Mycorrhiza (1994) 5:145-152

9 Springer-Verlag 1994

N. C. Talukdar 9 J. J. Germida

Growth and yield of lentil and wheat inoculated with three Glomus isolates from Saskatchewan soils

Abstract The vesicular-arbuscutar mycorrhizal fungi (VAMF) Glomus clarum (Nicol. and Schenck) isolate NT4, G. mosseae (Nicol. and Gerd.) Gerd. and Trappe isolate NT6 and G. versiforme (Karst.) Berch isolate NT7 coexist in wheat field soils in Saskatchewan. This study assessed the response of lentil (Lens esculenta L.) and wheat (Triticum aestivum L.) to monospecific and mixed cultures of these VAMF isolates. Seedlings were inoculated with 100 spores of a VAMF isolate, or an equal mixture of spores of two isolates, and grown in a sterile soil mix in a growth chamber. Both crops responded differently to these different VAMF isolates. In the case of lentil, G. clarum NT4 was more effective than G. mosseae NT6 and G. versiforme NT7, and significantly increased (P<0.05) the shoot dry weight (43%) and grain yield (57%) compared with the uninoculated control. There was a significant positive correlation between the percentage of VAMF colonized roots and shoot dry weight (r=0.672"**) and shoot phosphorus concentration (r=0.608"**) of lentil. In the case of wheat, G. clarum NT4 had no effect on shoot dry weight, but produced significant (P < 0.08) increases in grain yield (12%) and the phosphorus concentration of the shoot and grain. Although G. clarum NT4 and G. mosseae NT6 both produced similar levels of VAM colonization in wheat, the only response of wheat to isolate NT6 was an increase in plant height at harvest. The efficacy of G. clarum NT4 on both crops appeared to be related to its ability to produce more arbuscular colonization than G. mosseae NT6. Dual inoculation of seedlings with G. clarurn NT4 and G. mosseae NT6 resulted in competition between these two isolates. This was evident from a comparison of plant shoot dry weight and grain yield, and VAMF

N. C. Talukdar 9 J. J. Germida (~) Department of Soil Science, Universityof Saskatchewan, Saskatoon, Saskatchewan S7N 0W0, Canada Tel.: (306) 966-6836; Fax: (306) 966-6881; e-mail: [email protected]

spore production on the two crops inoculated either with isolate NT4 alone or in combination with NT6. G. mosseae NT6 reduced the efficacy of G. cIarum NT4 by 16% when dual inoculated on lentil, but had no effect when the host was wheat. Based on spore production, it was found that G. clarum NT4 was more competitive than G. mosseae NT6 when dual inoculated on lentil or wheat. Isolate NT4 produced ca. 2000 and 500 spores/ 100 g substrate, respectively, in the lentil and wheat pots, which was approximately 2-3 times more spores than those produced by isolate NT6 with either crop. When the plants were dual inoculated, there was a 1519% reduction in spore production by G. clarum NT4 and a 50-70% decrease in spore production by G. rnosseae NT6. Our results show that G. clarum NT4 was more competitive and effective in its ability to colonize and increase the growth and yield of lentil and wheat than G. mosseae NT6 or G. versiforrne NT7. The relative performance of isolate NT4 with different host plants suggests that this VAMF isolate exhibits a host preference for lentil. Key words Lentil 9 Wheat 9 Glomus spp. 9 Dual inoculation 9 Competition

Introduction Vesicular-arbuscular mycorrhizal fungi (VAMF) are an important group of soil fungi because they stimulate plant growth in agricultural soils with low levels of available P (Abbott and Robson 1981, 1984; Schubert and Hayman 1986). The many different VAMF species inhabiting agricultural soils (Hall 1977; Schenck and Smith 1981; Abbott and Robson 1984; An et al. 1990), have varying effects on crop growth. For example, VAMF inoculants differ in their ability to stimulate plant growth in sterile soils (Jensen 1982; Powell 1982; Haas and Krikun 1985; Medina et al. 1988; Louis and Lim 1988; Vierheilig and Ocampo 1991; Dhillion 1992) and in natural soils in the presence of indigenous

146

VAMF (Abbott and Robson 1981; Medina et al. 1988). Other VAMF species may depress host plant growth, either transiently or permanently depending upon the growth conditions (Bethlenfalvay et al. 1983; Koide 1985; Hung et al. 1990). Thus it is important to understand the relative contribution of a VAMF species, either alone or in the presence of other VAMF, to crop growth (Daft and Hogarth 1983; Koomen et al. 1987). Information on competition between VAMF to colonize host plants may be useful for practical exploitation of VAMF in crop production. The reasons why one VAMF species may be more effective than others in stimulating the growth of host plants are not well understood. The performance of VAMF symbiosis is influenced by combinations of host-endophyte-environment (Bethlenfalvay et al. 1982). In some cases, effective VAMF strains rapidly colonize roots (Sanders et al. 1977), exhibit a greater production and spread of extramatrical hyphae, and have a higher rate of P uptake (Graham et al. 1982; Jakobsen et al. 1992) than less effective strains. Increased uptake of relatively immobile mineral P by VAMF usually results in enhanced host-plant growth and yield (Bethlenfalvay et al. 1983). However, a VAMF strain that is effective under one set of growth conditions may not be effective under other conditions (Hayman and Tavares 1985; Medina et al. 1988), and thus competition between VAMF to colonize host plants will also vary (Bethlenfalvay et al. 1983; McGonigle and Fitter 1990; Sanders and Fitter 1992). Glomus species are widespread in soils of Saskatchewan, Canada (Talukdar and Germida 1993a), but their relative importance to crop growth is not known. In this study we assessed the effects of three isolates of G. clarum (NT4), G. mosseae (NT6) and G. versiforme (NT7), either alone or in combination, on the growth and yield of lentil and wheat in a growth chamber.

