Illlllllllllllll Hllllllll __ 3,r13_93_10714-217 "MLEJL ‘5 This is to certify that the thesis entitled Sane Aspects of the Ecology of (hion Myoorrhizae presented by Nicholas C. Bolgiano has been accepted towards fulfillment of the requirements for M.S. degree in Botany & Plant Pathology Major proLssor www— 0-7639 MS U is an Afiimmtive Action/Equal Opportunity Institution ~ MSU RETURNING MATERIALS: . Place in book drop to LIBRARIES remove this checkout from —_-_. your record. FINES will be charged if book is returned after the date stamped below. .a-u— ‘2') l W SOME ASPECTS OF THE ECOLOGY OF ONION MYCORRHIZAE by Nicholas C. Bolgiano A THESIS submitted to Michigan State University in partial fulfillment of the requirement for the degree of MASTER OF SCIENCE Department of Botany and Plant Pathology 1982 ABSTRACT SOME ASPECTS OF THE ECOLOGY OF ONION MYCORRHI ZAE BY Nicholas C. Bolgiano High levels of infection of onion roots by vesicular- arbuscular mycorrhizal fungi was only found in commercially grown onions when soil phosphorus (P) concentrations were below 30 pg/cma. Experimentally, added P decreased mycor- rhizal infection in the field. Soil P concentrations in the field experiments above 20 pg/cm3 were correlated with low mycorrhizal infection. Onions given adequate P for maximal growth exhibited high mycorrhizal infection only after mid- season. In pot experiments, onions grown in soils of low P were stimulated by mycorrhizae. Addition of P appeared to decrease mycorrhizal benefits in some soils. Differences in infection levels were found among 17 onion cultivars studied in the field. Application of irrigation to field-grown on- ions decreased the amount of soil P necessary to inhibit mycorrhizal infection. ACKNOWLEDGEMENTS I would like to thank Dr. Gene Safir for the privilege of working in his laboratory, and for the support he has provided me. I appreciate the efforts of Dr. Dean Haynes and Dr. Melvin Lacy for serving on my committee. Without the efforts of my fellow graduate students, this thesis would have not been possible. I would like to partic- ularly thank Emmett Lampert, Gary Whitfield, Frank Drummond and Ellie Groden of the Entomology Department, and Charles Nelsen, Scott Eisensmith, Carol Ishimaru and Maureen Mulligan of the Botany & Plant Pathology Department. Paige Sommers and Eric Hall provided valuable technical help, and Donna Maas and Nancy Bolgiano served as editors in this thesis preparation. ii TABLE OF CONTENTS Page LIST OF TABLES v LIST OF FIGURES vii Introduction and Literature Review 1 Chapter I. Mycorrhizal Infection in Commercially 44 Grown Onions Materials and Methods 44 Results 52 Discussion 66 Chapter II. The Effect of Four Phosphorus Levels, 70 With and Without Glomus etunicatus, in Two Fields ' Materials and Methods 71 Results 75 Discussion 88 Chapter III. The Potential of Mycorrhizal Substitu- 93 tion for Fertilizer P Materials and Methods 93 Results 98 Discussion 107 Chapter IV. Mycorrhizal Infection in 1? Onion 110 Cultivars Materials and Methods 110 Results 112 Discussion 112 iii Chapter V. Chapter VI. Bibliography Appendix Appendix Appendix Appendix Appendix Appendix Appendix 1. Response of Mycorrhizae to P Under Two Watering Regimes in the Field Materials and Methods Results Discussion Onion Plant Model Data Collection - 1980 Data Collection - 1981 Measurement of Leaf Surface Area Root Lengths The Effect of Plant Density on Yield Plant Emergence Plant Maturity Relative Growth as Related to Temper- ature Calculated Growth Rates Simulation Program Field locations of 1979 commercial field survey Procedure for Root-Staining and Spore Isolation Commercial onion fields sampled in 1980 Bray P1 Soil Phospohrus Extraction Pro- cedure Commercial onion fields sampled in 1981 Computer files containing onion data Onion simulation program iv Page 115 115 116 120 123 123 125 131 141 144 153 156 158 159 163 167 183 184 185 186 188 189 190 LIST OF TABLES Table 1. 10. 11. 12. 13. 14. Mycorrhizal ratings, spore counts and soil P concentrations from 1979 commercial field survey Mycorrhizal ratings of onions and soil P concentrations in autoclaved soil with inocu- lum and in nonsterile soil Spore counts in soils from 1980 commercial field survey Spore concentrations of soil extracted at distances from the onion row Soil P concentration from fields to which inoculum was added Soil P concentrations, mycorrhizal ratings and spore counts from 1981 commercial field survey Analysis of variance of the mycorrhizal ratings of the 1981 Muck Farm low P field experiment Mycorrhizal ratings of the 1981 Muck Farm low P field experiment with time Mycorrhizal ratings and plant weights from 1981 Muck Farm low P field experiment Mycorrhizal ratings and bulb fresh weights of 1981 Muck Farm high P field experiment Analyses of variance for the experiment exam- ining mycorrhizal benefits in two soils - I Plant dry weights, mycorrhizal ratings and soil P concentrations of the experiment exam- ining mycorrhizal benefits in two soils - I Analyses of variance for the experiment exam- ining mycorrhizal benefits in two soils - 11 Plant dry weights, mycorrhizal ratings and soil P concentrations of the experiment exam- ining mycorrhizal benefits in two soils - II Page 53 55 60 61 61 64 80 80 81 86 99 100 102 103 Table 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. Plant dry weights, mycorrhizal ratings and soil P concentrations of the mycorrhizal experiment conducted in marl soil Plant dry weights of the mycorrhizal exper- iment conducted in sandy muck soil Plant dry weight of the mycorrhizal exper- iment conducted in Houghton muck soil Mycorrhizal ratings of onion cultivars from 1981 Muck Farm experiment Analyses of variance of the 1981 irrigation-P experiment Mycorrhizal ratings and yields from Muck Farm irrigation-P experiment Regressions predicting bulb and leaf weights as a percentage of top and bulb weight Surface area regressions Regressions describing the green leaf surface area to total leaf surface area relationship Regressions describing the leaf surface area to leaf weight relationship Root length regressions Regression equations describing yield-density data Emergence data and regressions of two emer- gence trials 1980 and 1981 onion data Plant simulation results vi Page 104 105 106 111 117 118 127 134 136 141 143 148 154 161 164 LIST OF FIGURES Figure 1. 2. 10. 11. 12. 13. Patterns of plant growth of mycorrhizal and non- mycorrhizal plants versus soil P concentrations Patterns of plant growth of mycorrhizal and non- mycorrhizal plants versus soil P concencentrations under well-watered and water-stressed conditions Mycorrhizal rating of onions versus measured soil P concentrations from 7 fields sampled in the 1979 commercial field survey Mycorrhizal rating of onions sampled from 6 fields versus time during 1980 Mycorrhizal rating of onions versus measured soil P concentration from 1980 commercial field survey Mycorrhizal rating of onions versus measured soil P concentration from 1981 commercial field survey Mycorrhizal rating of onions versus measured soil soil P concentration from 1980 Muck Farm experiment Mycorrhizal rating of onion roots and bulb yield versus total P from the 1980 Muck Farm low P field experiment Mycorrhizal spore numbers versus mycorrhizal rating of treatments of the 1980 Muck Farm experiments Mycorrhizal rating of onions versus measured soil P concentration of the 1981 Muck Farm low P field experiment Mycorrhizal ratings of onions versus measured soil P concentration of the 1981 Muck Farm high P field experiment Mycorrhizal rating of onions versus measured soil P concentration from the irrigation-P experiment ln(bulb dry weight/total dry weight) versus total dry weight vii Page 27 34 54 57 58 63 76 78 84 87 119 128 Figure 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. ln(1eaf dry weight/total dry weight) versus total dry weight ln(bulb weight) versus total dry weight pre- dicted from regression and observed ln(1eaf weight) versus total dry weight pre- dicted from regression and observed ln(bulb dry weight) versus total dry weight observed in 1980 and predicted from regression ln(1eaf dry weight) versus total dry weight observed in 1980 and predicted from regression Green leaf surface area versus total leaf surface area ln(total leaf surface area/leaf dry weight) versus leaf dry weight ln(total leaf surface area) versus leaf dry weight ln(total leaf surface area) versus leaf dry weight observed in 1980 and predicted from regression Green leaf surface area versus total leaf surface area observed in 1980 and predicted from regression The reciprocal of the dry weight per plant versus density observed in 1981 Total plant dry weight per m2 predicted from regression and observed Percent onion emergence versus degree days after planting for two trials Relative growth rate function of onions as re- lated to temperature A. ln(plant dry weight) versus degree units after planting B. Calculated growth rates versus degree units from 1980 and 1981 onion data viii Page 128 129 130 132 132 138 139 139 140 140 152 152 155 160 160 Introduction and Literature Review 1, in particularly vesicular-arbuscular mycor- Mycorrhizae rhizae (VAM), hold great promise for improving growth of agricultural crops. Though these fungi have been extensive- ly studied in recent years, knowledge of the role of VAM in crop productivity is lacking. The goals of this research included determining the benefits of onion (Allium cepa L.) VAM in the field, as well as under controlled greenhouse and growth chamber conditions. By ascertaining which factors affect VAM formation, the potential of mycorrhizal benefits might be deduced. With the application of these results, symbiotic microorganisms could help man's efforts to grow his crops with higher economical and ecological efficiency. Most crop plants form VAM, except for members of the families Cruciferae and Chenopodiaceae. VAM fungi are members of the family Endogonaceae. Included are eight Acaulospora species, seven Gigaspora species, thirty-eight Glomus species and seven species of Sclerocystis (Miller, 1981). 1 - 'Mycorrhizae' refers to an association between fungi and roots of higher plants. Mycorrhizal fungi penetrate and ramify throughout host roots, and their hyphae branch out into the soil well beyond zones in which nutrient uptake by roots remains limited. Mycorrhizal hyphae absorb and transport nutrients, thus increasing the soil volume from which plants absorb those nutrients (Rhodes and Gerdemann, 1980 and Tinker, l975ab)2. Mycorrhizal roots appear to absorb nutrients from the Same labile pools as do nonmycorrhizal roots (Barrow et al, 1977, Hayman and Mosse, 1972, Mosse et a1, 1973, Pairunan et a1, 1980, Powell, 1975b, Sanders and Tinker, 1973 and Tinker, 1975ab). No insoluble sources of nutrients have been found which mycorrhizae can exclusively tap. Nutrient transfer occurs primarily across hyphal and arbuscular walls within host roots. In return, mycorrhizal fungi absorb carbon from host cells (Bevege et a1, 1975). Mycorrhizae can increase uptake of highly immobile ele- ments that cannot readily diffuse to roots. Phosphorus (P), is probably the most important element involved in mycorrhi- zal uptake, since it is a highly immobile nutrient needed in large amounts by host plants. The uptake of copper (Cu) and zinc (Zn), two highly immobile micronutrients, is also in- creased by mycorrhizae (Rhodes and Gerdemann, 1980). Since diffusion of these nutrients through soil is slow, most 2 - See these two articles, plus that of Safir and Nelsen (1981) for reviews on nutrient uptake by mycorrhizae. plants deplete their resources within 1-2 mm of root sur- faces (Owusu-Bennoah and Wild, 1979). Thus, mycorrhizae provide an advantage to plants by enabling nutrient absorp- tion beyond the root depletion zone. The uptake of more mobile elements can be increased if soil concentrations of these nutrients is extremely low. Sulfur (S) uptake can be increased by mycorrhizae in S deficient soil (Rhodes and Gerdemann, 1980). Likewise, potassium (K) uptake by mycorrhizae may be of benefit in extremely K deficient soil (Powell, 1975a). Mycorrhizae can improve nitrogen (N) nutrition of nodulating plants through beneficial effects of P on nitrogenous activity (Asimi et al, 1980, Carling et al, 1978, and Daft and El-Giahmi, 1974). Rhodes and Gerdemann (1975), using a split-plate system, found uptake of P by mycorrhizal hyphae from as far as seven cm from host onion roots. But Owusu-Bennoah and Wild (1979), using autoradiography techniques to examine uptake zones of P by hyphae unrestricted by split plants, noted uptake by mycorrhizal onion roots confined to a much smaller soil vol- ume. The effect of mycorrhizae was to increase the zone of P uptake from 1 to 2 mm from root surfaces, which is similar to the action of root hairs. Development of Mycorrhizae The life cycle of mycorrhizal fungi is synchronous with growth and development of the host. Being an obligate para- site, mycorrhizal fungi must rely almost exclusively on the host for a carbon and energy source. The fungus must colo- nize host roots and reproduce within the period of carbon availability. Patterns of fungal development have been noted in relation to host phenology. . Sutton (1973) noted a sigmoidal shape of the infection response curve over time. In a number of field-grown crops, a 20-25 day lag in infection was observed, supposedly caused by the time required for spore germination, germ tube growth and host penetration. This was followed by a period of extensive colonization, and then a leveling off of infec- tion. The great proportion of infected maize and bean roots was found in the top soil layers, and infection declined rapidly with increasing soil depth. Other hosts, however, exhibited more uniform infection throughout the root system (Sparling and Tinker 1978a). The period of rapid root colonization coincided with in- creasing amounts of external hyphae remaining attached to extracted roots. Extensive hyphal ramification throughout the soil probably accompanied the extensive internal coloni- zation. However, Bethlenfalvay et a1 (1982) found external hyphae to spread more quickly and to level off sooner than hyphae internal to roots. In low P soil, arbuscular forma- tion accompanies hyphal root colonization. In high P soil, arbuscles may be lacking and hyphae abnormal (Mosse, 1973), with hyphae confined to certain cells or in a profuse net- work of thin mycelium. Lack of arbuscles was also noted in shaded onions (Hayman, 1974). Since the colonization period occurred during highest nu- trient uptake, Sutton (1973) hypothesized that there may be a significant role for mycorrhizae in host nutrition. High levels of colonization of cereal crops occurring after peak P absorption and root growth were not thought to have much impact on host nutrition (Black and Tinker, 1979 and Hayman, 1970). Furlan and Fortin (1973) found the growth enhance- ment of onion by mycorrhizae to lag about four weeks behind the rapid rise in infection. Under high light intensity and favorable infection, a growth response of onion to infection was seen two weeks after transplanting seedlings over mycor- rhizal inoculum (Hayman, 1974). Sanders et al (1977) found that growth stimulation of onion lagged only slightly behind infection, and that P inflow increased with infection. The highest P inflow per percent length infected occurred when there was less than ten percent of the root length infected. Above ten percent infection, P inflow rose less steeply with increasing percent of root length infected. The rate of P inflow however, cannot be inferred from the extent of colonization, since Rhodes et a1 (1978) found zero P inflow through hyphae infecting 8% of roots in a high P soil. In general, P inflow through efficient hyphae follows a similar pattern to infection over time, while growth stimulation lags behind. It is probable that some P must accumulate to offset hyphal carbon demand, before growth stimulation begins (Stribley et al, 1980a). The leveling off of colonization of roots by mycorrhizal fungi coincides with plant maturation, and slower root and shoot growth rates (Sutton, 1973, and Saif, 1977). It is probable that cessation of colonization is governed by host physiology (Sutton, 1973). Bethlenfalvay et a1 (1982) linked the slowdown in hyphal colonization to the increased carbon sink of host reproductive structures. This is also the time reported to give greatest spore production (Sutton and Barron, 1972). The importance of early mycorrhizal infection to stimulate maximum growth has been illustrated (Smith and Smith, 1981). Infective units must be present in upper soil layers near young roots, as well as in lower soil layers to infect later developing roots. A high infection can result from a small amount of inoculum (Daft and Nicolson, 1969b). Early dif- ferences in infection may determine the extent of growth stimulation, regardless of final infection level. Carling et al (1979) found maximum infection of soybean by Glomus fasciculatus to occur at an average spore concentra- tion of 0.1 spores per gram of soil. Kotcon (1979) noted maximum plant growth of mycorrhizal onions infected with g; fasciculatus at a spore concentration of 0.4 spores per cm3 of soil, but counting spores may not give a good indication of a 5011's infectivity. For example, Porter (1979), using a soil dilution technique, observed good correlation between infectivity and spore counts in one soil but not in another. Powell (1976) found that as inoculum, infected roots were more effective than spores. Hyphae from infected pieces of root penetrated host roots rapidly, while formation of a pre-infection hyphal fan from germinated spores resulted in slower infection rates. Mycorrhizal hyphae may form exten- sive networks in the soil and serve as infection sources. Fungi that don't form spores (Baylis, 1969) may infect roots primarily through soil hyphae. Koucheki and Read (1976) thought this network important in seedling establishment. Hall's (1976) observations corroborated Powell's, but Kou- cheki and Read saw greater growth in plants with spores as inoculum. Manjunath and Bagyaraj (1981) also found equal growth stimulation of mycorrhizal plants given spores or infected root segments in sterile soil. In nonsterile soil, however, spore inoculum gave greater plant growth than root inoculum. Biological factors can reduce inoculum potential in nonsterile soils. Ross and Ruttencutter (1977) thought hyperparasitism might have played a role in significantly reducing spore levels of Glomus macrocarpus. Warnock et a1 (1982) found mycorrhizal benefits markedly reduced by the feeding of collembola on mycorrhizal fungi. Some biological factors also significantly reduced sporulation in nonsterile soil compared to sterile soil (Ross, 1980). The suppressant factor was able to move through sterilized soils, but not through 0.2 um pores. Mycorrhizal fungi may have some saprophytic capabilities, though they have not been grown in axenic culture separately from hosts. Mycorrhizal hyphae apparently became estab- lished in soil without a host by growing through pores in synthetic pouches and tubes through which spores could not pass (Warner and Mosse, 1980). After pouches and tubes con- taining inoculum were removed, host plants were subsequently infected when grown in the external soil. Soil containing inoculum and maintained at greenhouse temperatures for ten weeks had higher infectivity than soil containing inoculum and stored at 2 C (Ocampo and Hayman, 1981). If roots of a previous host remained in the soil during growth of the following crop, infectivity was higher than if roots were removed. This was true even if the roots were those of supposed nonhosts, members of the Cruciferae and Chenopodiaceae. Mycorrhizal hyphae externally colonized nonhost roots, particularly when a host was also present (Ocampo et al, 1980). Long-term survival in infected