— Rotation and Fertilization Influence on Vesicular-Arbuscular Mycorrhizal Fungi J. R. Ellis,* W. Roder, and S. C. Mason ABSTRACT been shown to affect the host-symbiont relationship Vesicular-arbuscular mycorrhizal fungi (VAMF) can reduce (Busse and Ellis, 1985; Harinikumar and Bagyaraj, stress resulting from nutrient deficiencies, drought, and other factors. 1989; Menge et al., 1980; Raju et al., 1988, 1990a; The objective of this work was to measure the effect of soybean \Gly- Ross, 1971; Saif, 1981), the effect of most of these cine max (L.) Merr.] and grain sorghum [Sorghum bicolor (L.) Moench] properties on VAMF colonization and development rotation and fertilization on plant response and VAMF root coloni- under field conditions is limited. This may be a result zation and diversity, and relate effects to soil environment. Fertilizer of problems in sampling roots under field conditions treatments consisted of no fertilizer, N, and manure. Rooting densities (Brown and Scott, 1984) to examine the relationship correlated with previous crop, VAMF colonization, and soil NO,. Root between VAMF colonization, root growth, and soil colonization by VAMF was affected by previous crop, rooting density, environmental conditions. N fertilization, soil P, and water-filled pore space. Root colonization Harinikumar and Bagyaraj (1988) found that grow- by VAMF ranged from 93% at 15 cm to 15% at the 120-cm soil depth. ing a nonmycorrhizal plant in soil for one season re- Root density and VAMF colonization were least when soybean was duced mycorrhizal colonization by 13% and a fallow grown the previous year and manure was applied. Root colonization period reduced colonization by 40%. Long-term con- by VAMF for control, N, and manure treatments were 54, 53, and tinuous cropping, depending on plant species, may 30%, respectively, for continuous soybean and 61, 55, and 44%, re- increase or decrease VAMF colonization and spore spectively, for soybean from rotation plots. Root colonization by VAMF numbers (Kruckelmann, 1975; Strzemska, 1975; Black for control, N, and manure treatments were 69, 59, and 54%, re- spectively, for continuous grain sorghum and 56, 48, and 31%, re- and Tinker, 1979; Harinikumar and Bagyaraj, 1988, spectively, for grain sorghum from rotation plots. These agricultural 1989). Soil fertility practices also may affect VAMF soils contained a diverse mixture of 26 VAMF species, which is prob- colonization of roots and root development (Menge et ably a major factor in the region's soil productivity. stressed al., 1980). Low-organic-matter soils tend to enhance due to cropping system or fertilizer practice have greater VAMF col- VAMF root colonization when green manure is ap- onization and VAMF activity. A diverse VAMF population could in- plied (Harinikumar and Bagyaraj, 1989). However, crease the ability of VAMF to respond to different stresses. Hayman (1980) and Strzemska (1975) showed N and P fertilizers depress root colonization by VAMF. The objectives of this work were to examine the diversity of VAMF in a crop-rotation system and de- OLONIZATION OF PLANT ROOTS by VAMF often termine the effect of soybean-sorghum crop rotation, C reduces adverse effects of plant stresses (water, fertilizer N, and manure on VAMF colonization and nutrient, and environmental factors) and has beneficial root development of soybean and grain sorghum, and effects in maintaining crop yields. Grain sorghum and to relate these effects to the soil environment. soybean are colonized by VAMF. Rotating crops and application of N fertilizer may affect VAMF coloni- MATERIALS AND METHODS zation and final crop yield. Clegg (1982) and Roder Research plots were established in 1980 on a Sharpsburg et al. (1989a) showed that rotating