Materials and methods VAMF inoculum preparation Spores of G. clarum NT4 (INVAM No. SA101), G. rnosseae NT6 (INVAM No. SA103) and G. versiforrne NT7 (lab isolate) were isolated from wheat field soils and have been described previously (Talukdar and Germida 1993a, b). Isolates NT4 and NT6 were from Outlook (Bradwell association, sandy texture, 10 lag available P g - l ) , and isolate NT7 was from Semans (Weyburn association, loamy texture, 35 lag available P g - l ) in the Dark Brown soil zone of Saskatchewan (Talukdar and Germida 1993a). Monospecific cultures of each VAMF isolate were produced in an autoclaved 1:1 soil:sand mix using maize (Zea mays L. cv Early Golden Bantam) as the host crop in a growth chamber (Talukdar and Germida 1993b). At harvest, soil:root mixtures were air dried at 25~C for 1 week and stored at 7~C for 6 weeks. Spores of each isolate were obtained by wet sieving the monospecific VAMF cultures (Gerdemann and Nicolson 1963). Debris was removed by sucrose density gradient centrifugation (Daniels and Skipper 1982). After density gradient centrifugation, considerable amounts of hyphal fragments still contaminated the VAMF spore preparation. The entire VAMF spore mixture was vortexed for

30 s in a 15-ml centrifuge tube and spores were allowed to settle for 2-3 min before hyphal fragments were decanted off. This step was repeated three times; G. clarurn NT4 and G. versiforme NT7 spore preparations appeared to be free of hyphae, whereas those of G. mosseae NT6 still contained small amounts of hyphae.

Treatments for lentil Five treatments were assessed: (1) G. clarum NT4, (2) G. mosseae NT6, (3) G. clarum NT4 + G. mosseae NT6, (4) G. versiforme NT7 and (5) an uninoculated control. Lentil seeds were surface sterilized by soaking in 70% (v/v) alcohol for 1 rain, 1.2% (w/v) sodium hypochlorite for 5 min, and washed eight times with sterile distilled water. Seeds were aseptically germinated on 0.3% Trypticase soy agar (0.3% Trypticase soy broth, 1.5% agar). Two 5-day-old seedlings were transferred to pots and inoculated with 100 VAMF spores by a funnel technique (Menge and Timmer 1982). Dual-inoculated plants received 50 spores each of G. clarum NT4 and G. mosseae NT6. Uninoculated control plants received 5 ml of a water filtrate, obtained by passing a mixed suspension of NT4 and NT6 spores through a 1.2 lain cellulose nitrate membrane filter (Millipore Cat. no. RAWP 025 00, Millipore Corporation, Bedford, Mass.). Plants were grown in 15-cm plastic pots containing 2 kg of a soil:sand mix (hereafter called soil mix). Outlook soil was mixed with grade 7 silica sand in a 1:1 ratio. The soil mix was amended with 50 ml kg -1 of a modified Hoagland solution to provide the original nutrient levels of the Outlook soil (Talukdar and Germida 1993a) and autoclaved at 121~C for 1 h to eliminate indigenous VAMF. The autoclaved soil mix had a pH of 7.6 and contained (lag g - l ) NO3-N, 30; available P, 10; available K, 320; SO4-S, 15.6; Cu, 0.5; Fe, 7.5; Zn, 1.4; and Mn, 75.0. Chemical analyses were performed by the Saskatchewan Soil Testing Laboratory as follows: pH determined on a 1:1 soil water suspension; NO3-N and SO4-S determined colorimetrically on 0.001 M CaClz extracts; P and K on 0.5 M NaHCO3 extracts (Olsen et al. 1954) and micronutrients on diethylenetriaminepentaacetic acid extracts (Lindsay and Norwell 1978). Pots were placed in a growth chamber (photosynthetic irradiance 250-300 laE m - 2 s-1) with a 16-h light (26~C): 8-h dark (20~C) cycle. Pots were watered daily with sterile distilled water to maintain the soil mix at 70% moisture holding capacity (ca. - 6 0 kPa soil water potential). Each pot received 50 ml of the modified Hoagland (minus P) solution (Talukdar and Germida 1993a) at 3, 4 and 5 weeks after planting and 50 ml of 75 ppm NO3-N, at 6, 7 and 8 weeks after planting. Lentil plants (n =4) were harvested at 31, 56 and 100 days after planting (DAP). The comparative effects of the three VAMF species were assessed by determining the shoot and root dry weights (72 h at 54~C) and shoot P concentrations at each harvest and the P concentration and yield of grain at the final harvest. Senesced leaves were collected from each pot, oven dried (72 h at 54~C), weighed and included in the total shoot dry weight of lentil at the various sampling intervals. Shoots and seeds were digested in a H2SOg-H202 solution (Thomas et al. 1967) and total-P in the digests determined colorimetrically using an autoanalyzer (Technicon Industrial Systems 1973). To assess VAMF colonization, the whole root system of each replicate sample was recovered, cut into 1-cm pieces and mixed; a representative 3-g (fresh weight) subsample was cleared and stained in lactoglycerol trypan blue (Koske and Gemma 1989). Colonization (percent colonized root length) was measured with a compound microscope ( x 100) using the gridline intersect method (Giovannetti and Mosse 1980). The number of intersections with arbuscular colonization was also noted. Spores of G. clarum NT4, G. mosseae NT6 and G. versiforme NT7 were recovered from the entire 2 kg soil mix at the final harvest by wet sieving and density gradient centrifugation (see above). The resulting spore suspension was brought up to 100 ml with distilled water and the number of spores in two 5-ml samples was determined using a stereomicroscope ( x 47).