roots points to some capability of survival without an actively growing host. Tommerup and Abbott (1981) showed that hyphae of some mycorrhizal fungi could survive in roots for six months if stored at -500 bars water potential. New hyphae regenerated from out or broken hyphal ends when the infected roots were used as inoculum for a host. Mycorrhizal fungi can spread fairly rapidly through soil. Powell (1979) observed mycorrhizal fungi spreading 0.6-3.2 m per year. Spread of an introduced mycorrhizal fungus through a host stand was slower if another species was established in the soil or the host. Warner and Mosse (1982) found faster spread of mycorrhizal infection in higher root densi- ties. Mosse et al (1982) observed an average expansion length of 7.5 cm for Glomus caledonius 13 weeks after infes- tation of inoculum. After 15 months, the fungus had bridged a 4.5 m span between infestation points. Menge (1982) found re-establishment of mycorrhizal fungi within a few growing seasons if small pockets of infective units escaped fumiga- tion. The spread of mycorrhizal fungi could conceivably span greater distances through transport by wind, water, or movement of soil by man or animals. Plant P content has been implicated in the control of mycorrhizal infection. Sanders (1975) showed that foliar 10 infected P reduced mycorrhizal colonization and weight of external hyphae per length of infected root, rendering the existing hyphae inneffective with a calculated P inflow of zero through this hyphae. Menge et al (1978b), using a split-plate technique, observed concentrations of plant P, not soil P, inhibited colonization of sudangrass by mycor- rhizal fungi. High soil P levels did not inhibit infection if plant P was low. Gianinazzi-Pearson and Gianinazzi (1978) found a mycorrhizae-specific alkaline phosphatase with activity paralleling mycorrhizal infection, and the enzyme repressed by P. The mechanism of P suppressing infection is unknown, but root exudation has been hypothesized to be involved. Rat- nayake et a1 (1978) found that root extracts of reducing sugars and soluble amino acids increased with increasing plant P. However, total exudation was much increased at low plant P. Leakiness of roots dropped dramatically with in- creases in internal plant P. A threshold response was seen with exudation below 0.034% root P in sudangrass, and 0.068% root P in Brazilian sour orange. At root P above these levels, exudation of metabolites and ions was low, though declining slightly with increasing P. Lipid P was inversely related to leakiness, further pointing to membrane dysfunc- tion as a result of low plant P. 11 In a study of 13 wheat cultivars (Azcon and Ocampo, 1981), no clear-cut relationship between plant P concentrations and the ability to form mycorrhizae was found, though generally the mycorrhizae-forming cultivars had lower plant P. However, cultivars with little or no mycorrhizal formation maintained a high level of root sugar, but very little capacity to exude sugar. Upon mycorrhizal formation in the mycorrhizae-forming cultivars, an increase in plant P and the percentage of reducing sugars in the exudate was seen. Because the level of infection in the mycorrhizae-forming cultivars was directly related to sugar in roots or exudates, a minimum level of exudation necessary for fungal colonization was postulated. Jasper et a1 (1979) also found that infection was related to plant P and root carbohydrate levels. Clover was grown in low or high P soil and transplanted to low or high P soil with mycorrhizal inoculum. No infection had occurred ten days after transplanting. Soluble carbohydrates levels of roots at this time was inversely related to plant P. Plants with high soluble carbohydrates and low plant P had the highest frequency of hyphal penetrations 31 days after transplanting. 12 Mycorrhizal.Infection as Related to Phosphorus Though high plant P is generally associated with reduced mycorrhizal infection when compared to low P plants, infec- tion must be assessed over a wide range of soil P to deter- mine the infection-P relationship. Mosse and Phillips (1971) noted that a small amount of P must be added to 151- folium parviflorum in agar culture for infection to take place. Adding inositol, a carbon source, eliminated this need. Apparently, a minimum amount of P was necessary to produce sufficient carbon to support fungal activity. A number of investigators have reported that the addition of a small amount of P to extremely P deficient soil in- creased infection (Abbott and Robson, 1977b, Pairunan et al, 1980, Porter et al, 1978 and Sanders and Tinker, 1973). Infection, after reaching a maximum at low soil P, declined to slight or negligible levels at higher soil P. Mosse (1973) found, in soils given additional P, new infection of pre-infected onions halted if the soil P amendment was great- er than 1.0 g Ca(H2PO4)2 HZO/kg soil. Infection tended to disappear in soils given 3-6 9 phosphate/kg soil. The amount of P necessary to cause reduction in infection and the shape of infection-P response curves depends on a number of factors. Diffusion of P through soil, uptake of P by hosts, infectivity of host by mycorrhizal fungi and the sensitivity of the fungi to plant P, or even soil P, can 13 influence the infection response at a given soil P concen- tration. Because mycorrhizal plants will have altered plant P compared to nonmycorrhizal plants, comparing infection to soil P concentrations rather than to plant P concentrations is a logical method on which to base infection-P responses. The method of measuring infection could also change infec- tion-P response curves. Gerdemann (1968) pointed out that plants in fertile soil having more roots but a lower percent of roots infected than plants in low P soils, may have a greater total root length infected. The trend of weight or length of mycorrhizal roots declining less sharply with in- creasing P levels than percent of root length infected has been verified (Abbott and Robson, 1977b and Asimi et al, 1980). In one case, increasing P levels were associated with increasing weight of infected roots, while percent of infected roots declined at high P levels (Abbott and Robson, 1977b). Biermann and Lindermann (1981) found that the measurement of infection by the percent of root segments colonized, gave a higher estimate of infection than the measurement of infec- tion by the percent of root root infected. The size of root segments can magnify differences between methods. Giova- netti and Mosse (1980) found similar results in comparing the two methods, and suggested use of a root-grid intersec- tion method to estimate percent infection. They discussed 14 how methods to measure percent root length infected probably greatly overestimated the percent of cortical tissue infect- ed, since uninfected cells above or below infected cells seen from a two dimensional view would be declared 'infect- ed'. Strzemski (1975) used transverse sections of root segments and a 1-6 scale to estimate cortical infection. Christie et a1 (1978) rated infection by a 0-4 scale with five microscope fields viewing 2 cm root segments. These two methods have proven to be quite accurate in the measure- ment of colonizing mycorrhizal hyphae, as discussed by Gio- vanetti and Mosse (1980). Estimates of the percent of cor- tical cells infected gave much lower estimates of infection then estimates of infection by the percent of root length infected. Measurement of actual fungal tissue, such as Hepper's assay of chitin (Hepper, 1977), may be ideal for use in sterile soils, but be subject to error in nonsterile soils. Methods accounting for intensity of infection, as well as incidence, should more accurately reflect amount of fungal biomass present than would methods measuring just incidence, such as percent of root length infected. Infec- tion measurements should account for both incidence and in- tensity of infections. Several workers have attempted to relate infection esti- mates to fungal biomass. Tinker (1975b) and Sanders et a1 (1977) found good correlation between infected root length 15 and weight of external hyphae. Bethlenfalvay et al (1981) compared the measurement of infection as percent root length infected to the measurement of hyphae by the chitin assay (Hepper, 1977). Only when percent infection measured less than 60% of the root length infected, was a good correla- tion found between the two methods. Above 60% infection, the amount of chitin-measured fungal tissue was higher per length of infected root. The use of percent root length infected underestimated infection above the 60% mark. Accurate measurement of mycorrhizal infection must more precisely measure the amount of fungal material present. New methods are needed for measurement of mycorrhizal infection, at least in field-grown plants. A number of infection-P response curves have been pub- lished. Sanders and Tinker (1973) saw a slight rise in percent root length infected with increasing soil P amend- ment, followed by a steep drop. Jaspar et al (1979) saw a sharp drop in hyphal penetrations of subterranean clover roots with P additions in virgin soil, but a more gradual decline of infection with P addition in clover grown in soil from an adjacent, cultivated soil. The authors speculated that a history of P fertilizers may have selected P-tolerant fungi in the cultivated soil. Porter et al (1978) found no difference in infectivity between inoculum from soils with different P histories, however. 16 Stribley et al (1980b) compared infection of leek in non- sterile soil from ten sites, and in sterile soil from the same sites, but infested with Glomus mossae inoculum. Per- cent infection of leek in g; mossae-amended soil decreased gradually with increasing measured soil P concentration, and then dropped sharply at 150 ppm P. Mycorrhizal infection in nonsterile soils was high in some low P soils, but low in others. Percent infection decreased more gradually with P, and seemed to be slightly more P-tolerant in higher soil P in the nonsterile soils. Pairunan et a1 (1980) observed gradual drops in percent infection of subterranean clover with P addition, whether in the form of superphosphate or rock phosphate. Jensen and Jakobsen (1980) found linear decreases of percent barley roots infected with measured soil P. Combinations of host, fungus and soil are likely to influence infection-P respons- es. It is not surprising that reports differ. Spore Production as Related to Phosphorus Production of spores with different soil P levels has been shown to often follow a pattern similar to that of infection-P responses. Daft and Nicolson (1969a, 1972) counted most spores produced by maize mycorrhizae in treatments given lowest P. The earlier the P application in one experiment, the lower the spore counts. Infection estimates and spore 17 counts followed similar patterns. Hayman et al (1975) and Porter er a1 (1978) isolated spores from soil with different P histories. The highest spore counts were at intermediate P levels. The P level associated with maximum infection may not give the highest spore production, since Hayman and colleagues found highest infection at lowest P. Interaction of host P uptake capabilities and spore pro- duction was illustrated in the results of Menge et al (1978a). Spore levels rose, then declined with P in Bra- zilian sour orange infected with Glomus fasciculatus. In Troyer citrange, a cultivar with greater capacity for root P adsorption, spore numbers dropped sharply with added P. A lower amount of P was necessary to inhibit infection and spore production in the more P-efficient cultivar. Mycorrhizal Infection as Related to Nutrients other than P Nutrients other than P can affect mycorrhizal infection. Crush (19733) and Hall (1977) showed that mycorrhizal plants given complete nutrient solution minus P (C-P) consistently produced higher infection than mycorrhizal plants given just P or no nutrients. Mosse and Phillips (1971) increased infection of mycorrhizal Trifolium parviflorum in agar cul- ture, by adding sufficient iron. In these cases, the more vigorous growth resulting from adding nutrients may have caused greater carbohydrate production and higher P demand, and thus increased mycorrhizal infection. C-P nutrients may 18 have little effect on infection if those nutrients are non- limiting. Crush (1973b) and Baylis (1970) saw similar infection levels between plants given C-P nutrients and those given zero nutrients. Mycorrhizal Infection as Related to Nitrogen Addition of nitrogen (N) to soil has decreased mycorrhizal infection in some cases. In wheat (Hayman, 1970, 1975), and in peanut (Porter and Beute, 1972), higher spore levels of mycorrhizal fungi were present in plots with low N compared to plots with higher N application. However, the effect of N is variable (Hayman, 1975), since Sparling and Tinker (1978a) found no effects on grassland infection. Nitrogen did not reduce infection at low P in agar culture of mycorrhizal Trifolium (Mosse and Phillips, 1971). But addition of N at higher P did reduce infection. Nitrogen similarly had little effect on infection of field-grown bar- ley at low P, but interacted with P to decrease infection at higher P (Jensen and Jakobsen, 1980). Interaction of N with P at intermediate P levels has a physiological explanation. Thien and McFee (1970) observed increased P uptake and translocation in corn seedlings after a pre-treatment of NH4 or N03. Grunes (1959) reviewed N effects on P absorption. Application of N to soil can increase P uptake, particularly when both N and P sources are in close proximity in the soil. Differences in P uptake 19 were greatest at intermediate P. Until a direct effect of N on mycorrhizal infection is shown, the most likely explana- tion of the inhibition of mycorrhizal infection by N is an indirect effect associated with P absorption. Mycorrhizal Infection as Related to Soil Water Soil water levels affect mycorrhizal development. Reid and Bowen (1979) studied infection responses to a range of soil water levels maintained for eight days. Highest infec- tion appeared in soil with slight water deficit (-l.9 bars). Infection then decreased with decreasing soil water content. Soil saturation also inhibited mycorrhizal infection, as reported by others (Powell and Sithamparanathan, 1977, Redhead, 1975 and Trinick, 1977). Nelsen and Safir (1982b) observed lower spore production in water-stressed onions than in well-watered plants, with the decrease comparable to differences in host size. Mycorrhizal Infection as Related to Soil Tillage Soil tillage has been shown to influence mycorrhizal infection (Kruckelmann, 1975). Shallow plowing produced more spores than did non-filling. Rotary hoe-tilled soil produced least spores. Addition of chopped straw increased spore numbers when soil was shallow-plowed. 20 Mycorrhizal Infection as Related to Pesticides The use of pesticides may restrict colonization of host roots by mycorrhizal fungi. A number of fungicides, at least one insecticide and several nematicides have been shown to inhibit mycorrhizal formation (Safir, 1980). Mycorrhizal infection can be adversely affected by applica- tion of fungi toxicants to soil or seed or even by foliar applications of some fungicides. Mycorrhizal Dependency as Related to Host Root Development 3 of plants has been The degree of mycorrhizal dependence associated with host P sinks and capacity for P uptake. Mycorrhizal infection would be expected to increase if mycorrhizae could aid in P uptake and lower host P demand. Baylis (1970) grew six plants species in steamed-sterilized (nonmycorrhizal) soil and in nonsterile (mycorrhizal) soil. Coprosma robusta, with a poorly developed root system, was the only species that did not grow well when nonmycorrhizal. The species that grew well in steamed soil had well devel- oped root hairs, rhizoid or proteid roots, which are roots with finely developed rootlets arising in clusters. Infec- tion occurred in some of these species, but was poorly developed. Plants with low root-soil lengths needed more 3 - 'Mycorrhizal dependence' has been defined as the size of mycorrhizal plants as a percentage of the size of non- mycorrhizal plants. 'Mycotrophy' has been defined as a positive growth response to mycorrhizae. 21 soil P to grow without mycorrhizae than did plants with root hairs and finely branched roots (Baylis, 1972b). These plants with lower root-soil interfaces had greater mycorrhi- zal dependency. Crush (1974, 1976) found that mycorrhizal dependency was not directly related to plant P concentra- tions of nonmycorrhizal plants, but was inversely related to root weights. The effect of root development on mycorrhizal dependency of citrus was seen by Nemec (1978) and Menge et al (1978a). Cultivars with highest mycorrhizal dependency had the lowest root development and leaf P concentrations of nonmycorrhizal plants. In the work of Menge and colleagues, Brazilian sour orange, a cultivar with low root development and P uptake capacity, and Troyer citrange, a cultivar with greater root development and P-uptake capacity, were compared over dif- ferent P levels. Sour orange was mycotrophic over a wide soil P range, while citrange was mycotrophic only in low soil P. Baylis (1975) noted plants with magnolioid4 roots were mycotrophic in a wider range of soil P than were plants with graminoid roots. A survey of eighty-nine Brazilian forest 4 - Magnolioid roots are coarsely branched roots, generally lacking root hairs, and with ultimate rootlets greater than 0.5 mm in diameter. 5 - Graminoid roots are finely branched roots, often with well developed root hairs, and with ultimate rootlets less than 0.1 mm in diameter. 22 trees showed plants possessing magnolioid-type roots were disproportionately mycorrhizal, while trees with graminoid- type5 roots were nonmycorrhizal or only lightly infected (St. John, 1980). Grasses tend to be mycotrophic only in extremely P deficient soil. Sparling and Tinker (1978b) found growth depressions due to mycorrhizae in two of three soils examined. Only in soil with very low P were grasses mycotrophic. Experiments indicated that only at high alti- tudes in New Zealand was soil P low enough to promote myco- trophy (Crush, 1973b). Indeed, growth depressions due to mycorrhizae in higher P soils were seen. Only two of five native grasses seemed to benefit from infection in their natural soils, and those two grasses had lower root hair frequency. Root hair development has been negatively associated with the extent to which a plant is mycorrhizally dependent. Baylis (1972a) suggested that root hairs may have evolved to serve in the uptake of water. Plant species in moist tropi- cal forests seem root hair deficient and mycotrophic in a wide range of soil P. Plants that have evolved in areas nor- mally subjected to water stress have developed root hairs, and are not generally mycotrophic, except at very low soil P. Host Carbon Balance and Growth Response to Mycorrhizae Environmental factors affecting host growth and carbon balance can influence infection and mycorrhizal dependency. 23 Infection rates of onion infected by Gigaspora calospora increased with increasing temperature (Furlan and Fortin, 1973). At low temperatures, mycorrhizae were parasitic. Schenck and Schroder (1974) studied mycorrhizal infection rates and soybean growth under a wide temperature range. Mycorrhizal infection and root growth responded similarly to warm temperatures (20-35 C), but infection was less tolerant than was root growth to cooler temperatures. Onions infec- ted with Glomus mossae at high temperatures, high light intensity and long daylength exhibited higher mycorrhizal development and growth stimulation than onions grown in less optimal conditions (Hayman, 1974). At low temperatures and limited light, mycorrhizae depressed growth. Nonmycorrhizal plants given phosphate grew as well as mycorrhizal plants at optimum temperature and light, but grew better than mycor- rhizal plants at suboptimal conditions (Hayman, 1974). Some mycorrhizal systems however, may be adapted to low light intensities. Baylis (1975) and Redhead (1975) did not find mycorrhizal infection reduced with shading of host. Mycorrhizal onions infected with Gigaspora calospora and grown at low light intensities (5-10 klux) demonstrated higher infection and mycotrophy (at 10 klux), than did onions grown at high light intensities (15-20 klux) (Furlan and Fortin, 1977). 