfa hay yield.

Nitrogen (N) and phosphorus (P) fertilizers change the morphological structure and effectiveness of P in the soil, which in turn affects crop growth, yield, and quality. However, the effects and mechanism of combined N and P application on the content of P fractions and the transformation of effective forms in alfalfa (Medicago sativa L.) production is unclear. This experiment was conducted with four levels of N: 0 (N0), 60 (N1), 120 (N2) and 180 kg·ha-1 (N3); and two levels of P (P2O5): 0 (P0) and 100 kg·ha-1 (P1). The results indicated that, under the same N level, P application significantly increased soil total N, and total P, available P, and content of various forms of inorganic P when compared to no P application, while decreasing the content of various forms of organic P and pH value. In general, under P0 conditions, soil total N content tended to increase with increasing N application, while total P, available P content, pH, inorganic P content in all forms, and organic P content in all forms showed a decreasing trend. When compared to no N application, insoluble P (Fe-P, O-P, Ca10-P) of the N application treatments was reduced 2.80 - 22.72, 2.96 - 20.42, and 5.54 - 20.11%, respectively. Under P1 conditions, soil total N and O-P tended to increase with increasing N application, while, pH, Ca2-P, Al-P, Fe-P, Ca10-P, and organic P content of each form tended to decrease. Total P, available P, and labile organic P (LOP) of N application reduced 0.34 - 8.58, 4.76 - 19.38, and 6.27 - 14.93%, respectively, when compared to no application. Nitrogen fertilization reduced the soil Ca2-P ratio, while P fertilization reduced soil Fe-P, moderately resistant organic P (MROP), and highly resistant P (HROP) ratios, and combined N and P elevated the Ca8-P to LOP ratio. The results of redundancy analysis showed that soil total N content, available P content, and pH were the key factors affecting the conversion of P fractions in the soil. Nitrogen and P reduced the proportion of soil insoluble P, promoted the activation of soil organic P, resulting in accumulation of slow-acting P in the soil, thereby improving the efficiency of soil P in alfalfa production.

This study explored the effects on the P content of various organs and parts of alfalfa and soil available P content, of inoculation with arbuscular mycorrhizal fungi（AMF） and phosphorus-solubilizing bacteria（PSB） in different proportions with and without phosphorus（P） application. The ultimate aim of this research was to provide a theoretical basis for improving the efficiency of phosphorus fertilizer utilization for the production of high-quality, high-yielding alfalfa by formulating a scientifically based and appropriate fertilization system. A pot experiment was conducted with a randomized block design. There were 10 treatments in total, consisting of five inoculation ratios（AMF∶PSB） treatments [3∶7（J 1 ）, 4∶6（J 2 ）, 5∶5（J 3 ）, 6∶4（J 4 ） and 7∶3（J 5 ）] and two phosphorus（P 2 O 5 ） application levels [0（P 0 ） and 150 mg·kg -1 soil（P 1 ）]. The plant P content, soil P content, dry matter yield of alfalfa and P use efficiency were measured. Correlation analyses were conducted to clarify the relationship between the alfalfa plant P content, soil P content, dry matter yield, and P use efficiency. The best combination of bacteria and phosphorus conditions for promoting alfalfa growth was identified using a membership function analysis. It was found that, under the same inoculation conditions, the contents of plant P, stem P, leaf P, flower P, root P, upper 1/3 plant P of aboveground part, middle 1/3 plant P of aboveground part, lower 1/3 plant P of aboveground part, rhizosphere soil available P, non-rhizosphere soil available P and the dry matter yield were all significantly greater in the P 1 treatments than in the P 0 treatments（P＜0. 05）. Under the same P application conditions, the plant P, stem P, leaf P, flower P, root P, upper 1/3 plant P of aboveground part, middle 1/3 plant P of aboveground part, lower 1/3 plant P of aboveground part were all significantly greater in the J 5 treatment than in the J 1 , J 2 , J 3 and J 4 treatments（P＜0. 05）, and the dry matter yields of J 1 , J 2 , J 3 and J 4 treatments were significantly greater than that of the J 5 treatment（P＜0. 05）. The P content of each organ of alfalfa ranked: flower＞leaf＞root＞stem; the P content of each plant fraction ranked: upper 1/3 plant of aboveground part＞middle 1/3 plant of aboveground part＞lower 1/3plant of aboveground part and the soil available P content ranked: rhizosphere soil＞non-rhizosphere soil. Across the bacterial-P coupling treatments, the total dry matter yield and P use efficiency of alfalfa both reached their maximum values(49. 31 g·pot -1 and 27. 23%, respectively) in the J 4 P 1 treatment. Correlation analysis showed that the P content of alfalfa plants, rhizosphere soil P content, non-rhizosphere soil P content, P use efficiency and total dry matter yield were positively correlated with each other. Among these, total dry matter yield was highly significantly positively correlated with rhizosphere soil available P content and P use efficiency（P＜0. 01）, and was significantly positively correlated with non-rhizosphere soil available P content（P＜0. 05）. According to the ranking of membership function values, the top three were J 1 P 1 , J 3 P 1 and J 4 P 1 . Therefore, a phosphorus application rate of 100mg·kg -1 P（P 2 O 5 ） and double inoculation of AMF and PSB in the ratio 3∶7 can greatly improve alfalfa P nutrition level, thereby increasing the dry matter yield.