147 Table 1 Means of shoot and root dry weights, and yield of lentil plants (n =4) inoculated with G. clarum NT4, G. mosseae NT6 or G. versiforme NT7 spores and grown in a growth chamber. Means Inoculant

Grain yield

Days after planting 31

56

100

G. G. G. G.

clarum NT4 mosseae NT6 versiforme NT7 clarum NT4 + G. mosseae NT6

31

56

100

rag/pot

% change

Root dry wt. (mg/pot)

Shoot dry wt. (mg/pot) Control

with the same letter within a column are not statistically different as determined by the least significant difference test (P=0.05)

523 c 735 a 585 b c 533 c

7860 12305 7888 9799

d a d c

7495 d 10710a 8451 c 9098 b

424 a 429 a 402 a 370 a

3687 a 3004 b 3121 b 3262 b

3808 a 2214 c 3132 b 3074 b

2800 d 4401 a 3530 c 3630 c

-57 26 30

668 a b

11085 b

9167 b

426 a

2935 b

2386 c

3963 b

42

Treatments for wheat Four treatments were assessed: (1) G. clarum NT4, (2) G. mosseae NT6, (3) G. clarum NT4 + G. mosseae NT6, and (4) an uninoculated control. The experiment was set up as for lentil, except that wheat plants did not receive NO3-N and were harvested at 35, 48 and 83 DAP. In addition to the parameters recorded for lentil, effect of inoculation on plant height and number of tillers at 83 DAP was also assessed. Statistics Analysis of variance (ANOVA), least significant difference (LSD), and Pearson correlation coefficients (r) were calculated using PC-SAS (SAS 1990); linear regression was performed and coefficients of multiple determination (R 2) were calculated using CoPlot (CoHort 1991).

Results Effect of monospecific cultures of Glornus isolates on lentil Plant biomass and yield

The three V A M F inoculants increased shoot biomass of lentil c o m p a r e d with the uninoculated control (Table 1). T h e a p p e a r a n c e of a positive growth response varied depending upon the V A M F isolate present. F o r example, shoot biomass was increased by inoculation with G. clarum N T 4 after 31 days whereas increased shoot biomass was evident in the plants inoculated with G. versiforme NT7 and G. mosseae NT6 only after 56 and 100 D A P , respectively. Isolate N T 4 was the most effective V A M F inoculant for increasing shoot biomass of lentil. T h e inoculants had no effect on root biomass until 56 D A P , after which biomass was significantly reduced c o m p a r e d with the uninoculated control. T h e reduction of root biomass was greatest in lentil plants inoculated with G. clarurn NT4. All V A M F inoculants increased grain yield significantly, with G. clarum NT4 producing the greatest increase (57%), followed by G. versiforme NT7 (30%) and G. mosseae NT6 (26%) (Table 1).

V A M F colonization

The percentage of root length colonized by the three V A M F isolates was different at different sampling intervals (Fig. 1A). G. clarum NT4 achieved the highest root colonization at all harvests followed by G. versif o r m e NT7 and G. mosseae NT6. Isolates NT4 and NT6 also f o r m e d the greatest n u m b e r of arbuscules, and the levels of arbuscular colonization by individual V A M F isolates (Fig. 1B) followed a trend similar to that for total V A M F colonization (Fig. 1A). T h e r e was a significant linear relationship b e t w e e n lentil shoot dry weight and V A M F colonized root length at 31 D A P (y=0.314+0.013x; R 2=0.650.**) and 56 D A P (y = 0.036 + 0.176x; R 2 = 0.802***). F u r t h e r m o r e , at all harvest dates there was a significant correlation ( r = 0 . 6 7 2 " * * ) between shoot biomass and the percent V A M F - c o l o n i z e d roots. A similar relationship was also noted for shoot dry weight and the percent arbuscular colonization ( y = 7 . 6 2 0 + 0 . 1 3 8 x ; R 2 = 0 . 6 6 4 ***) at 56 D A P . Moreover, there was a significant correlation between grain yield and percent V A M F colonized roots ( r = 0 . 8 2 5 " * * ) and percent arbuscular colonization ( r = 0 . 7 8 3 " * * ) at 100 D A P .