24 Environmental stresses can change a mycorrhizal relation- ship from symbiosis to parasitism. McCool (1981) subjected mycorrhizal and nonmycorrhizal tomatoes to ozone. Ozone eliminated mycorrhizal growth responses at optimum light, temperature and daylength. At suboptimal light, temperature and light conditions, a growth depression caused by mycor- rhizae was seen. Defoliation of mycorrhizal and nonmycor- rhizal maize, tomato and alfalfa decreased mycorrhizal in- fection and growth stimulation. The greater the extent of defoliation, the greater the growth depression (Daft and El-Giahmi, 1978). Generally, mycorrhizal benefits have been observed to be maximal in environmental conditions optimal for host growth, with the exception of low soil nutrition. The major processes involved in the mycorrhizal relation- ship are probably nutrient uptake through hyphae and trans- fer of an energy source from host to fungus. Host carbon-P balances will determine to a large extent, the role of mycorrhizae in plant growth. Mycorrhizal hyphae may have priority in the use of root carbon. Bevege et a1 (1975) calculated that mycorrhizal hyphae have an affinity for carbon four times that of host cells on a per weight basis. Environmental factors decreasing mycorrhizal benefits prob- ably decrease the carbohydrate needed for host growth, decreasing the host P demand, and thus tipping the balance of the host-fungus relationship from mutualism to 25 parasitism. Growth depressions at high P levels due to mycorrhizae have been attributed to toxicity from accumulated P (Mosse, 1973). Mycorrhizal plants had greater percent P in tissue at all P levels, however. The mechanism for P accumulation was operative at low P, but growth depressions were evident only at high P. Stribley et al (1980a), upon observing higher percent P in mycorrhizal plants than in nonmycorrhi- zal plants, calculated a 40-60% growth depression due to my- corrhizae. If plant P had been utilized to the same extent in mycorrhizal plants as in nonmycorrhizal plants, the size of the mycorrhizal plants would have been 40-60% larger. It is likely that some of the growth depressions observed in Mosse's (1973) experiments were probably due to carbon drain by mycorrhizal fungi. Crush (1973ab) observed that growth depressions in the greenhouse were either reversed or lowered in the field. Under greenhouse conditions, Glomus tenuis was thought to have a competitive advantage over other mycorrhizal fungi and to compete with the host plant for P. Cooper (1975) noticed growth depressions in some plant species when they were mycorrhizal in high P soil. The depressions tended to be short-lived. The onset of myco- trophy was thought to be caused by a drOp in soil P over time, and by increased plant P demand. Addition of C-P 26 nutrients increased host growth and decreased growth depres- sions due to mycorrhizae, probably from the increased host P demand and carbohydrate supply. Mycorrhizal Dependency as Related to Phosphorus Mycorrhizal benefits often disappear when plant P is increased (Sanders, 1975). Holevas (1966) was one of the first researchers to notice this effect when P was added to soil. If phosphate can manipulate mycorrhizal benefits, then a knowledge of these benefits under a wide range of soil P concentrations is helpful to understand the practical benefit of mycorrhizae. In agriculture, growers wish to gain the highest yield per economic input. By studying mycorrhizal benefits relative to optimal economic yields, the feasibility of exploiting mycorrhizae can be determined. In some soils, mycorrhizal plants have grown bigger than nonmycorrhizal plants in every P level tested (Figure 1a). This phenomenon may have several causes. In some cases (Kormanik et al, 1977 and Schultz et a1, 1979), mycorrhizal sweetgum grew better than nonmycorrhizal sweetgum at all P levels, though Kormanik (1981) ascribed a soil P threshold for sweetgum above which mycorrhizal and nonmycorrhizal plants grew equally well if other nutrients were nonlim- iting. In these cases, the authors speculated that hormonal factors were involved. Timmer and Leyden (1978, 1980) ob- served that when fumigation was used to kill Phytophthora, 27 A B Myc Nye N g ‘5’ t R a: K 5 NM NM O-a Soil P Soil P C D NM ‘V g Myc g Myc +’ E“ u N .5. 53 m R’ m NM Soil}> SoilP Figure 1. Patterns of plant growth of mycorrhizal (Myc) and nonmycorrhizal (NM) plants versus soil P concen- tration: Myc plants are larger than NM plants at all soil P concentrations (A). Myc plants are larger in a range of soil P concentrations, but Myc and NM plants are equivalent in size at some P concentration. Myc plants reach maximum size at a lower soil P concentration than do NM plants (B). Myc plants are larger than NM plants only in soil P, when plant size is submaximal (C). Myc plants are larger than NM in low P, but smaller at high soil P (D). 28 mycorrhizal fungi were also killed. High applications of P gave initial growth stimulation to citrus, but plants later became stunted from copper deficiency. Mycorrhizal plants given high P showed copper deficiency symptoms, presumably from inhibition of infection. Addition of moderate P and soil infestation of mycorrhizal inoculum gave highest growth. P seemed to stimulate growth of these plants until infection became established. If soils possess high rates of phosphate fixation, avail- able soil P may remain fairly low, and mycorrhizae may be of benefit to the host plant, despite the amount of P added. Hayman and Mosse (1971) and Mosse (1973) studied mycorrhizal benefits to onion in a number of soils. Two soils, with apparent high P fixation rates, consistently showed mycotro- phic growth of hosts, even with an amount of added P that caused growth depressions due to mycorrhizae in other soils. If mycorrhizal plants consistently outgain nonmycorrhizal plants at a P level giving maximal growth, practices to in- sure efficient mycorrhizal development may become necessary for maximal yields. In some experiments, maximal growth of mycorrhizal plants required a lower P level than did maximal growth of non- mycorrhizal plants (Figure 1b). In these situations, mycor- rhizae can partially substitute for P in that less P becomes necessary for maximal plant growth if mycorrhizae are 29 present. Maximum growth achievement in mycotrophic plants may involve nutrients other than P. Ross (1971) planted soybeans with and without mycorrhizal inoculum in fumigated field plots at three P levels. Initially, soil P was at an intermediate concentration. All mycorrhizal plants acheived a similar high yield, though the trend was toward decreasing yield with increasing P. Nonmycorrhizal plants with zero P addition were stunted. Addition of P increased yields of nonmycorrhizal plants to just below the level of the mycor- rhizal plants. Mycorrhizae thus compensated for the lack of P, and increased plant concentrations of P, N, Ca and Cu. Even if given high P, nonmycorrhizal plants may not attain maximal yield if concentrations of other nutrients are lim- iting growth. Mycorrhizal plants tended to decrease in yield with increasing P, presumably from mycorrhizal sup- pression, as Timmer and Leyden (1978, 1980) also found. A goal of optimum economic benefits may point toward main- taining soil P concentrations at intermediate levels and assuring that plants are mycorrhizal. Mycotrophy and maximal plant growth may only exist in a narrow P range. If P can be maintained within that range, substitution of mycorrhizae can be realized. For example, it appeared that if leek plants could be maintained within a soil P concentration of 60-120 ppm P, mycorrhizae could sub- stitute for fertilizer P (Stribley et al, 1980b). A narrow 30 P range also appeared to exist in an experiment with subter- ranean clover (Barrow et a1, 1977) for which mycorrhizae could substitute for P. Mycorrhizal dependence can vary among citrus cultivars, according to the soil P concentration. In the study of Menge and colleagues (1978a), mycorrhizae substituted for P and provided maximum growth (Figure 1b) in Brazilian sour orange, a cultivar with a poorly developed root system and low capacity for P uptake. However, mycotrophic growth was acheived at a P level providing for much less than maximum growth (Figure 1c) in Troyer citrange, the cultivar with a more highly developed root system and a greater capacity for P uptake. Well developed root systems can circumvent the need for mycorrhizae, and efforts to maximize yields of cul- tivars with this trait through mycorrhizae may be nonproduc- tive. Knowledge of host P uptake capabilities may give clues as to mycorrhizae substitution for P with particular hosts. Mycorrhizal benefits at P levels producing less than maxi- mal growth (Figure 1c) were described by Daft and Nicolson (1969a). Mycorrhizal maize, infected with Glomus macro- carpus, received one P application at l, 2, 5 and 7 weeks after transplanting with inoculum. Only with zero P appli- cation did mycorrhizal plants exceed nonmycorrhizal plants in size. Mycorrhizal plants of the treatment with zero P 31 applied were only half the maximum size. Daft and Nicolson (1966) saw decreased mycorrhizal dependency of tomato with increasing P levels, and with mycorrhizal dependency near zero at maximum growth. This trend of mycorrhizal growth benefiting plants only at low P and at submaximal plant growth was noted in subterranean clover by Abbott and Robson (1977b), Pairunan et al (1980) and Porter et a1 (1978). Powell (1977c) found that the greatest response of ryegrass to mycorrhizae was at intermediate soil P levels, but these P levels were not high enough to give maximal growth. Rye- grass showed no response to mycorrhizae at high P, and very little at low P. Hall (1978) found that differential re- sponses of maize cultivars to mycorrhizae were related to root growth. Two cultivars of maize and one of sweetcorn were grown at 2 P levels, with and without inoculum. At low P, one of the maize cultivars and the sweetcorn cultivar, with low root development compared to the other maize culti- var, responded to mycorrhizal infection. The second maize cultivar did not respond to infection, and mycorrhizae did not improve growth of any cultivar at high P. All cultivars responded to P, and grew double or triple the size of plants at low P. If mycorrhizae is of little benefit to producing optimal yields, cultural practices to insure maximal mycor- rhizal infection may be irrelevant. 32 Growth depressions due to mycorrhizae often occur at fair- ly high P levels, with mycotrophic growth occurring only at low P ( Figure 1d). Hall et a1 (1977) grew mycorrhizal and nonmycorrhizal plants of two white clover cultivars. They were grown at ten P levels, with mixed inoculum of three mycorrhizal fungi, placed under seeds of mycorrhizal treat- ments. Growth stimulation due to mycorrhizae occurred only at low P, and growth depressions caused by mycorrhizae were evident at high P. Crush (1973b) and Sparling and Tinker (1978b) observed mycotrophy of grasses grown in extremely P deficient soils with higher P levels inducing mycorrhizal parasitism. Mosse (1973), as discussed earlier, observed growth depressions in some soils at high P, though not in others. Baylis (1967) described growth depressions at high P in two host species, Coprosma robusta and Griselinia lit- toralis, which were highly mycotrophic at low P. Crush (1976) found that the probability of growth depressions was related to root development of the host species. Mycotrophy was found in red and white clover, though decreasing with increasing P levels. But alsike clover and lucerne (alfal- fa), species with long root hairs and high root/shoot ra- tios, showed 3-16% growth depressions when mycorrhizal. The degree of parasitism increased with increasing P levels. Crush cautioned, however, that an an initial growth depres- sion may be weighed against increased P absorption with 33 declining soil fertility, particularly in high plant densities. Growth depressions caused by mycorrhizae should be consid- ered when optimizing crop growth. Efforts to insure mycor- rhizal infectivity must be considered in light of the possi- ble detriment to maximal crop growth. Mycorrhizal Dependency as Related to Soil Water The improved uptake of water by mycorrhizal soybean (Safir et al, 1972) and onion (Nelsen and Safir, 1982a) over nonmy- corrhizal plants at the same P level has been linked to im- proved P nutrition. The effect of soil water on growth and uptake of P by mycorrhizal and nonmycorrhizal plants has been recently investigated by Nelsen and Safir (1982b). Imposition of water stress on mycorrhizal and nonmycorrhizal plants caused lower growth and uptake of P by nonmycorrhizal plants relative to mycorrhizal plants. The lower water availability appeared to decrease the availability of P to plants, probably through slower diffusion of P through the soil. Under water stress, plant P concentrations of nonmy- corrhizal plants apparently fell below a critical concentra- tion, thus inducing mycotrophy (Figure 2). Olsen et a1 (1961) examined the influence of soil moisture and soil P concentration on P uptake of corn seedlings. A linear relationship existed between P uptake and soil mois- ture content. An interaction of P uptake between soil Figure 2. 34 thu2Gnmuh SQfl.P PLmuanmnh Patterns of plant growth of mycorrhizal (Myc) and nonmycorrhizal (NM) plants versus soil P concen- trations under well-watered (A) and water-stressed (B) conditions. Under well-watered conditions, a maximum plant growth level (M1) is reached with adequate soil P. At some P concentration (Pl), the size of Myc and NM plants is equivalent. Upon water stress, the maximum level of plant growth (M2) is lower, and the soil P concentration (P2), at which sizes of Myc and NM are equivalent, is higher than under well-watered conditions. 35 moisture tension and soil P concentration was caused by a decrease in P diffusion through soil with decreasing soil moisture, as well as physiological effects on root uptake capabilities. In effect, uptake of P in a soil with a high P concentration but low soil moisture, may be equal to up- take of P in a soil with much lower P concentration, but higher soil moisture. In an agricultural system, management of soil P concentrations should account for frequency of water stress. More phosphate may be needed to achieve maximal yields in times of drought. Excess P may eliminate any mycorrhizal benefits, though if mycorrhizal, plants may be better able to withstand drought stress, because of higher P uptake capabilities. The interactions of P and water availability with mycorrhizae should be understood in order to acheive efficient resource conservation. Host Specificity and Adaptation oprycorrhizal Fungi VAM fungi, or endophytes as they are often termed, have extremely wide host ranges, but recent evidence points to some degree of host specificity. Nemec (1978) tested the mycorrhizal dependence of six citrus cultivars with the addition of three endophyte species at two P levels. The degree of mycorrhizal dependence for any cultivar depended upon the endophyte used. Upon addition of P, the order of mycorrhizal dependence among cultivars with a particular endophyte changed from the order observed at low P. Ocampo 36 and Hayman (1981) noted that Gigaspora margarita appeared to be less infective of lettuce than of onion, while Glomus fasciculatus infected both hosts well. Schenck and Kinloch (1980) monitored incidence of spores of mycorrhizal fungal species on six crop monocultures maintained for seven years after the clearing of woodland. Each crop appeared to be developing its own population of endophytes. While some species were increasing in some crops, others were decreas- ing. Acaulospora and Gigaspora species were predominately found on dicots, while Glomus species seemingly preferred monocots. Many species were specific to one or two crops. If a crop or crop rotation has been maintained over time, establishment of efficient indigenous endophytes may pre- clude necessity of introducing species. But certain endo- phytes may be desired, depending upon the host and on soil nutrition. Endophytes differ in their efficiency for nutrient uptake and in their ability to stimulate growth, though the source of soil P is apparently the same among species (Powell, 1975b). Some species greatly increase yields, particularly in low soil P, while other species provide little or no ben- efit (Abbott and Robson, 1977b, Gilmore, 1971, Jensen, 1982, Powell, l977ac, Powell and Daniel, 1978 and Sanders et al, 1977). Species giving highest infectivity may not always give greatest growth stimulation (Powell, l977a). 