these crops im- silty clay loam (fine, montmorillonitic, mesic Typic Argiu- proved crop yield, compared with continuous crop- doll) at the University of Agricultural Research ping. They were unable, however, to explain the and Development Center, Mead, NE. The treatments con- sisted of soybean and grain sorghum growing in rotation rotation effect based solely on N availability. Since and in continuous cropping receiving no fertilizer (control), VAMF colonize the root system and serve as a link cattle manure (15.8 Mg dry matter ha-1 yr-1 or 160-250 between plant roots and soil, changes in cropping sys- kg N ha-1 yr-1), and fertilizer N (44 kg ha-1 for soybean, tems that affect the soil environment should affect 90 kg ha-1 for grain sorghum). Nitrogen was broadcast as both roots and VAMF. NH4NO3 (33% N) 3 to 4 wk after planting. The experiment The rate and patterns of root growth vary with soil was conducted as a randomized complete-block design with water, soil strength, bulk density, soil aeration, soil a split-plot treatment arrangement. The cropping system temperature, chemical and biological soil properties, was the whole plot and N treatments were subplots. Sub- plant genetic potential, and climate (Brown and Scott, plots were 12 rows wide (0.75-m spacing) and 7.8 m long. Four replicates were used and all observations were limited 1984; Eavis, 1972; Jones, 1983; Robertson et al., to the four center rows, using the middle 5.4 m of row 1980). Although soil fertility, soil water, organic mat- length. Manure was applied in spring. Tillage was uniform ter, soil aeration, soil temperature, and plant host have for all treatments and consisted of disking after manure application and double disking before planting. Crops were J.R. Ellis, USDA-ARS, Keim Hall, East Campus, and S.C. Ma- planted with a six-row John Deere1 Max-Emerge planter. son, Dep. of Agronomy, Univ. of Nebraska, Lincoln, NE 68583; In 1986 and 1987, soybean (Pella 86 foundation seed, and W. Roder, International Research Inst., Vientiane, Laos. Contribution from the USDA-ARS in cooperation with the Agri- Univ. of Nebraska, Lincoln) inoculated with Bradyrhizo- cultural Research Div., Univ. of Nebraska, Lincoln. Published as Paper no. 9787 Journal Series. Partial financial support was pro- Abbreviations: VAMF, vesicular-arbuscular mycorrhizal fungi; vided by the International Sorghum and Collaborative Re- WFPS, water-filled pore space; HSD, highly significant differ- search Program (INTSORMIL) and USAID Grant DAN-1254-G- 00-0021. Received 3 June 1991. * Corresponding author. 1 Mention of companies or commercial products does not imply recommendation or endorsement by the USDA or the University Published in Soil Sci. Soc. Am. J. 56:789-794 (1992). of Nebraska-Lincoln over others not mentioned. 789 790 SOIL SCI. SOC. AM. J., VOL. 56, MAY-JUNE 1992

Table 1. Correlation coefficients (r) for selected soybean and sorghum parameters with vesicular-arbuscular mycorrhizal fungal colonization (VAMF), arbuscule formation (ARE), vesicle production (YES), and root length (RTL) for 1986 and 1987. Soybean Sorghum VAMF ARE VES RTL VAMF ARE VES RTL 1986 RTL 0.30 0.30 -0.16 _ 0.71 0.34 0.45* _ N Treatment -0.60 * -0.60 * -0.22 -0.17 -0.49 -0.26 0.39* -0.16 Previous crop 0.12 0.13 -0.39 0.88* 0.67 0.50* 0.21 0.63* Shoot wt. -0.33 -0.29 -0.40 * 0.06 -0.56 -0.08 -0.46* -0.58 * Grain wt. -0.30 -0.28 -0.49 * 0.19 -0.45 -0.25 0.36 -0.46* No. of heads -0.58 -0.39 0.51* -0.51 * 1987 RTL 0.31 0.38 -0.16 _ 0.41 0.29 0.09 _ N treatment -0.59 * -0.53 * -0.45 * 0.02 -0.69 ' -0.51 * -0.41 * -0.14 Previous crop 0.22 0.29 -0.49 * 0.64* 0.43 0.45* -0.20 0.59* Shoot wt. 0.22 0.25 -0.17 0.23 -0.16 -0.15 0.22 -0.43 * * Significant correlation at the P < 0.05 probability level.