为探讨不同施磷水平下接种丛枝菌根真菌(AMF)和解磷细菌(PSB)对紫花苜蓿光合特性及生物量的影响,提高紫花苜蓿的磷肥利用效率及优质高产研究、制定科学合理的施肥制度提供理论依据.采用随机区组设计进行盆栽试验,设置4个接菌处理[未接菌对照组(CK,J0)、单接巨大芽孢杆菌(Bm,J1)、单接摩西管柄囊霉(Fm,J2)和双接菌(Bm×Fm,J3)]和4个施磷(P2O5)水平[0(P0)、50(P1)、100(P2)和150 mg·kg-1(P3)],共计16个处理.通过对紫花苜蓿的叶片净光合速率(Pn)、蒸腾速率(Tr)、气孔导度(Gs)、胞间CO2浓度(Ci)、光能利用效率(LUE)、水分利用效率(WUE)、叶绿素[Chl(a+b)]含量和生物量进行测定,并通过相关性分析明确紫花苜蓿各光合指标与生物量之间的关系,通过隶属函数分析筛选出有利于促进紫花苜蓿生物量形成的最佳菌磷耦合模式,明确施磷和接种解磷菌对紫花苜蓿光合特性及生物量的影响.结果表明:相同接菌处理下,紫花苜蓿叶片Pn、Tr、Gs、LUE、WUE、Chl(a+b)和生物量均随施磷水平的提高而呈先升高后降低的趋势,在P2处理达到最大值,且施磷处理均显著大于未施磷处理(P<0.05),而Ci则随施磷水平的提高呈逐渐降低的趋势,且施磷处理均显著低于未施磷处理(P<0.05);相同施磷处理下,紫花苜蓿叶片的Pn、Tr、Gs、LUE、WUE、Chl(a+b)和生物量均为接菌处理显著大于未接菌处理(P<0.05),且除Chl(a+b)外,Pn、Tr、Gs、LUE、WUE、Chl和生物量均为在J3处理达到最大值,而Ci显著低于未接菌处理(P<0.05),在J3处理达到最小值.相关性分析表明,Ci与Pn、Tr、Gs、LUE、WUE、Chl(a+b)和生物量均呈显著负相关(P<0.05),其余各指标间均两两呈显著正相关(P<0.05).根据隶属函数值大小排序,排名前3位的分别为J3P2、J2P2和J3P1.双接种丛枝菌根真菌与解磷细菌,并施磷100 mg·kg-1的菌磷耦合模式,能够显著提高紫花苜蓿叶片的光合特性并增加叶绿素含量,进而有利于促进紫花苜蓿生物量的形成.

为探讨不同氮磷配施条件下紫花苜蓿细根周转及不同土层分布动态特征,分析苜蓿细根周转各指标之间的关系.采用双因素随机区组设计进行田间试验,设置4个施磷水平[0(P0)、50(P1)、100(P2)和150 kg·hm-2(P3)]和两个氮水平[0(N0)和120 kg·hm-2(N1)],共计8个处理,通过微根管根系监测0～60 cm的土层细根周转特征.结果表明:在相同施氮条件下,随着施磷量的增加,紫花苜蓿细根总现存量、细根表面积密度、细根生产量和死亡量呈先增加后降低的趋势,在P2条件下达到最大值,且P1、P2处理显著大于P0处理(P<0.05),在相同施磷条件下,N1处理显著大于N0处理.在不同土层中,在相同施氮条件下,随着施磷量的增加,苜蓿细根现存量在0～30 cm土层中呈先增加后降低的趋势,在0～15 cm土层中,P2处理苜蓿细根现存量显著高于其他处理(P<0.05).不同处理下,苜蓿细根现存量主要集中在15～30 cm土层.在相同施氮条件下,随施磷量的增加,苜蓿细根周转率呈先降低后增加的趋势.细根周转率受细根现存量与细根死亡动态变化的影响较大.细根死亡量与周转率拟合的相关系数最大,拟合效果最好.综上所述,当施磷(P2O5)量为100 kg·hm-2、施氮(N)量为120 kg·hm-2时,能够显著增加苜蓿细根的现存量和根表面积密度,进而促进苜蓿根系周转和生长.