37 Soil factors or adaptation of fungi to soils can alter mycorrhizal efficiency. Mosse (l972ab) found that fungal species generally exhibited poor efficiency in acid soils, with performance among endophytes dependent upon pH-soil combinations. Some species showed poor ability to infect and stimulate growth in all soils. Abbott and Robson (1977a) found Glomus mossae predominately in cultivated sites with pH greater than 5.3. Acaulospora laevis was favored by pH less than 4.6. Others have noted the higher efficiency of g. mossae in high pH soils (Mosse, 1972b, 1975, Powell, 1977a and Stribley et al, 1980b). Baylis (1975) recognized that the effect of endophyte on host appeared to differ according to soil-fungus combina- tions, with performance often best in the soil from which the fungus was isolated. This was not true as a rule, how- ever. Lambert et al (1980) also presented evidence for adaptation of endophytes to soil. When inoculum was added to sterile soils, growth of birds-foot trefoil was greatest if the inoculum had been isolated from the soil used. Upon addition of P, strains not isolated from test soils de- pressed growth slightly. Superiority of indigenous endo- phytes in nonsterile field plots was not evident, however. Trefoil preinfected with indigenous strains had average yields compared to trefoil infected with fungi from 41 other sites, while growth advantages caused by any strain in the 38 nonsterile soil tended to be short-lived. Yields were corre- lated with soil chemical concentrations from the soils pro- viding the inoculum source. Endophytes adapted to soil with high Cu and Zn generally performed well. If fungal species have adapted to their soil and host, introduction of new species or strains may prove ineffective. Indigenous species tended to perform as well as or better than intro- duced species in stimulating clover growth in well-developed pastures (Powell, 1976, l977b). Efficiency of introduced species, compared to indigenous species, tended to be high- est in soil that was recently or had never been cultivated. The success of introduced species through time must be weighed when considering soil infestation of introduced species. Lambert et a1 (1980) found that growth benefits to plants given introduced mycorrhizal fungi to be of short duration. In some cases, Powell and Daniel (1978) also benefit of introduced species to diminish with time. Ini- tial differences in growth stimulation by some introduced species compared to indigenous species may be due to high initial infection rates. A truly efficient species will maintain superiority over time. Mycorrhizal fungi may adapt to soil P. Jaspar et a1 (1979) compared growth and infection of ryegrass between a virgin soil with a spore count of 320 spores Acaulospora laevis per kg of soil, and a cultivated soil with a spore 39 count of 400 spores of Glomus monosporus per kg of soil, with six levels of P added. Fungi in the cultivated soil were not only less sensitive to high P, but provided greater growth stimulation at low P. Differences in growth stimula- tion due to inherent efficiency of species and to previous soil P history appeared to exist. Similarly, Abbott and Robson (1978) observed greater growth in subterranean clover grown with two isolates of Glomus monosporus with a high P history, than in clover infected with Q. fasciculatus or indigenous species. Porter et a1 (1978) however, did not find any effect of past P history on the ability of isolates to stimulate subterranean clover. The success of adding introduced mycorrhizal fungi to non- sterile soil could be due to low soil inoculum potential or inefficient indigenous endophytes. An increase in efficien- cy of introduced fungi compared to indigenous fungi, may be obscured by the method of infecting the host or by adding inoculum. Early infection caused by preinfecting trans- plants or adding inoculum beneath seeds, can increase infec- tion rates (Smith and Smith, 1981). Hayman and Mosse (1979), Khan (1972, 1975ab) and Saif and Khan (1977) transplanted preinfected plants to unsterilized field soil, and observed growth increases compared to uninfected transplants in low P soil. These experiments may confirm the importance of early infection to growth stimulation in low P soil, but should 40 not be used as a rationale for adding inoculum to compensate for inefficient indigenous endophytes. Tests of endophyte efficiency should be conducted over long time periods to insure equal fungal establishment. Addition of inoculum may prove successful if infectivity of indigenous endophytes is low. Infestation of mycorrhizal inoculum under the seed in field plots increased yields of lucerne (alfalfa), onion and barley (Owusu-Bennoah and Mosse, 1979), barley (Clarke and Mosse, 1981), potato (Black and Tinker, 1977), cotton (Rich and Bird, 1974) and alfalfa (Az- con-Aguilar and Barea, 1981). Black and Tinker (1977) added 3 to soil with inoculum with a spore density of 13 spores/cm a spore density of 1 sporqu3. Addition of inoculum in- creased yield without any added P, but not upon addition of P. Addition of P significantly increased yield. Rich and Bird (1974) added inoculum with higher spore densities and lower nematode numbers to cotton plots in low P soil. An increased host growth was attributed to early infection of mycorrhizal fungi. Azcon-Aguilar and Barea (1981) found increased growth of alfalfa over two cuttings, when Glomus mossae was added to a phosphate-fixing low P soil. Owusu- Bennoah and Mosse (1979) and Clarke and Mosse (1981) ob- served higher infectivity of host roots in soil with added inoculum. Mosse (1977) tested the effect of adding mycorrhi- zal inoculum to 11 unsterile soils in which Stylosanthes was 41 grown. In four out of five soils with high natural infec- tivity, indigenous and introduced species gave equal growth. Addition of inoculum gave increased growth to plants in soils of low natural infectivity. Addition of rock phos- phate altered the environment for indigenous endophytes. In three of the four cases in which growth of plants with indi- genous and introduced species was equivalent in low P soil, the introduced species proved superior with the addition of P. Indigenous species appeared less P tolerant. Addition of inoculum to soils in which indigenous endo- phytes have been killed by fumigation may be necessary in low P soils. Ross and Harper (1970) and Timmer and Leyden (1978) found it beneficial to add inoculum to fumigated soils growing soybean and citrus. Kleinschmidt and Gerde- mann (1972) found that citrus seedlings grew normally only when mycorrhizal fungi escaped fumigation. Additional experiments involving increased onion growth with introduced species of mycorrhizal fungi are those of Mosse et a1 (1969), Mosse and Hayman (1971) and Khan (1981). Mosse et a1 preinfected onions with mycorrhizal fungi, and transplanted mycorrhizal and nonmycorrhizal seedlings to both sterile and nonsterile soil. Highest growth stimula- tion was noticed in sterile soil, but preinfected plants were also larger than noninfected onions in nonsterile soil. Mosse and Hayman (1971) preinfected onions with Glomus 42 mossae or added g. mossae inoculum under seeds. Infection with Q. mossae improved plant growth in four soils, whether in sterile or nonsterile soil. Khan (1981) preinfected onions with four species of mycorrhizal fungi, and observed growth effects after 12 weeks in nonsterile coal wastes. Glomus fasciculatus and Q. mossae gave greatest growth, while g. macrocarpus produced intermediate results. Sclero- cystis rubiformis and indigenous endophytes yielded poor growth. Crop rotation could provide a management tool for manipu- lation of mycorrhizal fungi in soil. Powell (1979) investi- gated the spread of introduced fungi in soil. The introduced species became established more quickly if a cruciferous pre- crop depressed infectivity of indigenous mycorrhizal fungi. Although mycorrhizae can superficially colonize roots of nonhosts to some extent (Ocampo and Hayman, 1981), spore numbers may be so depressed by nonhost crops or fallow fields, that infection of subsequent hosts may be depressed (Black and Tinker, 1977, 1979). Monocropping for long periods can not only change composition of fungal species (Schenck and Kinloch, 1980), but may dramatically influence spore numbers. Kruckelmann (1975) counted more spores from plots of wheat and oats than from those of potatoes or beets. All were monocropped for 16 years. The sequence of hosts in rotations may give clues as to possible success, as well as 43 the need to add mycorrhizal fungi. Mycorrhizae probably cannot rival photosynthesis in impor- tance to plant development, as one author has stated (Kor- manik, 1981). But in some situations, mycorrhizae can play a significant role in nutrient uptake and plant growth. Mycorrhizae interact with pathogenic fungi and nematodes (Schenck, 1981 and Schenck and Kellam, 1978), but a review of of these effects has been omitted in this discussion. The feasibility of mycorrhizae improving plant growth must be examined before management of mycorrhizae to control diseases could ever become practical. Few studies of mycorrhizae have been conducted in the field, and the role of mycorrhizae in crop productivity is poorly understood. To ascertain it's importance, investiga- tion of mycorrhizal occurrence in field situations was de- sired, with onion (Allium ggpa L.) as the host that was chosen for study. Onion is highly responsive to mycorrhi- zae, probably because of a shallow root system and lack of root hairs. If any annual crop can be managed for mycorrhi- zal benefits, onion would be a good candidate. 44 Chapter I. Mycorrhizal Infection in Commercially Grown Onions Sampling of commercial onion fields was planned in order to study the occurrence of mycorrhizal infection, with the sampling designed so that mycorrhizal occurrence could be related to environmental factors. Hopefully, knowledge of the occurrence of infection and limiting factors in commer- cial fields would provide a framework for more careful in- vestigation of mycorrhizal ecology. Materials and Methods I A. 1979 Commercial Field Survey Onion fields near Grant, Eaton Rapids and Lainsburg, MI were selected for 1979 surveys1 (Appendix 1). The fields near Grant were sampled on 29 August 1979 and those near Eaton Rapids on 30 August 1979. The onions near Eaton Rap- ids and Lainsburg were organically managed (no pesticides used), while those near Grant commonly received fungicide applications to control foliar diseases. The survey was conducted just prior to harvest, so that large effects of pesticides on mycorrhizal infection, if present, would be likely to be observed. Within a field, 10 sites were ar- bitrarily chosen, and three plants with some roots intact were harvested from each site. Soil from around the roots 1 - Samples were taken with the cooperation of Tom Ellis, of the Entomology Department. 45 was collected. Ten roots from each site were stained for mycorrhizal infection using a modification of the Phillips and Hayman (1970) method (Appendix 2). The amount of mycor- rhizal hyphae in a 1 cm segment from each stained root was judged as high (3), moderate (2), low (1) or not present (0). Spores of Glomus and Gigaspora were extracted and counted from 5 grams of soil from each sitez. The spore extraction technique was a modification of the sucrose-density gradient method of Ohms (1959) (Appendix 2). Soil phosphate analysis was conducted on soil from 7 fields by the Michigan State University Soils Laboratory. I B. Mycorrhizal Infection in Nonsterile Soils An experiment was designed to compare mycorrhizal infec- tion of onions grown in a number of unsterile soils under uniform conditions. Soils were collected from five commer- cial onion fields and from the Michigan State University Muck Farm. The commercial sites included fields GO, GB, GS, 66 and ERl (Appendix 1). Soils were sieved through 1/4 inch screen and added in equal volumes to 8.6 cm Styrofoam cups. There were four replications per soil. Onion (Allium cepa L.) c.v. Downing Yellow Globe seeds were planted approxi- mately 2 cm deep. Pots were kept under artificial lights until emergence, at which time plants were thinned to two 2 - Infection ratings and spore counts were conducted with the help of Charles Nelsen of the Botany and Plant Pathology Department. 46 per pot, and pots were arranged in a randomized complete block design in a greenhouse. Nine weeks after planting, tops and roots were removed from the soil and the upper 8 cm of roots were stained for the purpose of rating mycorrhizal infection. One 1 cm segment from each of ten roots was rat- ed for mycorrhizal infection, on a 0-4 scale. The numbers represent an attempt to categorize mycorrhizal infection on a logarithmic scale. The estimation of infection in this manner was designed to eliminate the error inherent in mea- surement of infection by the percent root length infected method (Bethlenfalvay et al, 1981). By assessing the se- verity of infection in 1 cm root segments, the method esti- mated hyphal mass within the root, rather than the length of infected root. This system was used to rate infection of roots in all subsequent experiments. The rating key is as follows: 0 - no infection, 1 - only a few entry points visible, 2 - small areas of mycorrhizal hyphae within less than 5-10% of the root length, 3 - hyphae present in most of the root length, but not intensely, or intense hyphal colo- nization in less than 50% of the root length, and 4 - in- tense hyphal colonization throughout the root segment. After estimation of root infection, analysis of variance was performed on the average mycorrhizal rating per pot. 47 I C. Mycorrhizal Infection in Sterile Soils With Added Spores Another experiment was initiated to control mycorrhizal inoculum levels, and to eliminate the effect on infection of biological factors differing among the nonsterile soils. Soil from the identical sites as in the previous experiment was sieved and autoclaved3 for 50 minutes at 20 psi and 250 F. Two m1 of a spore suspension of Glomus etunicatus chla- mydospores were sprayed on the soil in three layers, with the spore concentration per 2 ml counted at 123.4 I 4.94 (mean 1 standard error). Inoculum was added in this way, since soil inoculum had introduced pathogenic microorganisms previous- ly. The soil layers onto which the spores were added were 15, 40 and 65 mm from the soil surface in 8.6 cm styrofoam 3 of soil. cups, with soil 90 mm deep and utilizing 320 cm Downing Yellow Globe onion seeds were planted approximately 2 cm deep and thinned to three plants per pot after emer- gence under artificial lights. Pots were then transferred to a greenhouse, and arranged in a randomized complete block design with four replications. Nine weeks after planting, ten roots from each pot were extracted and stained, and a 1 cm segment from each of the stained roots was rated for mycorrhizal infection. Analysis of variance was performed on the average rating per pot. A representative soil sample 3 - The length of autoclaving may not have sterilized the soil. 48 was analyzed for soil P by the Michigan State University Soils Laboratory. By taking into account the bulk density (grams of dry soil per cm3 of soil), soil P concentrations were calculated as pg/cm3. Soil P was calculated by the Soils Laboratory as pounds per acre, which was estimated by multiplying ppm P by 2. The calculation of pg/cm3 is done by multiplying ppm P by the bulk density. This calculation allows comparison of P concentrations among soils on a volume basis. I D. 1980 Commercial Field Survey Six commercial onion fields near Grant and Eaton Rapids, MI were selected for sampling during the 1980 growing sea- son. One to three two-foot plots were established in each field (Appendix 3). Again, the Eaton Rapids field was organically managed, while those near Grant commonly received fungicide applica- tions. The Eaton Rapids field was included for late season sampling to observe any dramatic differences in mycorrhizal infection that were evident between the two types of manage- ment. Sampling consisted of extracting soil cores from di- rectly beneath a plant, or 4, 12 or 20 cm away. Soil depth was sampled down to 20 cm. Plants were arbitrarily selected within a plot, and core locations were measured from the bulb center. Cores were taken throughout the season to ascertain the timing of infection. Sampling began in 49 mid-June, and terminated by early September, when plants were senescent. Five to seven root segments from each core were stained, and 1 cm lengths were rated for mycorrhizal infection. Four to eight cores were extracted from each plot at each sampling. Soil P was measured, using 1.7 grams of air-dried soil from cores extracted on the mid-June and the late-August to early-September dates. The extraction method was the Bray Pl method (Bray and Kurtz, 1945)( Appendix 4), which mea- sures the soil P that is soluble in dilute HCl. This is an accurate measure of plant available P in soil pH below approximately 7.6 (Bray, 1948). I E. Addition of Mycorrhizal Inoculum to Commercial Onion Fields A series of experiments was planned in which mycorrhizal inoculum was to be added to commercial fields in order to observe whether inoculum was limiting infection. Four fields near Grant, MI, coded GO, 61, GB and 65 (Appendix 5) were selected for these experiments. The experiment at each location was established after the commercial onions had emerged. The experimental design was a randomized complete block with four replications. Four 16-foot row lengths were staked, 6-8 rows apart, and were treated as blocks. Within each block, four 2-foot subplots were estabished, separated by 2-foot guard plots. Soil inoculum of Glomus etunicatus was added under two randomly 50 selected plots in each block. As a control, soil without Q. etunicatus was added under the other two plots in each block. The added soil was previously collected soil from the same field and autoclaved, with small amounts of mycor- rhizal pot culture soil or autoclaved pot culture soil mixed thoroughly within for the infested and noninfested treat- ments, respectively. Spartan Banner onions were seeded in a layer of soil above that which was added. A soil core sam- ple was taken adjacent to each experimental unit in order to estimate soil P concentrations. Experiments were set up in GO and GS fields on 14 May 1981, with an estimated 600 spores added per infested experimental unit. Experiments were set up in GO, 61 and GB fields on 21 May 1981, with an estimated 187,000 spores per infested experimental unit. Onions in GS soil were consumed by onion maggots, conse- quently infection ratings of onions were not available. The first experiment in the GO soil was sampled on 25 June 1981, and the other three experiments were sampled on 16 July 1981. Three soil cores, 4 cm from a plant and 0-15 cm below the soil surface, were taken from one infested and one nonin- fested treatment per block. Cores from an experimental unit were bulked, and six 1 cm root segments were extracted, stained, and rated for mycorrhizal infection. Soil P anal- ysis was conducted on a representative 1.7 g air-dried sub- sample from each bulked sample. 51 I F. 1981 Commercial Field Survey Surveying of mycorrhizal infection in commercial onion fields was repeated in 1981, and included soils with a wide range of soil P concentrations. Samples were taken twice to gain additional information on infection timing. Three two-foot plots were established in each of seven commercial onion fields near Grant, Michigan (Appendix 5). The fields were divided into approximate thirds, with one plot situated in each third. On two dates, 25 June 1981 and 20 August 1981, each plot was sampled for mycorrhizal infection, spore counts of 912- mps and Gigaspora species and soil P concentrations. Three soil cores, 4 cm from a plant and 0-15 cm below the soil surface, were taken arbitrarily within each plot. Two G9 plots were not sampled on 20 August, because onions had been harvested. Marker stakes were missing in field G8 on 20 August, so samples were taken from new plots within the field. Soil bulk density was estimated by weighing wet soil from each core, and then adjusting the soil weight to the equivalent dry weight after a soil subsample was weighed, dried at 85 C, and reweighed. Six roots were extracted from each core, then stained and rated for mycorrhizal infection. Spores were extracted and counted from a 5-7 gram soil sam- ple, and soil P analysis was conducted on 1.7 g of air-dried soil from each core. Spores collected from all fields were 52 bulked, in order to attain a large number of spores for identification of mycorrhizal fungi. Results I A. 1979 Commercial Field Survey Onions in some commercial fields exhibited low mycorrhizal infection ratings, while onions in other fields had higher mycorrhizal ratings (Table l). The average rating for Grant fields was 0.70, which was lower than the 1.25 average of the organic fields near Eaton Rapids and Lainsburg. How- ever, spores were present in all fields in relatively equal amounts (Table l). The lower ratings in the Grant fields may have reflected higher soil P concentrations in those fields, since an inverse relationship (R2=0.86) was found between soil P concentrations and mycorrhizal infection rat- ings (Figure 3). Onions in Grant field GS had the highest infection rating and the lowest soil P concentration, despite fungicide applications (Table 1). I B. Mycorrhizal Infection in Nonsterile Soils A significant difference in mycorrhizal infection ratings occurred among nonsterile soil treatments. High ratings were found in onions grown in the GS soil, while infection remained low in onions grown in the other five soils (Table 2). 53 Table 1. Mycorrhizal ratings, spore counts and soil P concentrations from 1979 commercial field survey Spore Count Mycorrhizal (spores/5 9 Soil P Field Rating of soil) (leha) GO‘l 0.76 29.2 - 60-2 0.13 96.6 — 61 0.13 41.6 - G2 0.82 63.8 - G3 0.11 27.6 392.6 G4 0.33 72.2 276.4 GS 1.98 26.4 10.6 G6 0.93 36.6 - ERl 1.11 26.0 248.8 ER2 0.89 30.6 248.5 ER3 1.39 16.5 221.5 MF 1.62 62.8 90.8 54 3.0 HYCURRHIZRL RRTING T r ‘3 ch 130 250 350 400 SOIL PHOSPHORUS Ohfl0 Figure 3. Mycorrhizal rating of onions versus measured soil P concentration (lb/A) from 7 fields sampled in the 1979 commercial field survey. The regression equation is Y = 2.12 - 0.005(X), where Y is the mycorrhizal rating and X is the soil P concentra- tion. 55 Table 2. Mycorrhizal ratings of onions and soil P concen- trations in autoclaved soil with inoculum and in nonsterile soil Sterile Soil With Inoculum Nonsterile Soil Soil P Mycorrhizal Mycorrhizal Soil ( cc) Rating Rating 65 113.4 0.48 ax 0.20 a G3 58.2 0.55 a 0.04 a GO 53.0 0.65 a 0.34 a ERl 36.8 0.53 a 0.52 a MF 14.8 2.60 b 0.38 a GS 1.6 3.45 c 2.75 b x - Numbers followed by the same letter are not significant- ly different, according to Duncan's Multiple Range Test (P=0.0S). 