P content o f shoots and grain G. clarum NT4 and G. versiforme NT7 increased the P concentration of lentil shoots at 56 D A P and of grain at 100 D A P relative to the uninoculated control. Inoculation with G. mosseae NT6, however, had no significant effect on the P concentration of lentil shoot but did result in an increase in the P concentration of lentil grain (Table 2). T h e r e was a significant correlation b e t w e e n P concentration of shoots and percent V A M F colonized roots ( r = 0 . 6 0 8 " * * for all dates) and percent arbuscular colonization ( r = 0 . 4 3 2 " * * for all dates).

148 Fig. 1 Effect of inoculation with G. clarum NT4 (A), G. mosseae NT6 (0), G. clarum NT4 + G. mosseae NT6 (0) and G. versiforme NT7 (r~) on the percentage of VAMF-colonized root length and arbuscular colonization in lentil (A, B) and wheat (C, D) roots at three harvests. Vertical lines represent least significant differences between treatment means (P = 0.05)

80 "A

[

80

C

o~

"~

60

60

40

40

20

20

i

g

0

i

50

50

40

40

30

30

20

20

,N O o

r

10

0

0 31 56 Days after planting

Table 2 Mean phosphorus contents of shoots and grain of lentil plants (n = 4) inoculated with G. clarum NT4, G. mosseae NT6 or G. versiforme NT7 spores and grown in a growth chamber. Means with the same letter within a column are not statistically different as determined by the least significant difference test (P=0.05). (DAP Days after planting) Inoculant

Phosphorus (mg g - l ) Shoot

Control G. G. G. G.

clarum NT4 mosseae NT6 versiforme NT7 clarum NT4 + G. mosseae NT6

Grain

31 DAP

56 DAP

2.76 2.93 2.68 2.81

2.67 3.47 2.90 3.04

a a a a

2.80 a

d a cd bc

3.27 a b

3.77 5.22 4.23 4.52

d a c bc

4.89 a b

100

0

35 48 Days after planting

83

Effect of m o n o s p e c i f i c cultures of G l o m u s isolates on wheat Plant growth, biomass and yield

Plant g r o w t h p a r a m e t e r s differed for w h e a t d e p e n d i n g on the V A M F inoculant (Table 3). F o r example, G. clar u m N T 4 increased the n u m b e r of tillers c o m p a r e d with the c o n t r o l and G. mosseae N T 6 - i n o c u l a t e d plants (Table 3). I n o c u l a t i o n with isolate NT6, h o w e v e r , caused a significant increase in plant height by 83 D A P . Isolate N T 4 stimulated early s h o o t g r o w t h of w h e a t at 48 D A P , w h e r e a s isolate N T 6 d e p r e s s e d shoot g r o w t h (Table 3). T h e s e effects w e r e transitory, h o w e v er, as there w e r e no significant difference in s h o o t dry weight at harvest (83 D A P ) . I n o c u l a t i o n with G. clarum N T 4 resulted in a significant increase in grain yield (12%), b u t inoculation with G. mosseae N T 6 h a d no significant effect on grain yield (Table 3). T h e r e was n o consistent effect of either V A M F inoculant o n r o o t dry weight. T h e N T 4 inoculant had no effect on r o o t

149 Table 3 Means of shoot and root dry weights, and yield of wheat plants (n = 4) inoculated with G. clarum NT4 or G. mosseae NT6

ter within a column are not statistically different as determined by the least significant difference test (P=0.05)

spores and grown in a growth chamber. Means with the same letInoculant

Days after planting 83

Control

Grain yield 35

48

83

35

48

83

mg/pot

% change

Tillers (numbers)

Height (cm)

Shoot dry weight (mg/pot)

4.00 b

63.5 b 63.6 b 71.3 a

4175 b 4205 b 3793 c

9053 b 9615 a 8355 c

8590 a 9053 a 9020 a

1675 b 1718 a b 1510 b

2627 a 2669 a 2374 a

2629 a 2101 c 2544 a b

5023 b 5638 a 4816 b

-12 -4

59.8 b

4545 a

9730 a

8938 a

1968 a

2577 a

2327 b c

5679 a

13

G. clarum NT4 5.74 a G. mosseae NT6 4.25 b G. clarum NT4 + G. mosseae NT6 6.00 a

Root dry weight (mg/pot)

g r o w t h at 35 and 48 D A P (Table 3) a l t h o u g h r o o t weight was significantly r e d u c e d at final harvest (83 DAP).

Table 4 Mean phosphorus contents of shoots and grain of wheat plants (n =4) inoculated with G. clarum NT4 or G. mosseae NT6

V A M F colonization

Inoculant

T h e p e r c e n t a g e of r o o t length colonized b y G. c l a r u m N T 4 and G. m o s s e a e N T 6 was n o t statistically different ( P < 0.05) at 35 and 48 D A P , a l t h o u g h G. c l a r u m N T 4 i n o c u l a t e d plants exhibited significantly g r e a t e r V A M colonization levels at 83 D A P (Fig. 1C). F u r t h e r m o r e , G. c l a r u m N T 4 - c o l o n i z e d roots c o n t a i n e d m o r e arbuscules t h a n G. m o s s e a e N T 6 - c o l o n i z e d r o o t s at all harvests (Fig. 1D). S h o o t biomass was not c o r r e l a t e d with % V A M F colonization. T h e r e was, h o w e v e r , a significant linear relationship b e t w e e n s h o o t biomass and the % arbuscular colonization at b o t h 35 D A P (y=3.618+0.079x; R 2=0.556.*) and 48 DAP ( y = 8 . 0 8 9 + 0 . 1 0 8 x ; R 2 = 0 . 8 2 0 " * * ) . G r a i n yield on the o t h e r h a n d was c o r r e l a t e d significantly with b o t h % V A M colonization (r = 0.645**) and % arbuscular colonization (r = 0.873***).