为了探究不同秋眠级紫花苜蓿(Medicago sativa L.)生产性能和越冬能力的差异,本研究以国内外26个不同秋眠级紫花苜蓿品种为试验材料,对紫花苜蓿生长性状、营养品质、越冬率及其综合评价进行研究.结果表明:紫花苜蓿的单株生物量、株高、茎粗、粗蛋白含量随着秋眠级的升高呈先增加后降低的趋势,而茎叶比、中性洗涤纤维、酸性洗涤纤维含量随秋眠级的升高呈先降低后增加的趋势,越冬率则随着秋眠级的升高呈逐渐下降的趋势.主成分分析综合评价得分由高到低依次为秋眠级4级＞3级＞2级＞5级＞6级＞8级＞7级＞9级＞10级,聚类分析表明秋眠级4级和3级紫花苜蓿为综合指标最优的一类.因此,秋眠级4级的紫花苜蓿最接近理想品种,其次为秋眠级3级,上述2个秋眠级紫花苜蓿均适合在新疆石河子地区种植.

为探究不同施氮量对紫花苜蓿根际土壤微生物数量、酶活性、干草产量的影响,设置4个施氮梯度,施氮量分别为:0kg/hm2(N0)、60kg/hm2(N1)、120kg/hm2(N2)和180kg/hm2(N3),测定紫花苜蓿根际土壤微生物数量、酶活性及干草产量,并进行综合分析.结果表明:紫花苜蓿根际土壤细菌、真菌、放线菌数量及根际土壤碱性磷酸酶、过氧化氢酶活性均随施氮量的增加呈先升高后降低的趋势.当施氮量达120kg/hm2(N2)时,紫花苜蓿根际土壤细菌、真菌、放线菌数量均达到最大值,分别为7.56、6.11和6.351g cfu/g;土壤碱性磷酸酶、过氧化氢酶活性均达到最高,分别为21944.44和944.04U/(g·d),且显著高于其他处理(P＜0.05).苜蓿干草产量在N2处理下达到最大值,为22.46t/hm2.通过综合评定,当施氮量为120kg/hm2(N2)时,有利于增加苜蓿根际土壤微生物数量及促进根际土壤酶活性,进而提高苜蓿干草产量.

为探讨滴灌条件下紫花苜蓿光合日进程及光合产物分配对不同施磷水平的响应,明确苜蓿地上光合产物与叶片光合参数之间的关系,通过田间试验,设置0(CK)、50(P1)、100(P2)和150 kg·hm-2(P3)4个施磷(P2O5)水平,在紫花苜蓿初花期选择典型晴天(10:00-20:00),采用Li-6400便携式光合仪测定紫花苜蓿叶片光合指标及环境因子日进程,并对苜蓿植株叶、茎、根中的可溶性糖和淀粉含量进行测定.结果表明,不同施磷水平下紫花苜蓿均有光合"午休"现象,气孔限制是净光合速率(Pn)下降的主要因素.光合有效辐射(PAR)对苜蓿叶片Pn影响最大,而大气CO2浓度(Ca)、大气温度(Ta)、大气相对湿度(RH)和叶面饱和蒸气压亏缺(Vpdl)对Pn的影响次之.主成分分析发现,苜蓿的光合效率由高到低依次是P2>P1>P3>CK.当施磷水平为100 kg·hm-2时,苜蓿的干草产量显著提高,紫花苜蓿叶片Pn、蒸腾速率(Tr)和水分利用效率(WUE)的日进程均较对照有所提高,叶片胞间CO2浓度(Ci)日进程均较对照显著降低.叶、茎和根的可溶性糖含量分别提高了11.6％、5.0％和4.6％,淀粉含量分别提高了15.2％、9.6％和5.3％,可溶性糖和淀粉的资源分配更多地表现为对叶的分配增加,对茎和根相对较少.因此,适当施磷能提高紫花苜蓿叶片的光合效率,进而显著促进苜蓿的生长发育,施磷(P2O5)量为100 kg·hm-2时对紫花苜蓿光合的提升效果最为明显.