56 I C. Mycorrhizal Infection in Sterile Soils With Added Spores Significant differences in mycorrhizal infection ratings of onions were found among the sterile soils with added spores. Infection was highest in G5 soil, which had the lowest soil P concentration (Table 2). The mycorrhizal infection of onions grown in MF soil was fairly high, and this soil also had a relatively low soil P concentration (Table 2). Infection ratings in the other four soils were uniformly low, with soil P concentrations greater than or equal to 36.8 ug/cm3. Generally, infection ratings were higher in the autoclaved soils given inoculum than in the nonsterile soils, particularly in the MF soil. I D. 1980 Commercial Field Survey Maximum mycorrhizal infection ratings were reached by July, approximately halfway through the growing season. Infection ratings of onions grown in some fields were much higher than infection ratings of onions grown in other fields (Figure 4). The location of cores relative to onion plants appeared to have no influence on the infection rat- ings of roots taken from the core. High mycorrhizal infection ratings were found only in onions grown in soils with P concentrations below 30 pg/cm (Figure 5). The lowest soil P concentrations and the high- est mycorrhizal infection ratings were observed in soil and in onion roots, respectively, collected from marl soils. 57 c.’ v- . GS 9 CD a)- Z d E q G4 + a: . ER1 :6 =2- hd at J: . 0: a: . D g d 1: ‘5’. ol 0. I 5 j j r fi I I T f I I 1 r I I 150 170 190 210 230 250 JULIRN DRTE Figure 4. Mycorrhizal rating versus time in onions sampled from 6 fields in 1980. Field codes are given in Appendix 3. NYCORRHIZRL RHTING 1.0 HYCORRHIZHL RRTING 58 4.0 2.0 -0 I I I I I I I I J O 0 0 j V I I I V V V 50 V I T 100 j V V I 1 V 150 V YT fl 1 V 200 son. PHOSPHORUS (pg/ems) 250 1.0 2.0 I cum I I I I I I I I I I I I I I I I I V V T T V V 50 V 1 V V 100 V V " fii 150 T V V l V 200 “9'I—I_l— 250 son. PHOSPHORUS (Jig/cm“) Mycorrhizal rating of onions versus measured soil P concentration (pg/cc) from midseason (A) and late season (B) 1980 commercial field survey. Soil P thresholds are estimated by dashed lines. Figure 5. 59 The low P concentrations probably reflect the high CaCO3 content and high P fixation rates of these soils. The marl soils also exhibited higher soil pH (6.5-7.5) than did the other soils with higher sand or muck contents. The highest soil P concentrations were found in the sandy soils. Onions in fields G4, GS and ERl showed equally high infec- tion ratings by the end of the season (Figure 4). The G4 and GS fields had received fungicide applications, while the ERl field had not. These three fields however, exhibited 3, while onions in soil P concentrations below 30 pg/cm 'fields GO, GI and GB, grown in soil P concentrations higher than 30 ug/cm3, had low mycorrhizal infection througout the season (Figure 4). Spore counts from the commercial field soils were quite low compared to concentrations of about 1000 spores per gram of soil that were observed from greenhouse pot cultures. 3 of soil. All ' Numbers ranged from 0.5 to 5.0 spores per cm plots contained spores, and plots containing onions with high infection ratings did not appear to have higher spore concentrations than did plots containing onions with low infection ratings (Table 3). However, field GS showed the highest spore concentration in late season. Analysis of variance showed that the horizontal distance of the sample from the row had a significant effect on spore numbers, while depths of 5-30 cm below the soil surface did not (Table 4). 60 Table 3. Spore counts in soils from 1980 commercial field survey Spore Countx Date Field Plot (spores/cc) Mean St.Dev. NY 6/10 Gl 1 1.49 1.73 6 GB 1 1.09 0.67 5 G5 1 0.57 0.63 3 6/25 61 1 1.54 1.80 6 Gl 2 1.49 1.68 6 7/01 G0 1 3.30 1.05 2 G4 1 0.79 0.92 2 G4 2 1.17 0.04 2 G4 3 4.22 1.97 2 GS l 5.80 5.08 2 GS 2 1.65 0.35 2 GS 3 2.47 0.17 2 7/23 GO 1 4.37 2.34 8 61 1 1.37 1.19 8 61 2 0.96 0.91 8 GB 1 1.03 0.70 7 G4 1 1.05 0.47 8 G4 2 3.34 3.61 8 G5 1 5.78 3.37 8 G5 3 4.97 2.43 8 x - Spores were extracted from soil 5-15 cm below the soil surface and either directly below a plant or 4 cm from the bulb center. y - Number of samples 61 Table 4. Spore concentrations in soil extracted at four distances from the onion row Spore Concentration Location (spores/cc) Center of double row 2.48 ax 4 cm from double row 2.47 ab 12 cm from double row 2.28 ab 20 cm from double row 0.54 b x - Numbers followed by the same letter are not significant- ly different according to Duncan's Multiple Range Test P=0.05 I E. Addition of Mycorrhizal Inoculum to Commercial Onion Fields Of 569 root segments from the three sampled fields ex- amined for mycorrhizal infection, only two showed any mycor- rhizal hyphae. The soil P concentrations from these fields (Table 5) were in the range of soil P concentrations (>30 pg/cm3) associated with minimal mycorrhizal infection in the field. Table 5. Soil P concentrations from fields to which inoculum was added Soil P Concentration Field (pglcc) co 52.6 i 1.4" Gl 57.1 1 1.8 GB 75.2 1 3.2 x - Mean 1 Standard Error 62 I F. 1981 Commercial Field Survey Mycorrhizal infection was only high in onions grown in soil with P concentrations below 30 pg/cm3 (Figure 6). In- fection ratings were higher in the low P soils (<30 pg/cm3) at the late season sampling date, than at the midseason sampling date (Figure 6). As in 1980, low soil P and high mycorrhizal infection were found in onions grown in marl soils. A few organic or sandy soils exhibited low soil P and high mycorrhizal infection (Table 6, Appendix 5), so evidently the mycorrhizal condition is not associated with just marl soils. In general, the sandier soils showed the highest P concentrations and low mycorrhizal infection. Mycorrhizal infection did not appear to be related to spore concentrations. Soil in all plots contained spores of mycorrhizal fungi, but plots with onions possessing high mycorrhizal infection did not appear to have higher spore concentrations than plots of onions with low mycorrhizal infection (Table 6). Onions in field G5, plot I, appeared to have the greatest spore production, as deduced from the differences in spore counts between the two sampling dates (Table 6). This plot, located in marl soil, exhibited pH values of 7.2-7.6. Elg- mgs mossae, favored by high soil pH (Abbott and Robson, 1977a, Mosse, 1972b, 1975), appeared to be the dominant en- dophyte in the Grant commercial onion fields and may have been favored by the higher pH soils. Glomus etunicatus was 63 4.0 2-0 1 HYCORRHIZRL RRTING 1.0 0-0 0 60 100 160 200 260 sou. PHOSPHORUS (pg/cm") 4.0 3.0 . 00' HYCORRHIZRL RHTING 2.0 1.0 °b so 100 150 200 250 SOIL Pnosrnoaus 019/ch) Figure 6. Mycorrhizal rating of onions versus measured soil P concentration (ug/cc) from midseason (A) and late season (B) 1981 commercial field survey. Soil P thresholds are estimated by dashed lines. 64 Table 6. Soil P concentrations, mycorrhizal ratings and spore counts from 1981 commercial field survey 25 June 1981 Soil P Mycorrhizal Spore Count Field Plot (pglcc) Rating (spores/cc) Mean St.Dev. GO 1 59.5 0.00 0.66 0.38 2 78.8 0.00 0.48 0.40 3 13.8 1.83 0.96 0.24 61 1 246.3 0.00 0.19 0.44 2 79.2 0.00 0.73 0.19 3 138.7 0.00 0.66 0.19 GB 1 105.3 0.00 0.30 0.36 2 181.7 0.00 0.06 0.10 3 140.2 0.00 0.32 0.43 GS 1 9.8 2.39 0.36 0.31 2 27.7 0.00 0.53 0.54 3 64.1 0.06 0.86 1049 G7 1 204.2 0.32 0.89 0.21 2 191.8 0.00 1.22 0.28 3 159.3 0.00 1.26 0.33 G8 1 69.4 0.00 1.21 0.71 2 104.7 0.22 2.15 0.80 G9 1 29.7 1.78 0.69 0.17 2 121.8 0.22 0.92 0.40 3 136.2 0.05 0.72 0.71 65 Table 6. (cont'd) 20 August 1981 Soil P Mycorrhizal Spore Count Field Plot (ngcc) Rating (spores/cc) Mean St.Dev. G0 1 70.6 0.39 0.52 0.16 2 82.1 0.00 1.65 0.68 3 21.2 3.03 1.56 0.76 61 1 172.5 0.11 0.27 0.24 2 70.2 0.11 1.15 1.07 3 71.5 0.00 0.83 0.61 GB 1 88.8 0.00 0.10 0.10 2 62.2 0.00 0.38 0.10 3 44.4 0.00 0.41 0.21 GS 1 10.8 3.06 4.52 1.94 2 16.2 1.94 0.31 0.00 3 22.9 1.50 1.08 0.35 G7 1 175.4 1.44 1.28 0.40 2 189.8 0.00 0.71 0.12 3 119.3 0.00 1.21 0.07 68x 1 71.5 0.00 6.95 3.10 2 107.2 0.00 1.93 1.39 3 102.0 0.00 1.02 0.17 G9 1 10.7 2.67 0.43 0.43 2 - - .. .. 3 - _. - .— x - plots within dates. field GB were not the same on the two 66 also frequently found. Spores of Gigaspora gigantea and Sclerocystis rubiformis4 were occasionally extracted. Sclerocystis spores were difficult to quantify, since their dark color blended with organic particles retained in the spore extracts. Discussion Mycorrhizal infection of onion roots occurred at high levels in commercial fields, but these levels were only observed in soils with P concentrations below 30 ug/cm3. Infection remained quite low in soils with P concentrations above 30 pg/cm3. In the soils with low P concentration, the highest infec- tion ratings were reached only past midseason. If growth stimulation lags behind infection (Furlan and Fortin, 1973), then growth benefit of mycorrhizae to field-grown onions probably occurs primarily after midseason. Mycorrhizae may still significantly increase nutrient uptake, since the bulk of dry matter production occurs after midseason (Sutton, 1973). Without proper experimental controls however, the function of mycorrhizal infection in commercial plant pro- duction remains a point of speculation. 4 - G. mossae and g. etunicatus were identified with the assistance of Dr. Barbara Daniels of Kansas State Uni- versity, and S. rubiformis was identified with the as- sistance of Dr. James Gerdemann of the University of Illinois. 67 There was no evidence of detrimental effects of pesticides on mycorrhizal infection, and soil P appeared to be the ma- jor factor that limited infection. However, the number of fields included in this study was too small to make definite conclusions about pesticide effects on mycorrhizal infection. The marl soils consistently exhibited high mycorrhizal infection and low soil P. Even if high P amounts were added to these soils, low P concentrations were consistently found, suggesting a high rate of P fixation. The high pH of these soils may have favorably increased mycorrhizal infection, especially since Glomus mossae was a predominant fungal spe- cies in the Grant area soils. The marl soils may be ideal for management of mycorrhizal benefits. Finding relatively small numbers of mycorrhizal fungal species in the Grant fields may suggest adaptation of these species to the common crops planted (Schenck and Kinloch, 1980), or adaptation of these fungi to the soils. 9. mossae has been previously linked to cultivated soils, particularly those with high soil pH (Abbott and Robson, 1977a). Spores of mycorrhizal fungi were present in all fields studied, though it is unknown how spore numbers related to inoculum potential, since mycorrhizal hyphae within the soil may be more efficient at infecting roots than spores, (Powell, 1976). The spores present in fields with soil P 3 concentrations above 30 pg/cm appeared viable, since their 68 volume was filled with oil droplets. Some spores may have been produced in soils of high P, even though mycorrhizal infection remained minimal. These spores could also have represented a holdover population of fungi from times of lower soil P. Several of the fields sampled showed little or no mycorrhizal infection observed in all three years of sampling. The presence of spores in these may reflect long-term survival. The inoculum present in nonsterile field soils gave lower mycorrhizal infection than when the soil was autoclaved and infested with Glomus etunicatus spores. The inoculum poten- tial of the nonsterile soils may have been too low to cause high mycorrhizal infection. It is possible however, that autoclaving the soil eliminated microorganisms that would naturally inhibit mycorrhizal infection. The experiment in which high levels of spores were added to commercial fields, demonstrated that soil inoculum may not limit mycorrhizal infection if soil P concentrations are too high. The concentrations of the soils to which spores 3 P concentration that were added were above the 30 pg/cm seemingly limited mycorrhizal infection. The trend of mycorrhizal infection with increasing soil P concentration showed a threshold type response in 1980 and 1981, with high infection only below a maximum soil P con- centration. The linear trend fitted to the 1979 data did 69 not agree with trends observed after more accurate methods of estimating mycorrhizal infection and soil P were adopted, and with inclusion of a wider soil P range. 70 Chapter II. The Effect of Four Phosphorus Levels, With and Without Glomus etunicatusypin Two Fields Onions have been reported to be highly responsive to in- troduced mycorrhizal fungi in pots (Mosse, 1973, and Mosse and Hayman, 1971), and in the field (Owusu-Bennoah and Mosse, 1979). Field experiments were planned to test the response of onions to Glomus etunicatus Becker and Gerdemann in muck soil. This species contributed significant growth benefits in laboratory experiments (Nelsen and Safir, 1982a). Four P levels were included in the experimental planning, to test the practicality of mycorrhizae replace- ment of fertilizer P. If mycorrhizae could give growth benefits at or near maximum plant growth (Figure lb), then the feasibility of mycorrhizal substitution for P could be further explored. However, if mycorrhizae could only increase plant size at submaximal growth levels (Figure 1c or 1d), then mycorrhizae might only serve as a sign of P deficiency. The experimental sites were chosen so that soil P concentrations were sufficiently low, to give a growth response to added P. A plot set up in 19791, showed no response of onions to Glomus etunicatus, or to two levels of added P in soil with a soil test reading of 97 kg/ha P. 1 - The plot was established by Dr. Bernard Zandstra of the Horticulture Department of Michigan State University. 71 Materials and Methods II A. 1980 Experiments 2, with one con- Two experiments were established in 1980 ducted in the same field used in 1979 (high P field), and the other conducted in a virgin field, cleared of vegetation the previous fall, with an initial soil P concentration of 3 pg/cm3 (low P field). The experimental design was a randomized complete block design with three replications and three thirty-foot rows per experimental unit. Treatments included four levels of added P, with and without soil inoculum of Glomus etunica- ggs. The inoculum was banded under the row at a rate of approximately 2500 spores per meter of row. Phosphate, in the form of triple superphosphate, was banded approximately 5 cm under the row at rates of 0, 30, 97 and 193 kg/ha P. N and K were added at rates of 70 kg/ha and 234 kg/ha, respec- tively. Spartan Banner onions were seeded on 23 May 1980. Root samples for the estimation of mycorrhizal infection were taken on 12 August and 10 September in the low P field, and on 30 July and 10 September in the high P field. Sam- pling on 30 July and 12 August, consisted of taking a soil core, 4 cm from a plant and 5-15 cm below the soil surface, from two locations in each treatment in the first block. 2 - The plots were established and maintained with the cooperation of Dr. Zandstra. 72 The 10 September sampling consisted of three cores extracted from block one treatments. Seven root segments were ex- tracted from each core, and were stained and rated for my- corrhizal infection. In addition, soil P concentrations were determined using 1.7 grams of air-dried soil from each core. Spores of Glomus and Gigaspora species were extracted and counted from soil taken from each row on 24 September3. On that date, onion bulb fresh weight of 10 m of row was determined from all experimental units from both fields. Analysis of variance was performed on the yield data. II B. 1981 Experiments 4 Experiments were established in 1981 , repeating the 1980 methods. A virgin low P field, cleared from vegetation in the fall of 1980, and the same high P field as in the previ- ous two years, was utilized. Initial soil P concentrations of the low field were minimal, measuring 1.0 1 0.16 ug/cm3 (mean 1 standard error) and the initial spore concentration of Glomus and Gigaspora species was 0.06 spores/cma. The initial P concentration of the high P field was 10.9 1 0.59 pg/cm3 P, with a spore concentration of 0.13 spores/cm3. Of the three thirty-foot rows per experimental unit in the low P field, one was divided into two-foot subplots, 3 - The spore counts were made by Charles Nelsen of the Botany and Plant Pathology Department. 4 - The plots were established and maintained with the cooperation of Dr. Zandstra. 73 separated by two-foot guard plots, and one subplot per experimental unit was randomly selected at 3, 6, 9, 12 and 15 weeks after planting for sampling. Plants were harvest- ed, counted and weighed after drying at 75 C., Ten arbitrar- ily selected roots from each plot were extracted and stained, and a 1 cm segment from each root was rated for mycorrhizal infection. By 21 weeks after planting, the tops had senesced, and bulbs from one meter in the center of the middle row in each experimental unit were topped and weighed. The bulbs were also dried at 75 C and then weighed. Soil P was measured from soil samples collected on four dates. At 6 and 9 weeks after planting, one core was ex- tracted from selected subplots. Core locations were 4 cm from a plant and 0-15 cm below the soil surface. Three of these cores were taken from selected subplots 9 weeks after planting, since variability was high in the previous sam- ples. Fifteen weeks after planting, three cores, 4 cm from a plant and 5-15 cm deep, were extracted from sampled sub- plots. The 0-5 cm soil depth was not sampled at this time, because it was thought that a lack of P in this profile con- tributed to the high soil P variabilities found previously. After soil was used for P measurement 15 weeks after plant- ing, soil from the three cores per subplot was bulked, and spores were extracted and counted from 5-7 grams of soil. 74 At 16 weeks after planting, three adjacent soil cores, 4 cm from a plant and 5-15 cm deep, were extracted from the center of the middle row of all experimental units in the high P field. Soil from the cores was bulked, and a portion used for soil P analysis. Spores were extracted and counted from another soil subsample. Bulb fresh and dry weight was measured from onions collected from one meter of row in the center of middle rows at 21 weeks after planting. Analysis of variance was performed on the mycorrhizal rat- ings from the high P field. Analysis of variance was also performed on the mycorrhizal ratings from all samples taken from the low P field, by treating time as a subplot factor in a split-plot design. Analysis of variance was also per- formed on spore data from both fields. If plant size was density-independent, analysis of vari- ance was performed on the mean dry weight per plant for each date. If plant size was density-dependent however, analysis of covariance was performed on the mean dry weight per plant after a reciprocal transformation, with plant density as a covariate. The test for density-dependence was a signifi- cant slope for regressions of the reciprocal of mean plant dry weight (w) against density. Frappell (1973) showed that the relationship of the reciprocal of 'w' to density was linear for onion growth. Plants harvested at 15 and 21 weeks after planting showed density-dependent plant size. Bulb fresh weights from the harvest 21 weeks after planting 75 were also subjected to analysis of covariance, after a re- ciprocal transformation of 'w'. If growth was not density- dependent, the reciprocal transformation biased means, and for this reason, analysis of variance or covariance was not performed collectively on the yield data collected from all dates in the low P field. Results II A. 1980 Experiments Mycorrhizal ratings declined sharply with increasing soil P concentrations at all sampling times. Mycorrhizal ratings were high only when soil P concentrations measured less than 5 ug/cm3. Ratings of roots from soil above this P concen- tration were quite low (Figure 7). The addition of 93 and 197 kg/ha P apparently inhibited mycorrhizal infection (Fig- ure 8), while ratings were low from all onions collected from the high P field. No effect of inoculum on mycorrhizal infection was noted (Figure 8). No significant differences in bulb yield due to the inocu- lum or P were found, though variability was high because of stand differences. However, the plants in the low P field treatment given zero P, seemed to be slightly submaximal in size (Figure 8). It appeared that maximal plant growth was achieved with a minimum of 30 kg/ha P. High mycorrhizal infection was found in plants achieving maximum growth in the 30 kg/ha treatment of the low P field, while low Figure 7. 