Table 2. Correlation coefficients (r) at different soil depths between soybean and sorghum vesicular-arbuscular mycorrhizal fungal root colonization (VAMF), arbuscules and vesicle production, root length density (RTL), available P (P), soil NO3-N, and soil water-filled pore space (WFPS) for 1986 to 1987. Soybean Sorghum Soil depth VAMF RTL P NO3-N WFPS VAMF RTL P NOj-N WFPS cm VAMF root colonization 7.5 0.71 -0.56* -0.35 * 0.26 - 0.50 * -0.37 -0.19 -0.01 15 0.75 -0.31 -0.28 * 0.37 - 0.57 * -0.48 * -0.15 -0.07 30 0.77 -0.20 -0.50 * 0.03 - 0.77 * -0.43 * -0.48 * 0.38 60 0.70 -0.30 * 0.51 * - 0.76 * -0.45 * -0.22 90 0.61 -0.19 0.47 * - 0.79 * -0.52 * 0.48 120 0.57 -0.02 0.44 - 0.53 * -0.19 0.27 Arbuscule formation 7.5 0.92* 0.64 -0.44 * -0.29 * -0.06 0.88 0.37* -0.21 -0.19 -0.01 15 0.94* 0.72 -0.10 -0.26 -0.30 0.87 0.64* -0.14 -0.13 0.33 30 0.90* 0.73 -0.18 -0.43 * -0.38 0.87 0.81* -0.06 -0.41 * 0.16 60 0.82* 0.64"( __ -0.24 -0.02 0.77 0.53* -0.38 * -0.49 90 0.94* 0.55 * __ -0.19 0.16 0.88 0.65* -0.38 * 0.26 120 0.97* 0.46" __ 0.03 0.34 0.89 0.37* -0.12 0.51 Vesicle production 7.5 0.88* 0.46" -0.32 -0.14 0.39 0.66 0.34* -0.31 -0.06 -0.08 15 0.75 * 0.47* -0.13 0.10 0.59 * 0.71 0.30* -0.41 * -0.08 -0.28 30 o:6i* 0.14 -0.19 0.05 0.32 0.53 0.41* -0.35 -0.25 -0.04 60 0.75* 0.42' _ 0.16 0.53 * 0.51 0.72* -0.17 0.01 90 0.75* 0.41* _ -0.21 0.38 * 0.62 0.52* -0.24 0.05 120 0.72* 0.33 -0.02 0.34 0.66 0.47* -0.18 0.32 Root length 7.5 0.71* -0.04 -0.22 0.27 0.50 -0.17 -0.17 0.22 15 0.75* -0.34 -0.26 0.10 0.57 0.07 -0.14 0.05 30 0.77* -0.34 -0.45 * 0.03 0.77 0.04 -0.39 * 0.20 60 0.70* -0.21 -0.21 0.76 -0.28 0.17 90 0.61 * -0.25 0.20 0.79 -0.47 * 0.28 120 0.57* -0.20 0.47 0.53 -0.18 0.21 ' Significantly correlated at the P < 0.05 probability level. bium japonicum (Nitragen Co., Milwaukee, WI), and grain to the same depth intervals from each plot to determine soil sorghum (Pioneer Hybrid 8358) were used in the experi- bulk density (Blake and Hartge, 1986), gravimetric soil ments. Root samples were taken in 1986 and 1987 using a water content (Gardner, 1986) and soil NO3 content (Kee- tractor-mounted Giddings hydraulic soil coring machine ney and Nelson, 1982). Hydropneumatic elutriation was (Giddings Machine Co., Ft. Collins, CO). Sampling dates used to extract roots (Smucker, 1984). Roots were collected were 31 July (Replicate 1) and 27 to 28 August (Replicates on a 0.42 mm sieve. After removing the debris (dead plant 2 through 4) in 1986 and 3 to 5 August in 1987. In 1986, material, seeds, etc.), root length was estimated by the sampling of Replicates 2 through 4 had to be delayed be- modified line-intercept method (Tennant, 1975). Root length cause of wet soil conditions. Three 4.46-cm-diam. cores was expressed in length per volume (cm cm-3) and length (subsamples) were taken from each plot. Samples were taken per area (km m~2). Roots were then dried at 65 °C and directly over the plants and at 19 and 38 cm from the planted weighed. Roots were stained using a modified trypan blue row. Total depth sampled was 120 cm. Cores were divided staining procedure (Kormanik and McGraw, 1982) and as- into depth intervals of 0 to 7.5, 7.5 to 15, 15 to 30, 30 to sessed microscopically at 200 x for VAMF root coloniza- 60, 60 to 90, and 90 to 120 cm. Subsamples of the same tion (Bierman and Linderman, 1981), arbuscule formation, depth were composited, placed in a plastic bag, and stored and vesicle production. The VAMF evaluations were ex- at -20 °C until root extraction. An additional core was taken pressed as a percentage of roots colonized then multiplied ELLIS ET AL.: VAM FUNGAL ROOT COLONIZATION 791 by the root length to obtain root-length colonization. Spores of VAMF were sieved from soil collected from rotation plots (Gerdemann and Nicolson, 1963) and identified from 30 original descriptions, published keys, and consultations with VAMF taxomonists. 