P c o n t en t o f shoots a n d grain

T h e s h o o t P c o n c e n t r a t i o n of G. c l a r u m N T 4 - i n o c u lated plants was significantly higher t h a n that of the controls at b o t h 35 and 48 D A P . Likewise, inoculation with N T 4 increased the P c o n c e n t r a t i o n o f the grain. Conversely, G. rnosseae N T 6 increased the P c o n c e n t r a tion of shoots only at 35 D A P (Table 4). T h e r e was a significant c o r r e l a t i o n b e t w e e n P c o n c e n t r a t i o n of shoots and p e r c e n t V A M F colonized roots ( r = 0 . 2 8 9 " for all dates) and p e r c e n t arbuscular colonization at 48 D A P ( r = 0 . 8 2 4 " * * ) and 83 D A P ( r = 0 . 7 7 5 " * * ) .

E f f e c t of dual inoculation with G. c l a r u m N T 4 and G. m o s s e a e N T 6 o n lentil and w h e a t T h e s h o o t b i o m a s s and grain yield of d u a l - i n o c u l a t e d lentil was significantly l o w e r t h a n that of G. c l a r u m

spores and grown in a growth chamber. Means with the same letter within a column are not statistically different as determined by the least significant difference test (P=0.05) Phosphorus (mg g - l ) Shoot

Control G. clarum NT4 G. mosseae NT6 G. clarum NT4 + G. mosseae NT6

Grain

35 DAP

48 DAP

2.56 b 2.94 a 2.87 a

1.61 c 2.06 a b 1.82 b c

3.38 c 3.74 b 3.41 c

2.71 a b

2.18 a

4.03 a

Table 5 Mean number of vesicular-arbuscular mycorrhizal fungi

spores recovered from the soil mix used to grow lentil and wheat (n = 4) in a growth chamber. Means with the same letter within a column are not statistically different as determined by the least significant difference test (P=0.05). (ND Not determined) Inoculant

G. ctarum NT4 G. mosseae NT6 G. clarum NT4 + G. mosseae NT6 G. versiforme NT7

Spores/100 g soil mix Lentil

Wheat

2068 a 945 b 1680 a 450 b 1918 a

577 a 215 b 487 a 75 b ND

N T 4 - i n o c u l a t e d plants but was significantly g r e a t e r t h a n that of the G. m o s s e a e N T 6 - i n o c u l a t e d plants (Table 1). P a t t e r n s of V A M F colonization in dual-inoculated lentil roots w e r e similar to those in G. c l a r u m N T 4 - i n o c u l a t e d roots (Figs. 1 A and 1C). A l t h o u g h w h e a t plants i n o c u l a t e d with a mixture o f N T 4 and N T 6 t e n d e d to g r o w b e t t e r and p r o d u c e m o r e total biomass t h a n plants inoculated with individual isolates, differences w e r e statistically significant only at 35 D A P (Table 3), w h e n the s h o o t weights of the dual-inocu-

150 lated plants were significantly greater than those of other treatments. In general, patterns of VAMF colonization of wheat roots inoculated with NT4 and NT6 in combination were similar to that of plants inoculated with NT4 alone. Relative differences in the numbers of spores produced in the lentil and wheat pots by the NT4 and NT6 isolates, inoculated either alone or in combination, indicated competition between the two isolates (Table 5). For example, whereas spore production by G. mosseae NT6 was 46% and 37% of that of G. clarum NT4 when the two VAMF strains were inoculated as monospecific cultures on lentil and wheat, respectively, it was about one-half these levels when the two isolates were dual inoculated. Furthermore, the increase in spore production by G. clarum NT4 was greater (i.e., 9- to 33-fold) when this isolate was dual inoculated with G. mosseae NT6 than when it was inoculated as a monospecific culture to either crop (i.e., 5- to 20-fold increase). In contrast, differences in spore production by G. mosseae NT6, inoculated as a monospecific culture or as a mixture with G. clarum NT4, were generally smaller, although not significantly so.

Discussion Our results indicate that G. clarum NT4, G. mosseae NT6 and G. versiforme NT7 isolated from Saskatchewan soils exhibit a degree of specificity in terms of their relative benefits to lentil and wheat plants. These V A M F are widespread in Saskatchewan soils and coexist in the wheat rhizosphere (Talukdar and Germida 1993a). However, when isolated from soil and cultured as monospecific cultures, these isolates differed in their ability to colonize and affect the growth and yield of lentil and wheat grown in a sterile soil mix. G. clarum NT4 was the most effective of the V A M F isolates studied and consistently increased plant growth, yield and P uptake. Both crops varied in the type and magnitude of their response to inoculation with these V A M F isolates. Lentil responded better to V A M F inoculants than wheat, demonstrating a greater dependence on these V A M F for P uptake during early plant growth, which resulted in enhanced shoot growth and increased yield. Differences in the ability of VAMF to stimulate plant growth have been reported previously by Mosse (1972) for Paspalum notatum, Carling and Brown (1980) for soybean (Glycine max L.), Koomen et al. (1987) for white clover (Trifolium repens L.) and strawberry (Fragaria vesca L.), and by Hung et al. (1990) for sweet potato [Ipomoea batatas (L.) Lam]. Furthermore, studies by Louis and Lim (1988) with soybean and by Dhillion (1992) with rice (Oryza sativa L.) demonstrate the importance of evaluating potential inoculant strains for endophyte specificity with plant species. Our results support these observations that some V A M F isolates that effectively stimulate growth of one host may not be