76 4-0 A ‘3 I c: 7 ° ' z 2 ° ' e ° ; 529- I E‘“ I I I m g I B o°' a: ‘33 I “ I I 0 ' 0 I 0. ‘ D II 1. g - . 1 j I v i ‘b IE_' 20 so 40 son. PHOSPHORUS (ug/cm") 5! V B °. CD to" | 2 O I— 0: 6° I O: l age- I N N ' H I I r: a: I O U I E oO-I ' ‘ 0 I .. I O I . ‘ ‘ O J A A o. V fl—Y ' V ' ' V o 10 20 so 40 sou. PHOSPHORUS (Hg/ems) Mycorrhizal rating of onions versus measured soil P concentration (pg/cc) from 1980 Muck Farm ex- periment: Data is from 30 July and 12 August sampling (A) and 10 September sampling (B). Circles represent data from low P field, and tri- angles represent data from high P field. Soil P thresholds are estimated by dashed lines. 77 In :1 014 'w J b 00-1 N * +Spores ' -9 1 ‘ _¢9 0‘) CD . ' -—spor68 I-Z-I . '-" I— ‘iqu_‘ ‘..'“\\\ . SE B "I _J V _l __°. E . 64 3 .1 :5 iii .1- ~ 3‘: >' I o ' £3 4 .5? z: "3- "' c: . I k 9‘ O V I V l V V V l I T r I I j I I V U V '— . c’o 40 so 120 160 200° PHOSPHORUS TREFITIIENT (kg/ha) Figure 8. Mycorrhizal rating of onion roots and bulb yield versus total P (initial+added P) of the 1980 Muck Farm low P field experiment. The upper two lines represent yield data and the lower two lines rep- resent mycorrhizal ratings. 78 3-0 SPORES/CII3 .0 + low P field +spores low P field -spores In high P field +spores 0 high P field -spores Figure 9. V 110 210 310 4.0 MYCORRHIZRL RRTING Mycorrhizal spore numbers in soil versus mycor- rhizal ratings of treatments of the 1980 Muck Farm experiments. The regression equation is Y = 0.44+0.43(X), R=0.84, where Y is the spore number and x is the mycorrhizal rating. 79 mycorrhizal infection was found in the 97 and 193 kg/ha treatments, and bulb size of these treatments was also maxi- mal (Figure 8). All P levels in the high P field appeared to give sufficient P, so that plant growth was not P-limit- ed, and mycorrhizal infection was quite low in these treat- ments (Figure 8). The observed soil P concentrations were much less than that expected if all added P remained soluble. It appeared that less than 10% of added P was recovered according to the soil P tests. Spore numbers were highest in treatments with high infec- tion ratings. Low spore numbers were found in the high P field and in the highest two P levels of the low P field (Figure 9). Addition of inculum may have increased spore numbers slightly in the treatments with high infection rat- ings (Figure 9). II B. 1981 Experiments The effect of P on mycorrhizal infection was significant, as was time, and the interactions of P and time, and spores and time (Table 7). Infection ratings of some treatments increased with time, as might be expected (Table 8). Infection ratings were relatively high in the treatments given zero P three weeks after planting (Table 9). The ad- dition of mycorrhizal inoculum significantly reduced infec- tion ratings in treatments given 30 kg/ha P. Interestingly, the addition of inoculum significantly increased plant 80 Table 7. Analysis of variance of the mycorrhizal ratings of the 1981 Muck Farm low P field experiment Source g; as 5 Block 2 - - Phosphorus=P 3 35.84 ** Spores=s 1 1.39 ns P x S 3 0.29 ns Error (a) 14 0.38 - Time=T 4 3.40 ** P x T 12 0.96 ** S x T 4 0.99 ** P x S x T 12 0.36 ns Error (b) 64 0.26 - ** - Significant at the 1% level ns - Not significant at the 10% level Table 8. Mycorrhizal ratings of the 1981 Muck Farm low P field experiments treatments with time Mycorrhizal Rating 0 kg/ha P 30 kg/ha P Week -spores +spores -spores +spores 3 2.27 ax 2.70 a 1.50 ab 0.23 a 6 3.30 b 3.43 a 0.97 a 2.00 b 9 2.90 ab 3.27 a 1.90 bc 1.93 b 12 2.93 ab 2.80 a 2.63 c 2.73 b 15 2.70 ab 3.20 a 2.77 c 2.83 b 97 kg/ha P 193 kg/ha P Week -spores +spores -spores +spores 3 0.67 a 0.37 a 0.20 a 0.23 a 6 0.40 a 1.83 b 0.13 a 0.83 a 9 0.70 a 1.27 b 0.17 a 0.53 a 12 1.13 ab 1.87 b 0.63 a 0.57 a 15 1.90 b 1.77 b 0.43 a 0.13 a x - Numbers in a group followed by the same letter are not significantlty different, according to Duncan's Multiple Range Test (P=0.05). Table 9. Mycorrhizal ratings and plant weights from the 1981 Muck Farm low P field experiment Time P Added (week) (kggha) Mycorrhizal Rating Plant Dry Wt (g) -spores +spores -spores +spores 3 0 2.27 a 2.70 a 0.0079 a 0.0070 a 30 1.50 ab * 0.23 b 0.0091 ab * 0.0119 c 97 0.67 be 0.37 b 0.0111 b 0.0097 b 193 0.20 c 0.23 b 0.0098 b 0.0109 bc 6 0 3.30 a 3.43 a y0.079 a 30 0.97 b 2.00 b 0.139 b 97 0.40 b 1.83 b 0.160 b 193 0.13 b 0.83 c 0.131 ab 9 0 2.90 a 3.27 a 0.52 a 30 1.90 b 1.93 b 1.58 b 97 0.70 C 1.27 be 1.67 b 193 0.17 C 0.53 c 1.17 b 12 0 2.93 a 2.80 a 2.48 a 30 2.63 a 2.73 ab 5.96 b 97 1.13 b 1.87 b 5.97 b 193 0.63 b 0.57 5.52 b 162 0 2.70 a 3.20 a 6.61 a 30 2.77 a 2.83 a 10.85 b 97 1.90 a 1.77 b 12.58 b 193 0.43 b 0.13 c 11.99 b Time 9 Added Bulb Freshy Bulb Der (week) (k ha) Weight (9) Weight (g) 21x 0 53.3 a 4.87 a 30 114.8 b 10.85 b 97 106.4 b 10.05 b 193 112.4 b 10.81 b. Denotes a significant difference between the -spore and +spore treatments according to the LSD test (P=0.05). Numbers in a group followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). - Represents the average of -spores and +spores - Plant weights adjusted for density by covariance analysis NM 82 weight at the 30 kg/ha P level, from 0.0091 grams per plant to 0.0119 grams per plant. This was the only case of inocu- lum affecting plant weights in either experiment. A minimum of 30 kg/ha P appeared to provide sufficient P to provide for maximum plant growth (Table 9). At six weeks after planting, infection was highest in treatments given zero P, while plant weights of this treat- ment were lowest (Table 9). The addition of inoculum sig- nificantly increased infection ratings of treatments given 30, 97 or 193 kg/ha P, as well as the average of all treat- ments with and without inoculum. Plant weight was not af- fected by the increases in infection ratings (Table 9). A minimum of 30 kg/ha P again appeared sufficient for healthy plant growth. Infection ratings were significantly increased by the addition of mycorrhizal inoculum in the treatments given 97 kg/ha P at 9 and 12 weeks after planting (Table 9). This significant increase was not found in other treatments at these times, and again this change in mycorrhizal infection ratings did not affect plant sizes. Infection ratings de- clined with increasing P levels at 9 weeks after planting, but by 12 weeks after planting, infection ratings of treat- ments given 30 kg/ha P were equivalent to those in treat- ments given zero P (Table 9). Maximum plant weight had been attained by the addition of 30 kg/ha P, with high mycorrhi- zal infection occurring simultaneously. Maximum plant size 83 also occurred upon addition of 97 and 193 kg/ha P, but my- corrhizal infection ratings were quite low (Table 9). No effect of inoculum on infection ratings or plants weights was evident at 15 weeks after planting (Table 9). Infection was high in treatments given 0 or 30 kg/ha P, relatively high in the treatments given 93 kg/ha P and low in the treatments with 193 kg/ha P (Table 9). Plant weight was again low at the 0? level, and high with the addition of P. The same trend in bulb fresh and dry weight was noted at the final harvest, which was 21 weeks after planting. High infection ratings were related to soil P concentra- tions at 6, 9, 12 and 15 weeks after planting. Highest in- fection ratings were observed from plants growing in soil with P concentrations of approximately 1.0 pg/cm3. Infec- tion ratings declined quite sharply with increasing soil P concentration. In soil with P concentrations above 20 ug/cm3 , infection ratings of onions were uniformly low (Figure 10). Unlike the 1980 experiments, soil spore concentrations did not differ among treatments. Treatments given inoculum tended to have higher spore counts than treatments without mycorrhizal inoculum. Treatments given 0 kg/ha P also tended to have higher spore counts than treatments given higher P. However, the differences were not significant. The mean spore concentration at week 15 in the low P field was 0.26 spores/cm3. The predominant species was Glomus 84 urcoanutznt nnriuo e 9 Q Q I A , B I I I l a a o. I o . I " I 3" I 0.0 I ,_ I o. 0' g 'I I I o 0 4° '5 " ' I .. I I I I 0 x .I g I . I g I °od . . gee-I . I - o , "' I C I .I 9' I .l . o 0' . OI . O. J A . O. l . A o 'V'I'f'fi"1"‘l"']"*1'*1 o ' 'I'rvrvvvuvvvyvvvlrrv r- o 20 40 so no no no 140 0 20 40 00 00 100 120 :40 SOIL PHOSPHORUS (pg/ans) SOIL PHOSPHORUS (pg/cm’) e. 9 Q Q C , D I I I ° 9: I 66 0' a” o z I z I o- "' I E i E q “ l °= . I .1 69d I God :I 2‘.“ I 2‘." I t z ' . a: I I: S I S ' u ' I . u I "e . "o g o-w.‘ t ".4 .I - o, "' I I. P b P o I 9 0 .0 0 a J a 1 . A a. 7vv'va'iwv'vvv‘YYV'vvvr"V a vv‘vvv17V1'Vvvv'vvv—ITVVIVYV 0 20 40 00 00 100 120 140 o 20 40 oo oo :00 no no Figure 10. son. Pnosrnonus (In/uni) son. PHOSPHORUS Wan“) Mycorrhizal rating of onions versus measured soil P concentration (pg/cc) of the 1981 Muck Farm low P field experiment. Data is from 6 weeks (A), 9 weeks (B), 12 weeks (C) and 15 weeks (D) after planting. Soil P thresholds are estimated by dashed lines. 85 etunicatus5 , the same species that was added in the soil inoculum. The addition of mycorrhizal inoculum significantly in- creased mycorrhizal ratings in the treatments given zero P, and measured at 16 weeks after planting (Table 10). Infec- tion ratings of the higher P treatments were very low and high mycorrhizal ratings in this field found only when soil P concentrations fell below approximately 20 pg/cm3 (Figure 11). The amount of P initially present in the soil was appar- ently sufficient to provide maximum plant growth. No dif- ferences in plant sizes were found upon addition of P (Ta- ble 10). There were no differences in the production of spores found among treatments, according to spore counts measured 16 weeks after planting. The mean spore concentration of 3 Glomus and Gigaspora species was 0.14 spores per cm of soil. Glomus etunicatus also predominated in this field. 5 - This species was identified with the assistance of Dr. Barbara Daniels of Kansas State University. 86 Table 10. Mycorrhizal ratings and bulb fresh weights of 1981 Muck Farm high P field experiment 9 Added Bulb Freshy (k ha) Mycorrhizal Rating Weight (g) "SEOI‘CS +spores 0 0.57 ax * 2.20 a 114.7 a 30 0.00 a 0.07 b 115.9 a 97 0.10 a 0.10 b 103.8 a 193 0.07 a 0.00 b 105.0 a x - Numbers in a column followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). * - A significant difference exists between -spore and +spore treatment, according to the LSD test (P=0.05). y - These numbers represent the average of the -spores and +spores treatments, and have been adjusted for plant density by covariance analysis '87 4.0 I‘IYCURRH IZFIL RFITING 2.0 L 1.0 0 I90 0 rI—r—o-oJe-I-w—poeo-ro-r-I—‘er—r—I—qe..9fl.p-o—I—I 20 40 60 80 100 120 140 16 sou. PHOSPHORUS (ug/em”) 0.0 I Figure 11. Mycorrhizal rating of onions versus measured soil P concentration (pg/cc) of the 1981 Muck Farm high P field experiment. The soil P threshold is estimated by the dashed line. 88 Discussion The response of mycorrhizal infection to soil P concentra- tion in the Muck Farm experiments paralleled the observa- tions from the commercial fields, in that infection ratings higher than 2.0 occurred only in onions grown in soil with low P concentrations. The rate of hyphal colonization appeared to be related to added phosphorus. Colonization was rapid in the treatments given zero P in the 1981 experi- ment in the low P field. If a lag phase (Sutton, 1973) in the infection process existed in the mycorrhizal coloniza- tion of these onions, it was of short duration. The treat- ments with higher P applications did exhibit apparent lag phases before a period of rapid root colonization by mycor- rhizal fungi. This lag time may be due less to the time required for spore germination and root penetration, than to the time required for root zone P to decline sufficiently to cause a drOp in the amount of P absorbed. At some point, host P would be low enough for infection to proceed rapidly (Menge et al, 1978a, and Sanders, 1975). Apparently, host P was sufficiently high to inhibit mycorrhizal infection in 3 in 1980, onions grown in soil P above approximately 5 ug/cm and above approximately 20 pg/cm3 in 1981. The soil P threshold restricting mycorrhizal infection appeared to be lower in Muck Farm soil than in the commer- cial fields. This could be due to several factors. Soil P may diffuse more quickly through Muck Farm soil, and onions 89 may attain a plant P concentration inhibiting mycorrhizal infection at a lower soil P concentration. A difference in soil P thresholds could also be attributed to measurement of soil P around the roots. If added P did not form a uniform concentration around the roots, then the surrounding soil of low P concentration at the Muck Farm may have diluted the amount of P extracted. If this occurred, then the soil P threshold limiting infection may have been underestimated in the Muck Farm soil. The soil P threshold may also have been higher in 1981 than in 1980, though this is difficult to judge because of the limited number of 1980 samples. In 1981, the addition of inoculum to the treatments given zero P had no significant effect on mycorrhizal infection ratings. Apparently, mycorrhizal fungi were colonizing the onions roots so rapidly that any differences were not dis- cernible. However, in higher soil P concentrations in both 1981 experiments, the addition of mycorrhizal inoculum was found to increase infection ratings, though this increase did not affect plant sizes. Mycorrhizal fungi in these P concentrations were colonizing roots at a much slower rate than in treatments given zero P, and the addition of inocu- lum may have provided more entry points per root length. Differences were then discernible because of the slow spread of hyphae in the roots. Microorganisms present in the soil inoculum may have favored the environment for colonizing mycorrhizal hyphae. 90 This cannot be overlooked, since soil without mycorrhizal inoculum was not added to the noninfested treatments. The effect of factors other than inoculum was minimized however, because of the small amount of added soil compared to the surrounding soil volume. High mycorrhizal infection resulted from low spore con- centrations. It is possible that high levels of infection occurring early in the season resulted from fungal hyphae within the soil, rather than from spores. The fungal species added in the soil inoculum, Glomus etunicatus, may have been ineffective in increasing yields, since it was the same species that predominated indigenous- ly. Infection from the indigenous fungi may have been suf- ficient to provide growth benefits at the P levels studied. Sanders (1975) and Rhodes et a1 (1978) found that low my- corrhizal infection at high soil P concentrations were not functional in increasing plant P. Presumably, the low in- fection observed in the treatments given 193 kg/ha P in the 1981 low P field experiment had little or no effect on plant weights, and these plants were probably gaining maximal plant size without benefit of mycorrhizal infection. The ineffectual increase in mycorrhizal infection with added inoculum, supports the idea that low or moderate infection is nonfunctional in increasing plant growth, at least in soil P concentrations near the observed soil P threshold. 91 Mycorrhizal infection ratings did not reach high levels in the treatments of the 1981 low P field given 30 kg/ha P, until near midseason. Infection only reached relatively high levels at the end of the season in treatments given 97 kg/ha P in the same experiment. Cooper (1975) found that mycorrhizal benefits increased with time, presumably as soil P concentrations dropped. By 12 weeks after planting in the 1981 low P field, the soil P concentrations in the root zone of the onions in the treatments given 30 kg/ha P may have dropped, and the high mycorrhizal infection may have been able to maintain maximum yields. However, without proper experimental controls, the role of mycorrhizal in providing growth benefits can only be a point of speculation. Both the 1980 and 1981 experiments in the low P fields demonstrate however, that high mycorrhizal infection and maximum yields can be achieved simultaneously, though this infection level may not be reached until midseason. The P concentrations providing high mycorrhizal infection and max- imum plant yields simultaneously, probably fall within a narrow range (Figure 10, Table 9). If soil P is below that range, plant P becomes insufficient for maximum yields. If soil P concentrations are above the soil P threshold limit- ing mycorrhizal infection, then mycorrhizal benefits are probably eliminated. In 1980, the soil P concentration of the treatment in the low P field given zero P was approxi- mately 4 ug/cm3, and yields were only slightly submaximal. 92 In 1981, the soil P concentrations of this treatment were approximately 1 ug/cm3, and yields were clearly submaximal. A critical plant P concentration is often necessary for maximum plant growth (Epstein, 1972), and a soil P concen- tration above these levels was evidently necessary to provide maximum plant growth in the Muck Farm experiments. The increase in plant weight, but a decrease in infection three weeks after planting with the addition of mycorrhizal inoculum in the treatment given 30 kg/ha P in the low P field, was unexpected. This observation may be similar to Cooper's (1975) findings of early parasitic effects of my- corrhizae. It may also be possible that a higher root growth rate of the plants given inoculum may have diluted the colonizing mycorrhizal fungi, so that differences in the hyphal biomass per root length were seen. Spore production may increase with high infection, as found in 1980, but this may not hold as a rule, as the 1981 results showed. The sampling at 15 weeks after planting may have been prior to the time of high spore production, how- ever. Perhaps the high variability in spore counts obscured any differences. Spore production may be triggered by envi- ronmental conditions, which were not as favorable in 1981 as in 1980. Spore counts probably should not be used as an in- dication of mycorrhizal infection, for they would have poor- ly predicted 1981 infection levels. 