60 Shoot dry-matter production was measured from 1 m of row at the time of root core collection. At maturity, grain 90 yield and number of heads (pods) per unit of row were also determined from these samples. Soil WFPS was determined 120 using the method of Linn and Doran (1984). Soil N was O, determined after sample digestion (Nelson and Sommers, Q 30 1980; Keeney and Nelson, 1982) and soil P by the Bray procedure (Olsen and Sommers, 1982) 60 Statistical analysis and mean comparisons were made fol- lowing guidelines by Steel and Torrie (1980) using Statis- 90 tical Analysis Systems procedures (SAS Institute, 1988) for correlations, analysis of variance, and Tukey's procedure 120 0 1 2 3 0 20 40 60 80 0 20 40 60 (HSD < = 0.05) to separate means. -3 Root length, cm cm VAM, % of root Arbuscules, % of root Fig. 1. Root length, vesicular-arbuscular mycorrhizal (VAM) RESULTS AND DISCUSSION fungi colonization. VAM Fungal arbuscule production in continuous-soybean and sorghum-soybean rotation plots with The samples for roots and VAMF were taken at or N-fertilizer or manure treatments. after flowering and heading phases of grain sorghum and soybean development. Although earlier observa- tions showed that 70% of the root system can be col- for grain sorghum, VAMF root colonization and ve- onized 4 wk after seedling emergence in this soil, the sicle production increased with increased WFPS. heading, pod production, and filling period were tar- Both high P and N have been reported to decrease geted for this experiment because this is the period VAMF colonization (Abbott and Robson, 1984; John- during which water, heat, and nutrient stress would son et al., 1984). In this experiment, VAMF coloni- most likely affect plant yield and the plants had the zation rates at the soil depths where there was a negative largest quantity of roots. Barber and Silberbush (1984) correlation with N and P were as great as 93% in related soybean grain yield of several soybean culti- soybean (Fig. 1) and 89% in grain sorghum (Fig. 2), vars to total root length at the R6 stage, when plants in spite of high N and P levels. The negative effects are sensitive to stress. Busse and Ellis (1985) and Ellis of P, therefore, were of little significance at these high et al. (1985) showed that VAMF reduced water stress values. When the average VAMF colonization of soy- at this stage. bean for all depths is compared during the 2 yr (Table Vesicular-arbuscular mycorrhizal fungal coloniza- 3), the results are not consistent. In 1986, the contin- tion and plant root development varied between years. uous-soybean control had 15 and 33% increases in Differences in plant and root between years VAMF root colonization and arbuscule formation, re- were attributed to differences in temperature and rain- spectively, compared with the rotation plots, whereas fall patterns (Roder et al., 1989b). However, differ- in 1987 there were 39 and 44% decreases. Root-length ences between rotation and manure treatment effects density of the continuous-soybean treatments was re- were consistent. Root and VAMF parameters