as effective for other hosts under similar conditions. These differences might be related to soil environment factors or, alternatively, to the genotype of the host plant or the VAMF. For example, the G. mosseae NT6 isolate used in our study was not as effective as the G. clarum NT4 isolate in colonizing wheat roots, producing arbuscules and increasing plant growth. In contrast, Vierheilig and Ocampo (1991) found that a G. mosseae isolate was more effective than G. fasciculatum or G. aggregatum in stimulating the growth of several wheat cultivars, although there was no difference in the percentage of V A M F colonization caused by the three VAMF. G. clarum NT4 enhanced the growth and yield of lentil more than G. versiforme NT7 and G. mosseae NT6. The relative effectiveness of these isolates (i.e., N T 4 > N T 7 > NT6) on lentil was correlated to the extent of their colonization. Other workers (Sanders et al. 1977; Abbott and Robson 1981; Medina et al. 1988) have reported similar results and reached the same conclusion. It should be noted, however, that isolate NT7 was originally isolated from a soil containing three times more plant-available P level (i.e, 35 p~g g-1) than isolates NT4 and NT6. Such an environment might have selected for VAMF strains inefficient in P uptake. Our studies were conducted in a soil mix with a low level of available P (i.e., 10 txg g - l ) and the relative performance of these isolates in soils with higher levels of available P is unknown. Abbott and Robson (1978), Hayman and Tavares (1985) and Hung et al. (1990) suggested that early colonization of plant roots by V A M F is important if plants are to benefit from this relationship. We also found a significant correlation between the percentage of V A M F colonized root length and arbuscule production with shoot dry weight (at all harvest dates) and grain yield of lentil at the final harvest. This suggests that early V A M F colonization of lentil roots and production of arbuscules are important for good plant growth and yield. Furthermore, the highly significant correlations between shoot P concentration at all harvest dates with the percentage of V A M F colonized root length and arbuscule production indicates that V A M F were essential to provide an adequate supply of P to lentil plants. In contrast, there was no relationship between wheat growth and the percentage of VAMF colonization, and only wheat grain yield was significantly correlated with the percentage of V A M F and arbuscular colonization levels. The P concentration of lentil shoot and grain increased in response to inoculation with the VAMF isolates, and P concentration followed a trend similar to VAMF colonization of the roots, i.e., N T 4 > N T T ~ N T 6 . Hung et al. (1990) also observed a positive correlation between the P content of sweet potato shoots and colonized root length for three VAMF isolates. Such relationships are expected because they reflect both the importance of VAMF to P uptake by plants and the effect of an adequate P supply on biomass production. The difference in the relative efficacy

151 of G. clarurn NT4 and G. rnosseae NT6 might be explained by functional differences at the level of the host-fungus interface. G. clarum NT4 produced significantly higher levels of arbuscular colonization in wheat and lentil roots compared to G. rnosseae NT6. McGonigle et al. (1990) noted that arbuscules are the principle V A M fungal structures that function in nutrient transfer between the host and fungus, and predicted that arbuscule development by a V A M F may reflect its relative benefit to the host crop. Our results support this hypothesis. The presence of more than one V A M fungus in the host rhizosphere may lead to competition between endophytes (Daft and Hogarth 1983; Wilson and Trinick 1983; Wilson 1984), and may even lead to multiple infections. Evidence for competition between specific V A M F pairs is usually obtained by comparing the growth of host plants inoculated either with one V A M F species alone or in combination with another (Bethlenfalvay et al. 1982). We found that the effects of dual inoculation of lentil and wheat with G. clarum NT4 and G. rnosseae NT6 were generally negative with respect to the relative efficacy of the most beneficial V A M F fungus, isolate NT4. In other words, the more beneficial the isolate NT4 was for a host, the greater the impact of isolate NT6. For example, inoculation of lentil with G. clarum NT4 alone produced a 57% increase in yield, whereas dual inoculation of this isolate with G. mosseae NT6 only produced a 42% increase in the yield (a 36% decrease). In contrast, NT4 increased the yield of wheat by only 12% and dual inoculation with NT6 had no impact on this increase in yield. Similarly, Koomen et al. (1987) found that competition between some V A M F in mixed inocula reduced the plant growth response that individual components of the mixture were capable of producing. Precise methods for evaluating competition between V A M F species should be based on the relative proportion of colonized root area occupied by each individual V A M fungus when coinoculated to a host plant (McGonigle and Fitter 1990; Sanders and Fitter 1992). However, methods to distinguish anatomical structures produced by different V A M F species inside roots are relatively scarce (Abbott and Robson 1978; Wilson 1984) and it is difficult to draw firm conclusions about competition between VAMF. Competition between V A M F has been assessed using hyphae diameter (McGonigle and Fitter 1990), hyphae length and biomass (Bethlenfalvay et al. 1983) or sporulation (Koomen et al. 1987). Daft and Hogarth (1983) used V A M F spore production, when inoculated alone or in combination with other V A M F , as a measure of competition. They found that spore production by G. clarum in a maize host was greater than that of G. caledonium, G. geosporum or G. rnosseae. However, spore production by each individual V A M F was greatly reduced when coinoculated with other VAMF. Since spore production by G. clarum was affected least by coinoculation, these authors suggested that this V A M fungus was