93 Chapter III. The Potential of Mycorrhizal Substitution for Fertilizer P A series of greenhouse and growth chamber experiments was planned to further the potential of mycorrhizal substitution for fertilizer, since nonmycorrhizal controls were not pres- ent in the field experiments. Because the commercial soils apparently differed in P retention capacities, the benefit of mycorrhizae was to be investigated in several soils. Materials and Methods III A - I. Mycorrhizal Benefits in Two Soils With Three P Levels Soil from fields GS and GO was sieved through 14 inch screen, autoclaved for 50 minutes at 20 psi and 250 F, and a portion was analyzed for soil P. Initial soil P concentra- tions for the G5 and GO soils were 26.2 1 2.73 ppm P (mean 1 standard error) and 62.2 1 1.12 ppm P, respectively. The concentrations corresponded to 20.4 and 45.0 kg/ha P for GS and GO soils, respectively. The chosen P amendments were 0.0, 2.58 x 10.3 and 5.16 x 10.3 grams of P per pot, with P added in the form of KH2P04. These treatments were selected to give overlapping soil P concentrations in the two soils, if all added P became available, with GS soil with P concen- trations of 20.4, 45.0 and 69.6 kg/ha, and GO soil with P concentrations of 45.0, 69.6 and 84.2 kg/ha. Soils were added to 6.5 cm styrofoam cups to give equiva- 3 lent 160 cm volumes. Soil inoculum of Glomus etunicatus, 94 weighing 3.0 1 0.05 grams was added to half the pots, 2-3 cm below the soil surface. The spore count of the soil inocu- lum was 695 1 46 (mean 1 s.e.) per 3.0 grams of soil. Three Spartan Banner onion seeds were planted in a layer of soil over the inoculum. Nutrients were added to the soil in enough water to saturate the soil. Potassium (K) was added in the fOrm of KNO3, so that each pot received 6.51 x 10-3 grams of potassium (K) from the KH2P04 and from the KNO3. Similarly, each pot received 2.34 x 10.3 grams of nitrogen (N) by balancing the N received in the KNOB, and by adding N in the form of NH4NO3. Plants were thinned to one per pot after emergence. Pots were arranged in a randomized com- plete block design with 2 replicates in each of 4 blocks in a growth chamber, conditions being 22 C day/l6 C night tem- perature, and a 14 hour daylength. The light intensity was 5.2 x 104 ergs/cmz/sec at one-half of maximum plant height. At 5 and 7 weeks after planting, plant tops from four rep- licate pots of each treatment were harvested, dried at 75 C, and weighed. At 5 weeks after planting, six 1 cm root seg- ments from each pot were rated for mycorrhizal infection. Representative soil samples from pots harvested at 5 weeks after planting were analyzed for soil P. Analysis of var- iance was performed on the dry weight per plant, after a natural logarithm transformation to insure homogeneity of variance. Analysis of variance was also performed on the infection and soil P data. 95 III A - II. Mycorrhizal Benefits in Two Soils With Three P Levels A second experiment was established in a similar manner. Initial soil P concentrations were 37.4 1 1.26 ppm (mean 1 s.e.) or 24.5 kg/ha in GS soil, and were 70.3 1 0.65 ppm or 50.9 kg/ha in G0 soil. P treatments were again chosen to give overlapping soil P concentrations of the two soils, if all added P became solu- ble. The pots received either 0, 2.77 x 10.3 or 5.54 x 10-3 grams of P per pot. This corresponded to total P levels (added P + initial P) of 24.5, 50.9 and 77.3 kg/ha in G5 soil and 50.9, 77.3 and 103.7 kg/ha in G0 soil. Each pot received a total of 6.97 x 10.3 grams of K and 2.51 x 10-3 grams of N. The measurement and analysis of plant weights, mycorrhizal infection, and soil P was conducted as in the previous ex- periment. III B. Effect of Four Spore Levels and Four P Levels in a Marl Soil Mycorrhizal soil inoculum from sorghum pot cultures was mixed with approximately 2 kg of GS soil to give a spore concentration of 366.6 spores/g of soil. Enough GS soil to 3 was seived and fill 48 21.2 cm clay pots, utilizing 3100 cm autoclaved. This pot size was standard in all further ex- periments, to eliminate any effects of small pots on root expansion and P uptake. 96 Four batches of autoclaved soil were mixed in a cement mixer with the inoculum to give calculated spore concentra- tions of 0, 1, B and 9 spores per cm3. Four samples were taken from each of the l, 3 and 9 spore/cm3 treatments and the spores were extracted. Spore counts were numbers com- parable to the desired concentrations. P levels of 0, 50, 100 and 200 kg/ha P were chosen to in- clude a wide range of added P. P, in the form of KH2P04, was added per pot in the amount of 0, 0.102, 0.203 or 0.407 grams of P per pot. The nutrients were added to the soil dissolved in 80 m1 of water. In addition to P, each pot received 0.78 9 K, 0.28 g N, 0.005 9 Cu, 0.005 9 Zn and 0.018 9 Mn. After adding nu- trients, the soil was allowed to dry for approximately 24 hours, and then mixed with a trowel in a large basin to dis- tribute the nutrients evenly. The soil was then returned to the pots. Filtrate from the mycorrhizal inoculum was added to each pot in an amount of water to saturate the soil, and then was allowed to incubate several days. Five Spartan Banner onion seeds were planted in each pot, approximately 2-3 cm deep. The pots were arranged in a randomized com- plete block design, with four replications, on a greenhouse bench under three metal halide lamps set for a 16 hour day- length. The onions were thinned to two plants per pot upon establishment. 97 Seven weeks after planting, plant tops were harvested, dried at 75 C, and weighed. Ten 1 cm root segments from each pot were rated for mycorrhizal infection. Representa- tive soil samples from block one were air-dried and analyzed for soil P. Analysis of variance was performed on the com- bined dry weight of the two plants per pot, and on infection ratings. Analysis of variance was also performed on soil P according to P level. III C. Effect of Fgur Spore Levels and Four P Levels in a Sandy Muck Soil An identical experiment was conducted, except that the soil used was a sandy muck soil from field GO. Seven weeks after planting, plant dry weight was measured. Analysis of variance was performed on the combined dry weight of the two plants per pot. III D. Effects of Four Spore Levels and Four P Levels in a Houghton Muck Soil Soil was obtained from the site of the 1981 low P field experiment at the Michigan State University Muck Farm. Af- ter the soil was seived and autoclaved, soil inoculum of Glomus etunicatus was added and mixed to give spore concen- trations of 0, 0.1, 1.0 and 10.0 spores per cm3. Because relatively small amounts of stock inoculum were added to the sterilized soil, soil for each pot was mixed in a small ro- tary mixer. The P levels chosen were similar to those in the 1980 and 1981 field experiments. Amounts of KH2P04 98 were added, so that pots received 0, 0.093, 0.203 or 0.407 grams of P. These amounts corresponded to 0, 30, 100 and 200 kg/ha P. Each pot received 0.117 grams of K and 0.238 grams of N. The remaining experimental conditions were sim- ilar to those described for the previous two experiments. Plant dry weight was measured seven weeks after planting. Analysis of variance was performed on the combined dry weight of the two plants per pot after a natural logarithm transformation, to insure homogeneity of variance. Results III A - I. Mycorrhizal Benefits in Two Soils with Three P Levels Addition of mycorrhizal inoculum had a significant effect on onion dry weights (Table 11). Soil infestation of inocu- lum increased plant dry weight 82.2%, from 0.135 grams per plant to 0.246 grams. Neither soil nor P had an effect on dry weights. Though the mycorrhizae x time interaction was not significant, mycorrhizal dependency decreased with time in GS soil, though not in GO soil (Table 12). As would be expected, the addition of mycorrhizal inoculum had a significant effect on infection ratings, but no dif- ferences existed among soil-P combinations (Table 12). The P levels did not significantly increase soil P concentra- tions in the two soils (Table 12), and it appeared that very little of the added P remained soluble five weeks after planting. Table 11. 99 mycorrhizal benefits in two soils - I Analyses of variance for the experiment examining Plant Dpy Weight 04 HI Source Block Soil=S P S x P Inoculum=M S x M S x P x M Error(a) Time=T x (A x 368 8 x a mNNHHNNHHwNHHNNHw mmmemvm nocxaexacx nozzevae trxaex on MS 0.016 0.025 0.175 10.949 0.013 0.020 0.103 52.955 0.220 0.332 0.030 0.096 0.128 0.039 0.135 0.159 B ns HS ns ** ns ns ** ns ns ns ns ns ns ns Mycorrhizal Rating Source g; M§ Block 3 - Soil 1 0.095 P 2 0.180 S x P 2 0.127 M 1 147.631 S x M 1 0.096 S x P x M 2x 0.126 Error 9 0.424 Soil P Source _£ gs Block 3 - Soil 1 3405.79 P 2 3.76 S x P 2 0.74 Error 15 8.52 Im ns ns ns ** ns ns W ** ns ns ** - Significant at 1% level x— Since uninfected plants contributed zero sums of squares to the error term, 24 degrees of freedom were subtracted. 100 Table 12. Plant dry weights, mycorrhizal ratings and soil P concentrations of the experiment examining mycorrhizal benefits in two soils - I Time of Mycorrhizalv Plant Dry Mycorrhizalx Harvest Soil Treatment Weight (g) Dependengy Week 5 GS NM 0.044 aY 138.6 Myc 0.105 b GO NM 0.044 a 86.4 Myc 0.082 b Week 7 GS NM 0.209 a 70.3 Myc 0.356 b GO NM 0.207 a 99.5 Myc 0.413 b P Added Mycorrhizal Soil P Soil (kggha) Rating ( cc) GS 0.0 3.09 a2 13.0 a 24.6 3.50 a 12.8 a 49.2 3.68 a 14.2 a GO 0.0 3.63 a 36.3 b 24.6 3.33 a 37.3 b 49.2 3.83 a 37.8 b v - NM = nonmycorrhizal plants, Myc s mycorrhizal plants x - Mycorrhizal dependency = ((weight of Myc-weight of NM)/weight of NM) x 100 y y - Numbers in a pair followed by the same letter are not significantly different, according to LSD test (P=0.05). z - Numbers in a column followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). ~ 101 III A - II. Mycorrhizal Benefits in Two Soils with Three P Levels The effect of mycorrhizal inoculum on plant dry weight was significant in this experiment also, and was different between the two soils (Table 13). Mycorrhizal dependency was higher in the GO soil than in the G5 soil, and decreased with time in both soils (Table 14). Overall, mycorrhizal inoculum increased plant dry weight 32.2%, from 0.292 grams to 0.386 grams. No difference in infection ratings at five weeks after planting could be attributed to either soil or P levels, though infection appeared to be slightly higher in the GS soil (Table 14). The addition of P increased soil P concentrations five weeks after planting in the G0 soil, but not in the GS soil (Table 14). III B. Effect of Four Spore Levels and Four P Levels in a Marl Soil Plant weight increased linearly with spore concentrations, but showed no response to P in the marl soil (Table 15). No increase in infection ratings was evident above a spore lev- el of 3 spores/cm3, though plant weight had increased (Table 15). Significant differences in soil P concentrations ex- isted among P treatments in block 1, though concentrations were quite low (Table 15). 102 Table 13. Analyses of variance for the experiment examining mycorrhizal benefits in two soils - II Plant Dry Weight Mycorrhizal Rating Source g; gs F Source g; as 3 Block 3 - - Block 3 - - Soil=S 1 0.408 ns Soil 1 0.363 ns P 2 0.069 ns P 2 0.034 ns 5 x P 2 0.070 ns S x P 2 0.004 ns Inoculum=M 1 3.103 ** M 1 78.822 ** S x M 1 0.320 * S x M 1 0.363 ns P x M 2 0.021 ns P x M 2 0.068 ns 5 x P x M 2 0.199 ns S x p x M 2 0.007 ns Error(a) 33 0.070 Error 9x 1.957 Time=T 1 50.199 ** S x T 1 0.069 ns Soil P P x T 2 0.123 ns S x P x T 2 0.079 ns Source g; pg F M x T 1 0.254 ns S x M x T 1 0.025 ns Block 3 - - P x M x T 2 0.125 ns Soil 1 2579.23 ** S x P x M x T 2 0.121 ns P 2 31.35 * Error(b) 36 0.070 S x P 2 23.93 ns Error 15 8.40 * - Significant at 5% level ** - Significant at 1% level x - Since the uninfected plants contributed zero sums of squares to the error term, 24 degrees of freedom were subtracted. Table 14. 103 Plant dry weights, mycorrhizal ratings and soil P concentrations of the experiment examining mycorrhizal benefits in two soils - II Time of Mycorrhizalv Plant Dry Mycorrhizalx Harvest Soil Treatment Weight (g) Dependency Week 5 GS NM 0.112 aY 46.4 Myc 0.164 b GO NM 0.086 a 72.1 Myc 0.148 b Week 7 GS NM 0.517 a 11.6 Myc 0.577 a G0 NM 0.413 a 49.9 Myc 0.619 b P Added Mycorrhizal Soil P Soil (k ha) Rating (u cc) cs 0.0 2.88 a2 11.9 a 26.4 2.67 a 12.7 a 52.8 2.67 a 12.6 a GO 0.0 2.46 a 30.5 b 26.4 2.37 31.5 b 52.8 2.34 37.3 b v - NM = nonmycorrhizal plants, Myc = mycorrhizal plants Mycorrhizal dependency = ((weight of Myc-weight of NM)/weight of NM) x 100 Numbers in a pair followed by the same letter are not significantly different, according to the LSD test (P=0.05). Numbers in a column followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). 104 Table 15. Plant dry weights, mycorrhizal ratings and soil P concentrations from the mycorrhizal experiment conducted in marl soil Spores Plant Added Dry Mycorrhizal P Added Soil P (spores/cc) Wt (9) Rating (k ha) (u cc) 0 0.306 ax 0.00 a 0 14.5 a 1 0.423 a 1.98 b 50 17.7 b 3 0.580 ab 2.80 c 100 21.0 c 9 0.828 b 2.75 c 200 25.0 d x - Numbers in a column followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). III C. Effect of Four Spore Levels and Four P Levels in a Sandy Muck Soil The effect of P on onion dry weight was significant, though the effect of spores was not. Dry weights doubled from low to high P (Table 16). However, with zero P added, plant dry weights increased with higher spore levels, and the addition of spores appeared to increase plant weights in treatments given 50 or 100 kg/ha P, though not significantly (Table 16). III B. Effect of Four Spore Levels and Four P Levels in a Houghton Muck Soil The effect of spores, P and the spores x P interaction on plant dry weight were significant in the Houghton muck soil. Plants responded to the 200 kg/ha P by a six-fold growth increase (Table 17). Spore concentrations had a positive effect on onion weights, though not as dramatically as with P (Table 17). Increases in plant weights with increasing spore levels occurred in treatments given zero P. The 105 Table 16. Plant dry weights of the mycorrhizal experiment conducted in sandy muck soil P Added Plant Dry P Added Spores Added Plant Dry (k ha) Weight (g) (k ha) (spores/cc) Weight (g) 0 0.557 ex 0 o 0.243 a 1 0.603 a 1 0.385 ab 2 0.873 ab 3 0.593 ab 3 1.053 b 9 1.008 b 50 0 0.445 a 1 0.598 a 3 0.713 a 9 0.655 a 100 0 0.675 a 1 0.950 a 3 0.938 a 9 0.923 a 200 0 1.160 a 1 1.205 a 3 0.835 a 9 1.015 a x - Numbers in a group followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). 106 Table 17. Plant dry weights of the mycorrhizal experiment conducted in Houghton Muck soil P added Plant Dry P Added Spores Added Plant Dry (k ha) Weight (g) (k ha) (spores/cc) Weight (g) 0 0.225 ex 0 0.0 0.090 a 30 0.485 b 0.1 0.173 ab 100 0.962 c 1.0 0.268 b 200 1.402 c 10.0 0.625 30 0.0 0.456 a 0.1 0.394 a Spores Added Plant Dry 1.0 0.633 a (spores/cc) Weight (g) 10.0 0.633 a 0.0 0.436 a 100 0.0 0.546 a 0.1 0.621 ab 0.1 1.094 ab 1.0 0.650 b 1.0 1.350 b 10.0 0.839 b 10.0 1.060 b 200 0.0 1.613 a 0.1 1.983 a 1.0 1.013 a 10.0 1.195 a x - Numbers in a group followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). 107 addition of 10 spores/cm3 appeared to increase the growth of onions given 30 kg/ha P, though this effect was not signifi- cant (Table 17). Significant increases in plant weight with the addition of spores were also found in treatments given 100 kg/ha P. Plant weight of the treatment given 1.0 spore/ cm3 at this P level was significantly greater than that of the nonmycorrhizal control (Table 17). No effect of spores on plant weight was apparent at the highest P level (Table 17). Discussion The first two experiments showed that mycorrhizae can sig- nificantly increase onion growth in low P soil. Apparently the amounts of P added were not enough to increase the solu- ble P concentrations noticeably, since little of the added P was recovered as soluble P. It was erroneous to expect the added P to be retained as soluble P. In a separate experi- ment, only about 10% of the P added to G5 soil could be ex- tracted as soluble P after two weeks. A higher percentage of added P was extracted from G0 soil, this being 20% of the amount added. Generally, mycorrhizal dependency was greatest when the size of the nonmycorrhizal plants was largest and it decrea- sed with time in 3 out of 4 cases. This may mean that the small nonmycorrhizal plants have a poorer capacity for nu- trient uptake relative to mycorrhizal plants, as compared to 108 the large nonmycorrhizal plants. Infection ratings-of all treatments given inoculum were equally high in each of the first two experiments. Infec- tion ratings were higher in the first experiment than in the second, but these experiments were among the first to employ the 0-4 rating system on a large scale, and the differences in ratings can be attributed to defining the rating scale. Variability in the experiments that were conducted in the large pots was high. The pots were old, and seemed to dif- fer in their capacity to dry out. Consequently, with equal watering, soil in some pots became waterlogged. Waterlog- ging reduces both mycorrhizal infection (Redhead, 1975) and plant growth. Because variability was high, and because the experiments were not repeated, the results should be viewed as preliminary, but some useful information was found. The potential of mycorrhizal substitution for fertilizer P appeared to differ among the three soils studied. This potential may be highest in the marl soils. Much of the P added to these soils becomes unavailable to plants, and my- corrhizae could play a significant role in producing ade- quate plant growth in these soils. Mycorrhizal infection in commercial onions grown in marl soils always appeared high, and soil P concentrations were usually quite low. The plants in the marl soil experiment (III B) did not respond to added P, apparently because little P was retained as plant-available. Plant growth in this experiment increased 109 linearly with increasing spore concentrations. If hyphae present in the soil do not considerably increase mycorrhizal infection in the commercial marl soils, then maximum growth benefits from mycorrhizae may not be possible with observed spore concentrations, unless fungal species other than Q19- mus etunicatus can provide high growth benefits at low spore levels. Mycorrhizal benefits in marl soils may be similar to the reports of Hayman and Mosse (1971) and Mosse (1973) for onions grown in soils with high P fixation rates, in that the benefits are present at all levels of added P (Fig- ure la). Onions grown in sandy muck and Houghton muck soils responded to added P, presumably because much of the added P became available for plant growth. Growth benefits from my- corrhizae were quite large at low P, and seemed to disappear at the highest P levels. More experimentation needs to be done to discover if mycorrhizae can provide growth benefit and maximum growth simultaneously (Figure lb), or only at P levels providing for submaximal plant sizes (Figure lcd). 110 Chapter IV. Mycorrhizal Infection in 17 Onion Cultivars The extent of mycorrhizal colonization and dependence can differ among cultivars of a crop, as Azcon and Ocampo (1981) found among thirteen wheat cultivars, and Menge et al (1978a) found in citrus. The tendency to form mycorrhizae has been linked to host root development, and to the capacity of roots to absorb P. The ability of onion cultivars to form mycor- rhizae should be understood if onions are to be managed for mycorrhizal benefits. Materials and Methods Seventeen yellow globe-type onion cultivars (Table 18) were grown in a field with different soil nutrient levels at the Michigan State University Muck Farml. Half of the field included soil with P and K available at low concentrations, while the other half contained higher levels of P and K. Both sections were seeded on 23 May 1981, with four replica- tions of the seventeen cultivars. On 9 September, each experimental unit was sampled for mycorrhizal infection and soil P, by extracting two adjacent soil cores from a site arbitrarily chosen within a four met- er row length. Cores were taken 4 cm from a plant and 5-15 cm below the soil surface. Soil from the two cores was 1 - Dr. Darryl Warnke of the Crop and Soil Science Depart- ment established and maintained the experiment. Table 18. 