mea- duced compared with that for the rotation treatments. sured were usually correlated for grain sorghum but seldom for soybean (Table 1). Almost all VAMF pa- rameters were negatively correlated with N treatment 0 in both years. In grain sorghum, there were positive correlations between total root length and average 30 VAMF colonization and vesicle production in 1986 60 Continuous and with VAMF root colonization in 1987. In 1986 Sorghum and 1987, grain sorghum as the preceding crop, com- 90 pared with soybean as the preceding crop, stimulated VAMF colonization, arbuscule formation, and root 120 o. growth of both grain sorghum and soybean. 01 The 2 yr of data were combined when correlations Q 30 at different depths of soil were examined (Table 2). Control At almost all depths, most VAMF parameters were 60 Nitrogen Manure significantly correlated with VAMF root colonization 90 and root length for both crops. Soybean and sorghum root length were greater if the preceding crop was 120 - grain sorghum rather than soybean. There was a trend 0 1 2 3 4 5 6 7 0 20 40 60 80 0 20 40 60 (significant in several instances) for reduced values of Root length, cm cm VAM, % of root Arbuscules, % of root VAMF and root parameters with increased soil NO3- Fig. 2. Root length, vesicular-arbuscular mycorrhizal (VAM) N values. Except in a few negatively correlated cases, fungi colonization, VAM fungal arbuscule production in available soil P had little relationship to VAMF and continuous-sorghum and soybean-sorghum rotation plots with root parameters. At a few depths for soybean, but not N-fertilizer or manure treatments. 792 SOIL SCI. SOC. AM. J., VOL. 56, MAY-JUNE 1992

Table 3. Effect of continuous and rotated soybean and N treatment on average vesicular-arbuscular mycorrhizal fungi (VAMF) root colonization and total soybean roots. 1986 1987 Roots with Roots with Roots with Roots with Treatment VAMF Arbuscules Roots VAMF arbuscules VAMF Arbuscules Roots VAMF arbuscules km m- • km m~ Continuous soybean Control 73 at 40 a 14.8 c 10.8 be 5.9 abc 35 ab 20 a 7.2 b 3.5 ab 2.0 ab N fertilizer 63 ab 26 ab 15.2 c 9.6 be 3.9 c 42 ab 21 a 9.3 ab 4.2 ab 2.1 ab Manure 39 b 18 b 14.3 c 5.6 c 2.6 c 21 b 11 a 7.2 b 2.1 b Mb Sorghum-soybean rotation Control 63 ab 30 ab 24.8 ab 15.8 ab 7.4 ab 58 a 36 a 10.8 a 5.9 a 3.7 a N fertilizer 66 a 32 a 26.3 a 17.4 a 8.5 a 43 ab 25 a 9.7 ab 4.3 ab 2.5 ab Manure 52 ab 25 ab 19.3 b 10.1 be 4.8 be 38 ab 22 a 10.2 ab 3.8 ab 2.2 ab t Numbers with same letters in a column are not significantly different at P < 0.05 according to Tukey's procedure.

Table 4. Effect of continuous and rotated sorghum and N treatment on average vesicular-arbuscular mycorrhizal fungi (VAMF) root colonization and total sorghum roots. 1986 1987 Roots with Roots with Roots with Roots with Treatment VAMF Arbuscules Roots VAMF arbuscules VAMF Arbuscules Roots VAMF arbuscules

at. 1,™ ™-2 Of. Continuous sorghum Control 73 at 28 a 36.8 a 26.8 a 10.1 a 64 a 38 a 17.6 a 11.3 a 6.8 a N fertilizer 62 ab 23 a 29.7 ab 18.6 abc 6.8 ab 56 a 32 a 17.1 a 9.6 ab 5.5 ab Manure 59 ab 21 a 31.6 ab 18.8 ab 6.5 ab 48 a 27 ab 17.9 a 8.7 be 4.9 ab Soybean-sorghum rotation Control 58 ab 17 a 24.1 ab 14.1 be 4.1 b 54 a 25 ab 15.6 a 8.4 be 4.1 b N fertilizer 51 be 17 a 22.8 b 11.6 be 3.9 b 44 a 23 ab 15.2 a 6.6 c 3.6 be Manure 33 c 10 a 23.2 ab 7.8 c 2.3 b 29 b 13 b 13.9 a 4.1 d 1.9 c t Numbers with same letters in a column are not significantly different at P < 0.05 according to Tukey's procedure.