more competitive. Similarly, our results suggest that G. clarum NT4 was more competitive than G. mosseae NT6 when dual inoculated to lentil or wheat. This was evident from the similarity in V A M F colonization patterns of roots inoculated with a mixture of isolates to those of plants inoculated with NT4 alone, and the relative effects of individual isolates and a mixture on plant biomass and grain yield, and also on spore production. For example, in the case of spore production, inoculation with a monospecific culture of G. clarum NT4 produced about 2-3 times more spores (per 100 g substrate) than were produced when the inoculant was G. mosseae NT6. However, when these isolates were dual inoculated to lentil or wheat, isolate NT4 produced about 4-7 times more spores than did NT6. Our study shows that three V A M F isolates of G. clarum NT4, G. rnosseae NT6 and G. versiforme NT7 which coexist in Saskatchewan field soils differ in their competitiveness and contribution to the growth and yield of lentil and wheat. G. clarurn NT4 was a very competitive and a highly effective V A M fungus for lentil and wheat grown in a sterile soil mix in the growth chamber. The effectiveness of NT4 on lentil and wheat growth in unsterile Saskatchewan soils containing indigenous V A M F is currently being evaluated. Acknowledgements This work was supported in part by grants from the Saskatchewan Agriculture Development Fund and the Natural Science and Engineering Research Council of Canada. N. C. T. was supported by the Commonwealth Scholarship and Fellowship Plan, Canadian Awards, Government of Canada. We thank R. E. Farrell for advice and consultation on statistical analyses. Contribution no. R 723, Saskatchewan Institute of Pedology.

References Abbott LK, Robson AD (1978) Growth of subterranean clover in relation to the formation of endomycorrhizas by introduced and indigenous fungi in a field soil. New Phytol 81:575-585 Abbott LK, Robson AD (1981) Infectivity and effectiveness of five endomycorrhizal fungi: competition with indigenous fungi in field soils. Aust J Agric Res 32:621~530 Abbott LK, Robson AD (1984) Colonization of the root system of subterranean clover by three species of vesicular-arbuscular mycorrhizal fungi. New Phytol 96:275-281 An Z-Q, Grove JH, Hendrix JW, Hershman DE, Henson GT (1990) Vertical distribution of endogonaceous mycorrhizal fungi associated with soybean, as affected by soil fumigation. Soil Biol Biochem 22:715-719 Bethlenfalvay GJ, Ulrich JM, Brown MS (1982) Plant response to mycorrhizal fungi: host, endophyte, and soil effects. Soil Sci Soc Am J 49:1164-1168 Bethlenfalvay GJ, Bayne HG, Pacovsky RS (1983) Parasitic and mutualistic associations between a mycorrhizal fungus and soybean: the effect of phosphorus on host-plant-endophyte interactions. Plant Physiol 57:543-548 Carling DE, Brown MF (1980) Relative effect of vesicular-arbuscular mycorrhizal fungi on the growth and yield of soybeans. Soil Sci Soc Am J 44:528-532 CoHort Software (1991) CoPlot: scientific graphics software, version 2.02. Cottort Software, Berkeley, Calif Daft MJ, Hogarth BG (1983) Competitive interactions amongst four species of Glomus on maize and onion. Trans Br Mycol Soc 80: 339-345