通过研究不同氮素水平下滴灌苜蓿叶片形态特征、光合日变化规律,分析不同施氮水平下滴灌苜蓿光合日变化、叶片形态与干物质产量的关系,以期进一步揭示施氮对紫花苜蓿干物质及产量形成的影响机制,进而为优化实际生产中紫花苜蓿的氮管理策略提供理论依据.采用单因素随机区组设计,设置0(CK)、60(N1)、120(N2)和180 kg·hm-2(N3)共4个施氮水平,在紫花苜蓿初花期对光合日变化、叶片形态、叶片氮含量和苜蓿产量构成进行测定.结果表明,施氮处理下苜蓿的叶片净光合速率、蒸腾速率和水分利用效率均高于不施氮处理,施氮处理的苜蓿叶片胞间CO2浓度低于不施氮处理.对净光合速率和蒸腾速率综合影响最大的环境因子是光合有效辐射.随着施氮量的增加,紫花苜蓿的叶长、叶宽、叶面积、比叶重,以及叶片干重、茎秆干重、干物质产量、叶片氮含量、淀粉和可溶性糖含量均呈先增加后降低的趋势.不同施氮水平下,对叶片形态结构影响最大的为叶面积,其次分别为比叶重、叶长和叶宽,对苜蓿干物质产量影响从大到小依次为叶片氮含量>净光合速率>叶面积>蒸腾速率>比叶重.不施氮和高氮处理下光合速率下降主要是因为光合活性受到抑制,属于非气孔因素.基于主成分分析,干物质产量、叶片形态以及光合作用综合得分最高的为N2处理,其次分别为N3、N1和CK处理.因此,施氮肥有助于紫花苜蓿光合面积和光合速率的协同改进,有利于光合产物的生成,从而促进苜蓿干物质产量的增加,在施氮量为120 kg·hm-2时提升效果最为明显.

为探究不同施磷量对滴灌紫花苜蓿根际土壤微环境及干草产量的影响,试验设置4个施磷(P2O5)梯度:0(P0),50(P1),100(P2)和150 kg·hm-2(P3),采用滴灌模式随水滴施,测定紫花苜蓿根际土壤微生物数量、酶活性、土壤理化性质及干草产量,并进行综合分析.结果表明,紫花苜蓿根际土壤真菌、放线菌数量及脲酶、过氧化氢酶活性在P2处理下达到最大值,分别为6.04,6.15 lg cfu·g-1及933.54,945.91 U·g-1·d-1;细菌数量在P3处理达到最大值,为7.14lgcfu·g-1;碱性磷酸酶活性在P1处理达到最大值,为14 939.80 U·g-1·d-1.根际土壤全磷及有效磷含量随着施磷量的增加呈逐渐增加的趋势,至P3处理达到最大值,但根际土壤pH及含水率均为未施磷显著高于其他处理(P＜0.05).苜蓿干草产量在P2处理达到最大值,为21.82 t·hm-2.综合评定,适宜的施磷量(100 kg·hm-2)有利于改善紫花苜蓿根际土壤微环境,进而提高苜蓿干草产量.

为明确不同秋眠级紫花苜蓿品种间营养元素限制性和分配格局的差异性,以国内外收集的10个不同秋眠级紫花苜蓿品种为研究对象,分析其叶片和茎秆的碳(C)、氮(N)、磷(P)含量和化学计量比变化特征.结果表明,紫花苜蓿叶片和茎秆中N和P含量均随着秋眠级的升高呈先降低后升高的趋势,C含量以及C:N、C:P和N:P呈相反的趋势.不同秋眠级苜蓿品种叶片中各元素含量为:秋眠级为4级的新疆大叶苜蓿叶片C含量、秋眠级为9级的WL656HQ叶片N含量均显著大于其他各秋眠级苜蓿品种(P＜0.05).苜蓿茎秆C含量在各秋眠级苜蓿品种间均差异不显著,N、P含量变化与苜蓿叶片相同,且第2茬叶片和茎秆C、N和P含量的变异系数均低于第1茬,两茬表现出相同的规律.除第2茬苜蓿的叶片外,秋眠级对苜蓿叶片和茎秆P含量的影响最大,其次为N含量,对C含量的影响最小.不同茬次间,秋眠级对苜蓿叶片C:P的影响最大,对N:P的影响最小.苜蓿叶片C含量与叶片和茎秆C ∶ N、C∶P呈显著正相关(P＜0.05),而与叶片和茎秆N含量呈显著负相关(P＜0.05);叶片N、P含量分别与茎秆N、P含量呈显著正相关(P＜0.05).叶片C ∶N、C ∶P分别与茎秆C ∶N、C ∶P呈显著正相关(P＜0.05),与茎秆N含量呈显著负相关(P＜0.05);茎秆C含量与茎秆C ∶N呈显著正相关(P＜0.05),茎秆N含量与茎秆C ∶ N呈显著负相关(P＜0.05);茎秆C∶P与茎秆N ∶P呈显著正相关(P＜0.05).非秋眠型和秋眠型苜蓿的生长受N元素限制,中等秋眠型苜蓿的生长受N和P元素共同影响,在不同秋眠级苜蓿生产过程中,应重点考虑N、P的供需平衡.