Plant roots from the continuous-soybean manure plots Table 5. Previous crop and fertilizer effects on soybean and had 28 and 44% reductions in VAMF colonization, grain sorghum yield (from Roder et al., 1989). compared with the manure rotation plots. 1986 1987 In contrast to soybean plots, there was an increase N N of 12 to 107% in VAMF root colonization, arbuscule Treatment Control fertilizer Manure Control fertilizer Manure production, or root length in grain sorghum due to Mgha -i crop rotation and N treatment (Table 4). Mycorrhizal Soybean colonization ranged from 15 to 89% of the roots col- Continuous 2.49 2.34 2.64 2.29 2.29 2.22 onized at all depths (Fig. 2). The large number of Rotation 2.64 2.62 2.69 2.63 2.52 2.55 arbuscules produced in the roots, 8 to 67%, indicate Son'hum high VAMF activity. Some root segments contained Continuous 4.62 7.18 8.04 3.25 4.47 5.52 arbuscules in every cortical cell. Continuous grain Rotation 7.41 7.71 7.89 5.09 4.92 5.24 sorghum increased root-length density, VAMF root colonization, and arbuscule production at almost every depth, compared with soybean-grain sorghum rota- Both soybean and grain sorghum residues have been tion. The unfertilized control treatment also had greater reported to have an allelopathic effect on seedlings VAMF root colonization than the manure treatment at (Huber and Abney, 1986; Kalantari, 1981; Guenzi et almost all soil depths. This indicates that there was a al., 1967) and Roder et al. (1989b) speculated that greater need for VAMF due to nutrient or environ- reduction in soybean and grain sorghum root devel- mental stress on the plants. The grain yield of plants opment under rotations was due to soybean-residue in rotation with soybean and treated with manure ex- phytotoxicity. However, the soil system is dynamic cept for the 1987 soybean, had greater yield (Table and VAMF colonization can be stimulated by plant 5) and lower VAMF root colonization then plants after stress, whether environmental or nutrient deficiency, sorghum or with the other fertilizer treatments. and VAMF have been shown to stimulate root growth Soil water content, soil aeration, soil bulk density, (Raju et al., 1990b). The positive association found size distribution of soil aggregates, porosity, infiltra- between WFPS and VAMF soybean root colonization tion capacity, and soil strength are considered factors in 1986 indicates water could stimulate VAMF colo- influencing crop root development (Fahad, 1979; nization and enhance root development. Without VAMF Robertson et al., 1980; Jones, 1983; Gerik et al., colonization data, interpretation of root growth and 1987). In this study, no associations were observed environmental effects plant response can be limited. between root density, WFPS, and soil bulk density. Spores of VAMF were isolated from rotation plots ELLIS ET AL.: VAM FUNGAL ROOT COLONIZATION 793 to determine VAMF population. Although isolates were Table 6. Vesicular-arbuscular mycorrhizal fungi identified in not correlated with individual treatments, the agricul- rotation plots in Nebraska. tural soils have been found to have great diversity of Acaulospora scrobiculata Trappe species within populations and deserve to be studied Endogone sp. (orange-gold with a sinuous hyphal mantle) Endogone sp. (gold angular) (Table 6). Acaulospora, Endogone, Entrophospora, Entrophospora infrequens (Hall) Ames & Schneider Gigaspora, Glomus, Sclerocystis, and Scutellospora Gigaspora margarita Decker & Hall species were identified in these soils. The diversity of Glomus caledonium (Nicol. & Gerd.) Trappe. & Gerd. Glomus claroideum Schenck & Smith the VAMF in this soil is in sharp contrast to most Glomus clarum Nicol. & Schenck reports, where usually only four or five species were Glomus diaphanum Morton & Walk. found (Berch, 1989). The VAMF diversity in these Glomus etunicatum Becker & Gerd. soils could accunt for the high colonization rates and Glomus fasciculatum (Thaxter) Gerd. & Trappe emend. Walker & Koske Glomus intraradix Schenck & Smith productivity of the soils in this region. Glomus microcarpum Tul. & Tul. Differences between grain sorghum and soybean Glomus mosseae (Nicol. & Gerd.) Gerd. & Trappe could be related to differences in mycorrhizal depend- Glomus occultum Walker Glomus tenue (Greenall) Hall ency of the two crops. Unpublished research has shown Glomus sp. (small orange-red) that soybean, but not grain sorghum, had significant Glomus sp. (medium red-brown) reduction in plant growth and grain yield due to the Glomus sp. (large red-brown) Glomus sp. (like Glomus macrocarpum Tul. & Tul.) loss of VAMF after fumigation. In addition, plant spe- Sclerocystis rubiformis