152 Daniels BA, Skipper HD (1982) Methods for the recovery and quantitative estimation of propagules from soil. In: Schenck NC (ed) Principles and methods of mycorrhizal research. American Phytopathological Society, St Paul, Minn, pp 2936 Dhillon SS (1992) Host-endophyte specificity of vesicular-arbuscular mycorrhizal colonization of Oryza sativa L. at the pretransplant stage in low or high phosphorus soil. Soil Biol Biochem 24:405-411 Gerdemann JW, Nicolson TH (1963) Spores of mycorrhizal Endogone species extracted from soil by wet sieving and decanting. Trans Br Mycol Soc 46:235-244 Giovannetti M, Mosse B (1980) An evaluation of techniques for measuring vesicular arbuscular mycorrhizal infection in roots. New Phytol 84: 489-500 Graham JH, Linderman RG, Menge JA (1982) Development of external hyphae by different isolates of mycorrhizal Glomus spp. in relation to root colonization and growth of troyer citrange. New Phytol 91:183-189 Haas JH, Krikun J (1985) Efficacy of endomycorrhizal-fungus isolates and inoculum quantities required for growth response. New Phytol 100: 613-621 Hall IR (1977) Species and mycorrhizal infections of New Zealand Endogonaceae. Trans Br Mycol Soc 68:341-356 Hayman DS, Tavares M (1985) Plant growth responses to vesicular-arbuscular mycorrhiza. XV. Influence of soil pH on the symbiotic efficiency of different endophytes. New Phytol 100: 367-377 Hung LL, Sylvia DM, O'Keefe DM (1990) Isolate selection and phosphorus interaction of vesicular-arbuscular mycorrhizal fungi in biomass crops. Soil Sci Soc Am J 54:762-768 Jakobsen I, Abbott LK, Robson AD (1992) External hyphae of vesicular-arbuscular mycorrhizal fungi associated with Trifolium subterraneurn L. I. Spread of hyphae and phosphorus inflow into roots. New Phytol 120:371-380 Jensen A (1982) Influence of four vesicular-arbuscular mycorrhizal fungi on nutrient uptake and growth in barley (Hordeum vulgate). New Phytol 90:45-50 Koide R (1985) The nature of growth depressions in sunflower caused by vesicular-arbuscular mycorrhizal infection. New Phytol 99: 449--462 Koomen I, Grace C, Hayman DS (1987) Effectiveness of single and multiple mycorrhizal inocula on growth of clover and strawberry plants at two soil pHs. Soil Biol Biochem 19:539544 Koske RE, Gemma JN (1989) A modified procedure for staining roots to detect VA mycorrhizas. Mycol Res 92:486-488 Lindsay WL, Norwell W A (1978) Development of a DTPA soil test for iron, manganese, copper, and zinc. Soil Sci Soc Am J 42: 421.428 Louis I, Lira G (1988) Differential response in growth and mycorrhizal colonization of soybean to inoculation with two isolates of Glomus clarum in soils of different P availability. Plant Soil 112: 37-43 McGonigle TP, Fitter A H (1990) Ecological specificity of vesicular-arbuscular mycorrhizal associations. Mycol Res 94:120122

McGonigle TP, Miller MH, Evans DG, Fairchild GL, Swan JA (1990) A new method which gives an objective measure of colonization of roots by vesicular-arbuscular mycorrhizal fungi. New Phytol 115:495-501 Medina OA, Sylvia DM, Kretschmer A E Jr (1988) Response of Siratro to vesicular-arbuscular mycorrhizal fungi. II. Efficacy of selected vesicular-arbuscular fungi at different phosphorus levels. Soil Sci Soc Am J 52:420423 Menge JA, Timmer LW (!982) Procedures for inoculation of plants with vesicular-arbuscular mycorrhizae in the laboratory, greenhouse and field. In: Schenck NC (ed) Principles and methods of mycorrhizal research. American Phytopathological Society, St Paul, Minn, pp 56-59 Mosse B (1972) Effect of different Endogone strains on growth of Paspalum notatum. Nature 239:221-223 Olsen SR, Cole CV, Watanabe FS, Dean LA (1954) Estimation of available phosphorus in soils by extraction with sodium bicarbonate. US Department of Agriculture, Circulation 939 Powell CL (1982) Selection of efficient VA mycorrhizal fungi. Plant Soil 68: 3-9 Sanders IR, Fitter A H (1992) The ecology and functioning of vesicular-arbuscular mycorrhizas in co-existing grassland species. I. Seasonal patterns of mycorrhizal occurrence and morphology. New Phytol 122:517-524 Sanders FE, Tinker PB, Black RLB, Palmerley SM (1977) The development of endomycorrhizal root systems. I. Spread of infection and growth-promoting effects with four species of vesicular-arbuscular endophytes. New Phytol 78:257-268 SAS Institute Inc (1990) SAS procedures guide, version 6, 3rd edn. Cary, NC Schenck NC, Smith GS (1981) Distribution and occurrence of vesicular-arbuscular mycorrhizal fungi on Florida agricultural crops. Soil Crop Sci Soc Fla Proc 40:171-175 Schubert A, Hayman DS (1986) Plant growth responses to vesicular-arbuscular mycorrhiza. XVI. Effectiveness of different endophytes at different levels of soil phosphate. New Phytol 103: 79-90 Talukdar NC, Germida JJ (1993a) Occurrence and isolation of vesicular-arbuscular mycorrhizae in cropped field soils of Saskatchewan, Canada. Can J Microbiol 39:567-575 Talukdar NC, Germida JJ (1993b) Propagation and storage of vesicular-arbuscular mycorrhizal fungi isolated from Saskatchewan agricultural soils. Can J Bot 71:1328-1335 Technicon Industrial Systems (1973) Orthophosphate in water and wastewater. (Industrial method no 94-70W) Technicon Industrial Systems, Tarrytown, NY Thomas RL, Sheard RW, Moyer JR (1967) Comparison of conventional and automated procedures for nitrogen, phosphorus, and potassium analysis of plant material using a single digestion. Agron J 59:240-243 Vierheilig H, Ocampo JA (1991) Susceptibility and effectiveness of vesicular-arbuscular mycorrhizae in wheat cultivars under different growing conditions. Biol Fertil Soils 11:290-294 Wilson JM (1984) Competition for infection between vesiculararbuscular mycorrhizal fungi. New Phytol 97:427-435 Wilson JM, Trinick MJ (1983) Infection development and interactions between vesicular-arbuscular mycorrhizal fungi. New Phytol 93: 543-553