为探讨滴灌条件下不同花期紫花苜蓿的农艺性状特征,明确苜蓿各农艺性状指标与干草产量之间的关系,本研究选取苜蓿生长的4个花期,即孕蕾盛期(S1)、初花前期(开花5％,S2)、初花后期(开花10％,S3)和盛花期(开花50％,S4),进行苜蓿不同花期的株高、茎粗、茎叶比及干草产量的测定.结果 表明:相同茬次,随着花期的推进,苜蓿的株高、茎粗、茎叶比及干草产量均呈逐渐增大的趋势,且盛花期各指标均显著大于孕蕾盛期(P＜0.05).紫花苜蓿的总干草产量为S2(刈割5茬)最大,其次分别为S3、S4、S1.不同花期滴灌苜蓿各农艺性状与苜蓿干草产量的关联度不同,相同花期,茎粗与苜蓿干草产量的关联度最大.在盛花期(开花50％,S4)刈割有利于苜蓿单茬干草产量的提高,但实际生产中往往还需要考虑苜蓿的营养品质、年总干草产量及刈割成本.因此,可适当提前刈割期,在初花后期(开花10％)刈割效果最佳.

Alfalfa (Medicago sativa L.) is an important forage legume in arid areas, but limited water resources and low fertilizer utilization have restricted its agricultural development. Meanwhile, studies on the effects of integrated water and phosphorus on production performance and water-use efficiency and phosphorus-use efficiency of alfalfa, especially on hay yield, phosphorus accumulation, and total phosphorus uptake are rarely reported under drip irrigation. The treatments were a factorial combination of three irrigation rates (5,250, 6,000, and 6,750 m3/ha per year) and four P rates (0, 50, 100, and 150 kg/ha per year) and consisted of 12 treatments for water and P management, arranged in a randomized complete block design with three replicates. Total hay yield and water-use efficiency and phosphorus-use efficiency of alfalfa in P2 treatment were significantly greater than those in the P1 and P3 treatments (p < .05), and the total hay yield of alfalfa with phosphorus application increased by 7.43%-29.87% compared with that in the nonphosphorus (P0) treatment under the same irrigation amount. The total phosphorus and available phosphorus concentrations in the 0-20 cm soil layer were greater than those in the 20-40 cm and 40-60 cm soil layers compared with those in the P0 treatment. Correlation analyses showed that total hay yield was significantly positively correlated with total phosphorus uptake and water-use efficiency (p < .01). The accumulated phosphorus concentration was significantly positively correlated with total phosphorus and available phosphorus concentration (p < .01) and was positively correlated with the phosphorus-use efficiency (p < .05). The membership function method was used to evaluate all the indicators, and the three treatments that had the greatest influence on the production performance of alfalfa were, in order, W2P2 > W3P2 > W1P2. Therefore, an irrigation rate of 6,000 m3/ha and a phosphorus application rate of 100 kg/ha per year should be considered as the best management for both high yield and water-use efficiency and phosphorus-use efficiency of alfalfa.

Alfalfa (Medicago sativa L.) is an important forage legume in farming and animal husbandry systems. This study assessed the effects of arbuscular mycorrhizal fungi (AMF) and phosphate-solubilizing bacteria (PSB) on alfalfa growth under different phosphorus application levels. In this experiment, a complete randomized block design was used. The following four bacterial applications were used: inoculation of Funneliformis mosseae (Fm), inoculation of Bacillus megaterium (Bm), inoculation of mixed species (Fm × Bm) and noninoculation treatment (CK). Phosphorus (P) treatment was applied at the following four levels: 0 mg kg−1 (P0), 50 mg kg−1 (P1), 100 mg kg−1 (P2) and 150 mg P kg−1 (P3). The results showed that with the increase in phosphorus application, each index increased first and then decreased. The J2 treatment was significantly greater than the J0 treatment (P < 0.05) under the same bacterial treatment. In each cropping period the difference in each index to alfalfa was extremely significant under J, P treatment and J × P interactive treatment (P < 0.01). The indexes were compared by membership function. The priority order was as follows: J3P2 > J1P2 > J3P1 treatment. Therefore, when phosphorus was applied at 100 mg kg−1, the mixed inoculation of Fm × Bm was optimal, benefitting mycorrhiza growth and the production performance of alfalfa.