Gerd. & Trappe cies differ in mycorrhizal dependencies related to the Sclerocystis sinuosa Gerd. & Bakshi type of root system developed by the plants (Baylis, Scutellospora sp. Unknown (Glomus or Entrophospora—spore with subapical vesicle) 1975; St. John, 1980). With the positive correlation Unknown (sessile endospore) between root length and VAMF, some differences could be related to the root system and P demands of the two plants. In this experiment, manure treatments consistently reduced root length and VAMF coloni- zation. Green manuring with winter (Triticum mation in roots of soybean and sorghum under most aestivum L.) also reduces VAMF colonization (Bal- tillage treatments. However, manure treatments pro- truschat and Dehne, 1988); however, Harinikumar and duced the greatest yields and this suggests that manure Bagyaraj (1989) found that green manuring stimulated treatment may have decreased plant dependency of VAMF colonization and infective propagules in a soil VAMF and resulted in a decrease in VAMF root col- with a pH of 5.4 and an organic-matter content of onization. Crop rotation also affected VAMF root col- 0.5%. The addition of manure can buffer soil pH and onization and activity as grain sorghum roots had higher improve the physical and nutritional properties of poorer VAMF root-colonization rates than soybean roots and soils. For poorer soils, manure could stimulate VAMF root-colonization rates were greater in both colonization, number of infective propagules, and plant and grain sorghum after grain sorghum. The large yield. In soils where nutrients and VAMF propagules number of arbuscules associated with colonization in- are adequate, manuring could reduce VAMF coloni- dicate an active VAMF-plant association at sampling zation. Vesicular-arbuscular mycorrhiza appear to re- with all tillage and fertility treatments. spond to plant nutritional and environmental stress, In contrast to other studies, root development of and VAMF root colonization is decreased as plant both gain sorghum and soybeans showed no associa- stress is decreased. Manure amendment, to continuous tion with bulk density and WFPS, probably due to the soybean, grain sorghum-soybean rotation, continuous limited range of soil conditions in this study. Water- grain sorghum, and soybean-grain sorghum rotation filled pore space and VAMF colonization had a pos- treatments increased plant yield but decreased root itive association at some soil depths in soybean, but colonization. This experiment supports the concept that neither bulk density or WFPS appeared to be a sig- VAMF importance is increased as different types of nificant factor in VAMF colonization and activity. stresses are applied to the plant, whether they are nu- However, there was a negative correlation between tritional, water, or environmental. Thus, crops in these VAMF colonization and soil NO3 and P. Although highly fertile and productive soils are buffered against soil NO3 has been shown to affect rooting density, it adverse effects by the activity of VAMF. In VAMF- was not a significant factor in root development in this species-rich soils, such as this soil, the function and experiment. relationships between species and host response needs Root length and VAMF colonization were pos- to be examined to determine the role, function, and itively correlated at all depths. Root colonization was interaction of the different VAMF species. as high as 95% in this soil and there was a diverse population with a mixture of seven genera and 16 species identified from rotation plots and eight other specimens that have been described to the genus level SUMMARY AND CONCLUSION or have not been described in the literature. The di- Vesicular-arbuscular mycorrhizal fungal coloniza- versity of the VAMF in these soils may contribute to tion and total root length of both crops were reduced the high productivity of soils of this area by supplying when soybean was grown as the previous crop. Root a VAMF population capable of responding to varying colonization of soybean and sorghum by VAMF was environmental conditions. This study emphasizes the greatest when fertilizer was not applied to the soil. importance of below ground mycorrhizal-plant-soil When manure was applied as a fertilizer, there was a interactions in any attempt to explain crop-rotation reduction in VAMF colonization and arbuscule for- effects on crop yield. 794 SOIL SCI. SOC. AM. J., VOL. 56, MAY-JUNE 1992

ACKNOWLEDGMENTS We are grateful to D. Watson, research technologist at the University of Nebraska, for work on this project and Drs. J. Morton and S. Berch for their help in confirming some of the VAMF identifications.