111 Mycorrhizal ratings of onion cultivars from 1981 Muck Farm experiment Cultivar Granada Ontario M. Mucker Spartan Sleeper Cima Sentinel Autumn Pride Spartan Banner Harvestmore Downing YG Rocket W-54 Pronto S. XPH 25 8155 x 826 Surecrop Early YG Low Nutrients Mycorrhizal Rating 2.18 2.38 2.38 2.45 2.48 2.68 2.68 2.68 2.70 2.75 2.83 2.90 2.93 2.95 2.95 2.98 3.30 ax ab ab ab ab ab ab ab ab ab ab ab ab ab ab ab b Cultivar Rocket Downing YGY Cima Granada Spartan Sleeper Pronto S. Mucker Harvestmore W-54 XPH 25 Spartan Banner Ontario M. Autumn Pride 8155 x 826 Early YG Surecrop Sentinel High Nutrients Mycorrhizal Rating 1.78 1.88 1.93 1.95 2.03 2.18 2.18 2.18 2.20 2.45 2.45 2.80 2.93 3.05 3.08 3.08 3.15 a ab ab ab abc abcd abcd abcd abcd abcd abcd abcd bcd cd cd cd d j x - Numbers in a column followed by the same letter are not y - YG = Yellow Globe significantly different, according to Duncan's Multiple Range Test (P=0.05). 112 mixed and weighed, with a 5-7 gram subsample dried at 80 C to ascertain dry weight/weight ratios and bulk density. A 1.7 gram sample of air-dried soil was analyzed for soil P. Six root segments from each bulked sample were stained and rated for mycorrhizal infection. Analysis of variance was performed on the infection ratings. Results The low nutrient soil had a measured soil P concentration of 4.31 1 0.22 ug/cm3 (mean 1 standard error), while soil P measured 9.01 1 0.40 ug/cm3 in the higher nutrient soil. Among cultivars, differences in infection ratings were found in the low and high nutrient treatments (Table 18). Signi- ficant differences in infection ratings of onions grown in the low nutrient soil were found only between the two culti- vars with the lowest and the highest mycorrhizal ratings (Table 18). However, a greater range of differences existed at the higher nutrient level (Table 18). Cultivars with high ratings tended to maintain that distinction, when grown in either nutrient level. Discussion Because of different capacities of onion cultivars to become mycorrhizal, potential exists for the manipulation of cultivars, in order to achieve different degrees of infec- tion. If onion mycorrhizae respond to plant P in the same 113 manner as wheat and citrus mycorrhizae, then all the onion cultivars studied were at a sufficiently low plant P concen- tration to become highly infected when grown in soil of 4 ug/cm3 P. In a slightly higher soil P concentration of 9 3, the onion cultivars showed more detectable differ- ng/cm ences in root colonization by mycorrhizal fungi. Cultivars with high infection ratings probably had a lower capacity for P uptake than cultivars with low mycorrhizal ratings. If soils are to be managed for high mycorrhizal benefits, then plants with a capacity for early infection and also a high response to infection, probably should be utilized. Cultivars with the highest mycorrhizal ratings might be most useful. However, if mycorrhizae only provide growth benefit in soil P concentrations providing submaximal plant growth, then the degree to which a plant is mycorrhizal may predict how P deficient the soil is. Cultivars with a high capacity to absorb P and low capabilities for mycorrhizal formation, could be grown with much better results. If mycorrhizae were shown to feasibly control a disease or a nematode infestation (Schenck and Kellam, 1978), but not necessarily provide growth benefits and maximal sizes simul- taneously (Figure 1b), then perhaps a cultivar with the ability to become mycorrhizal and still retain high P ab- sorption, might be chosen. Therefore, a moderately infected cultivar might well be selected. Since inherent differences exist in the degree to which cultivars form mycorrhizae, 114 promise for breeding onions for this trait also exists. The shape of infection-P response curves may differ among cultivars, as Menge et al (1978a) found for citrus. The concept of a soil P threshold common to all cultivars may not hold true. 115 Chapter V. Response of Mycorrhizae to P Under Two Watering Regimes in the Field The report of Nelsen and Safir (1982b) showed that mycor- rhizal onions were more drought tolerant than nonmycorrhizal onions. A field experiment was designed to test the re- sponse of mycorrhizal infection and plant growth to soil P, under well-watered and water-stressed conditions. Materials and Methods The experiment was established at the Michigan State Uni- versity Muck Farm1 , with one half of the field plot reserved for irrigated treatments, and one half for nonirrigated treatments. Within an irrigation level, there were three replications of each of four treatments. The treatments were zero P added, 60 lb/A P205 added, 60 lb/A P205 added along with Glomus etunicatus soil inoculum banded under the row at a rate of approximately 2500 spores per meter of row, and 200 lb/A P205. P was also banded under the rows. Spar- tan Banner onions were seeded on 21 May 1981, with each treatment row bordered by two guard rows on either side. Tensiometers were placed at 3-5 inch and 8-10 inch depths, and monitored throughout the season. Irrigated plots re- ceived water whenever tensiometer readings reached 15 centi- bars. Irrigated onions received a total of 5.1 inches of l - The experiment was established and maintained by Dr. Darryl Warnke, of the Crop and Soil Science Department. 116 water from four irrigations. This was in addition to 10.94 inches of rain received from 15 June to 18 September, the period during which the soil was dry. On 8 September, soil core samples were taken from each treated row. Three adjacent soil cores, 4 cm from a plant and 5-15 cm below the soil surface, were taken from the mid- dle of the rows, and 1-2 meters from each end. Soil from each site was mixed and weighed. A 5-7 gram subsample was dried at 80 C to gain dry weight/wet weight ratios and bulk density. Soil P analysis was conducted on 1.7 grams of air-dried soil. Six roots from each bulked soil sample were stained, and 1 cm segments were rated for mycorrhizal infec- tion. The mean rating of the three sites per row was used as the infection rating per experimental unit. The final fresh weight of bulbs per row was measuredz, and expressed as hundredweight per acre (cwt/A). Analysis of variance was performed on the infection and yield data. Results The level of irrigation had a significant effect on mycor- rhizal infection ratings, as did the P-inoculum treatments (Table 19). The P-inoculum treatments also had a signifi- cant effect on yield, but irrigation levels did not (Table 19). 2 - Measured by Dr. Warnke. 117 Table 19. Analyses of variance of the 1981 irrigation-P . experiment. Mycorrhizal Rating Yield Source g1 gs E Mg F Irrigation l 10.14 ** 9480.4 ns Error (a) 4 0.31 3994.3 P-inoculum 3 2.34 ** 23729.2 ** Error (b) 12 0.33 2629.1 ** - significant at 1% level Irrigation decreased mycorrhizal infection ratings from a mean of 2.35 to 1.05. Within irrigated treatments, no sig- nificant differences were found (Table 20). But addition of P significantly decreased infection ratings in the nonirri- gated treatments (Table 20). The addition of mycorrhizal inoculum did not appear to affect infection ratings. High mycorrhizal ratings of onions appeared to be influenced by both soil P concentrations and by irrigation. High infec- tion ratings occurred in irrigated onions only when the soil P concentrations were below approximately 15 pg/cm3 (Figure 12). However, high infection ratings were found in the non- irrigated onions at soil P concentrations up to 30 pg/cm3 (Figure 12). Onion yields did not respond to irrigation. A high water table did occur at the Muck Farm, and may have reached the deeper roots, thus diminishing the effect of irrigation on onion growth. Yields did increase with increased P Table 20. 118 Mycorrhizal ratings and yields from Muck Farm irrigation-P experiment P-inoculum Treatment Mycorrhizal Ratingx Yield (cwt/A) NIRRx IRRx NIRRx IRRx o lb/A 2.49 aY 2.12 a 399.3 a 335.0 a P O 2 5 60 lb/A 2.80 a 1.09 b 482.3 ab 470.7 b P O 2 5 60 lb/A P O + 2.54 a 0.91 b 428.3 a 406.3 ab ingcglum 200 1b/A 1.59 a 0.09 b 539.3 b 478.3 b P205 x - NIRR = nonirrigated, IRR = irrigated y - Numbers in a column followed by the same letter are not significantly different, according to Duncan's Multiple Range Test (P=0.05). Figure 12. 119 ‘3’ A g I 92 I O n I 2 o o z: ' . C .l a: g .o' Eiql I N N I H 0 g I a: 0 q 8 so. 0 ,_ o =:9. I I 0 O I .d o 00 o 9- °o 15 so 45 35 76 SOIL Pnosrnoaus (pg/an”) O 4 B I o o o o ' csg' ' "l 2 o oo ' : 0 0 O I m 0 o I a: o o o no ' é 9- 000 o 0' 5°“ . I E o ‘ a I o u I :59. I I ' T ’ I 9 I j 1 O I 15 'I 30 I 45 'l 50 I 75 sou. PHOSPHORUS (In/cm“) Mycorrhizal rating of onions versus measured soil P concentration (pg/cc) of the 1981 Muck Farm irrigation-P experiment. Data is from from irrigated treatments (A) and nonirrigated treat- ments (B). Soil P thresholds are estimated by dashed lines. 120 application (Table 20). The addition of spores decreased yields to a similar extent in both irrigated and nonirri- gated treatments, by approximately 12%. However, this ef- fect was not significant. Discussion The shift in the response of mycorrhizal infection to P has not been previously reported. Reid and Bowen (1979) found that lower soil water content decreased mycorrhizal infection, but this experiment was short-term. Their results may have been due to direct differences of soil moisture on the colonization process, and did not involve the effect of soil water on diffusion of P. Also, their experiment was only conducted at one P level. Over a longer period of time, the effect of soil water on diffusion and uptake of P may override the more direct effect of soil water on infection. Thus, decreasing soil moisture may in- crease mycorrhizal infection because plant P levels may fall from the decreased P diffusion rates. Verification of yield responses to P under different water regimes (Figure 2) was not found, because of the lack of plant response to irrigation, and the lack of nonmycorrhizal controls. The magnitude in the shift of soil P thresholds limiting mycorrhizal infection may point out the difficulty in the management of soils for mycorrhizal benefits. If onions are 121 to be maintained within a narrow soil P range to provide for mycorrhizal growth benefits and maximal plant growth simul- taneously (Figure lb), then the effect of water availability on mycorrhizal benefits must be understood. The soil P con- centrations allowing high infection and maximal yields will differ depending upon the diffusion of P within the soil. This soil P range will shift, depending upon the frequency and the length of water stress. If soils are managed within this range, on the premise of an abundant water supply, then the occurrence of drought might give decrease yields, if soil P diffusion is sufficiently slowed. If enough P is added to provide high yields in times of water stress, my- corrhizal benefits might be precluded when soil P is above the limiting concentration. The high soil P concentrations in the commercial Grant fields may safeguard yields against the probability of water stress. More research is needed to ascertain whether the observed soil P concentrations are economically optimal for the irrigation practices used in those fields. Since an extremely wide range of soil P concentrations was observed in commercially used soils, potential probably ex- ists for more careful management of P. The key to manage- ment might be the frequent analysis of soils with the assis- tance of the Michigan State University Soils Laboratory. Large errors in the calculation of soil P concentrations can occur if soil densities are not taken into account. Bulk 5.: (”J N densities of all fields fell within a range of 0.19-1.03. The same ppm values from soils of the two extreme bulk den- sities would have at least 5-fold differences in soil P concentrations on a volume basis. Measurements of the soil bulk density can be easily made for provision of accurate soil P analysis. 123 Chapter VI. Onion Plant Model One goal of an onion pest management study was the forma- tion of an onion growth model. The effects of the weather and cultural factors most influencing onion growth would be included in the model. The weather factors identified as having the greatest effect on onion growth were temperature, rainfall, relative humidity, light intensity and photoper- iod. Cultural factors included onion cultivar, fertiliza- tion, cultivation, irrigation, plant density and plant ar- rangement. From these variables, sizes and weights of onion tops, bulbs and roots would be predicted. . Data Collection - 1980 Onions were collected from field plots near Grant, Mi., during the 1980 growing season, for the purpose of measuring onion growth. The plots were located in field 61 (Appendix 3). Four double-rows of onion cultivar 192 were seeded on 7 May 1980. Seventy-two two-foot plots were staked out on 10 June 1980 in each of two areas of the field. The areas dif- fered as to soil type, with one soil a sandy muck, and the other an organic muck soil. On 12 June, 25 June, 8 July, 23 July and 15 August, six plots were randomly selected from each area. A plant near the center of the plot was selected as the sample plant. This plant was collected, and leaf lengths measured and recorded, so that leaf surface area could be estimated. If the leaf had senesced partially or 124 completely, the green leaf length and total leaf length were measured and recorded so as to estimate green and senescent leaf surface of those leaves. On 25 June and 8 July, the plants on either side of the sample plant were collected, and leaf lengths were measured and recorded. Dry weights of leaves and bulbs, as well as bulb lengths and diameters, were measured and recorded. The plants were dried at 70 C, until no weight change was found. The upper end of the onion bulb was recognized as the point between bulb and sheath at which an oval form would be complete. Weights of leaves were distinguished from weights of leaf sheaths. Soil cores were taken directly below and at several dis- tances from the sample plant. The distances from sample plants at which the cores were extracted were 4 cm on all dates, 8 cm on 8 July and 23 July, 12 cm on 23 July and 15 August and 20 cm on 15 August. The distances were at right angles from the onion row and extended within an 86 cm space between double rows. Cores were 1.94 cm in diameter, and 15-20 cm long. Cores were divided into 5 cm segments, so that 3-4 segments were extracted from each core. From each segment, all root fragments were extracted, and measured for length. Shortly before the 15 August date, the leaves of onions in all plots became diseased with downy mildew. As a result, the leaf lengths were impossible to measure, because of . 125 brittleness. The lengths of the leaf sheaths were measured and recorded, in order to estimate leaf surface area. On this date, 21 plots not previously sampled were selected, and all bulbs within the center foot were collected. Bulb lengths and widths, and bulb dry weights were measured and recorded. Daily maximum and minimum air temperatures were obtained for the sample sitel. Analysis of variance was performed on plant weights from each sampling date to compare growth ef- fects of the two areas within the field. Data Collection - 1981 During the 1981 growing season, data was collected on onions grown in the experiment conducted in the low P field at the Michigan State University Muck Farm (Chapter II B). Plant dry weights of tops and bulbs were recorded from all treatments, as previously described, at 3, 6, 9, 12, 15 and 21 weeks after planting. Onions in treatments given 0 and 193 kg/ha P plus Glomus etunicatus inoculum were measured more extensively. Leaf lengths were measured, as in 1980, on all plants collected from the two selected treatments at 3 weeks after planting. On all subsequent dates, leaf lengths were measured on ten arbitrarily selected plants from each sampled subplot. Leaves, leaf sheaths, and bulbs of the chosen plants were 1 - The temperatures were received from Gary Whitfield of the Entomology Department. 126 dried at 70 C, and weighed. By 21 weeks after planting, leaves were almost totally dead from a combination of senescence and foliar diseases. Leaf surface area was not estimated from these plants. The bulbs were topped at this harvest, and fresh and dry weights were recorded. Since the 1981 data was more extensive than the 1980 data, it was selected for further analysis. The onions given 193 kg/ha P apparently acheived better nutrition than the onions given zero P, so analysis was concentrated on data collected from these plants. Since growth rates were to be estimated from total weights of tops and bulbs, the percentage of dry weights apportioned into bulbs and tops was compared to total dry weights. Bulb weight, as a percentage of total weight, increased until it appeared that a constant percentage was acheived. A nonlin- x, was chosen to fit the observed ear function, Y = a + be.c data, with Y equal to In (bulb dry weight/total weight of top and bulb), x equal to the total weight of top and bulb, and a, b and c representing constants. The natural loga- rithm of the dependent variable was taken to insure homoge- neity of variance. The regression model was fitted by find- ing the minimum sums of squares associated with selected c values through repetitive linear regressions of Y a a + bz, where 2 was equivalent to e cx. 127 The resulting equation (Table 21) fit the data well (Fig- ure 13). Bulb weight, as a percentage of the weight top and bulb, was approximately 5% three weeks after planting. This percentage increased until a relatively constant percentage of 47% was reached. The same regression model was fit to leaf data in a simi- lar manner (Table 21, Figure 14), with the dependent varia- ble being ln(1eaf dry weight/total weight of top and bulb) and the x variable being the total weight of top and bulb. Initially, leaf weight, as a percentage of top and bulb, was approximately 74%. This figure declined to about 34%, when the total weight of top and bulb reached 23.2 grams, the maximum weight observed. The variability was quite high, though the coefficients of determination (R2) (Table 21) can be misleading, since they are statistics of the regressions of Y against e-cx. But these equations accurately predicted trends the ln (bulb weight) and the ln (leaf weight) from the weight of top and bulbs (Figure 15,16) (data-Appendix 6). Table 21. Regressions predicting bulb and leaf weights as a percentage of top and bulb weight Number Equation 33 of cases X Range Y1=-0.7520-2.0383*e-o’2425*x 0.472 190 0.0063-23.20 yz=-1.0961+0.7914*e’°°“29*" 0.548 190‘ ’ 0.0063-23.20 Y1 refers to bulb data, and Y2 refers to leaf data. 128 In (BULB HEIGHT/TOTHL REIGN?) O D d 1 94 ° 0 O o g o o 4 . . . on an " . : “"" - o 7" " ' \ o 4 . C , o a O N I O O D I O I 7' ‘V'VrYV'VYVV'Vrjj’TT’YTt O 5 IO 15 20 2" PLRNT DRY HEIGHT (g) Figure 13. 1n (bulb dry weight (g)/total dry weight (g) of top and bulb) versus the total dry weight (g) of top and bulb with fitted regression line. I In (LEHF HEIGHT/TOTHL HEIOHU Figure 14. PLHNT DRY H510!" (g) 6 0 fl- ' o I O ‘ I e ' ° 7‘ ° 0 9 o 4 O O f-Fm—H...Vw.rv‘v..vjv... I 10 I! 20 :5 ln (leaf dry weight (g)/total dry weight (g) of top and bulb) versus the total dry weight (g) of top and bulb with fitted regression line. 129 In (BULB HEIGHT) O. O I O. O Tjtfifiv‘vwvv'vvvwifvftTWV'j 0 5 10 15 20 25 PLHNT DRY HEIGHT (g) D 4 q B N— In (BULB HEIGHT) r V V V T 1 1 fi I T .0 0.5 11.0w ‘ ‘ 11:5j ' ' 2.0 PLRNT DRY WEIGHT (9) Figure 15. In (bulb dry weight (9)) versus the total dry weight (g) of top and bulb predicted from re- gression (solid line), and observed (circles). Plant dry weights are in a range of 0.006-23.2 grams (A) and 0.006-2.0 grams (8). 130 I. ILEHF HEIGHT) “200 -4-0 I I V V T Y I V V W j' 1 V Y 1 T V U V Y I V T o 5 10 1 15 20 25 PLHNT DRY HEIGHT (g) 2.0 l ’200 1 In (LEHF HEIGHT) -4-0 01.5' 1110' v ' '11.5‘r ' 1 ‘2.0 PLRNT DRY HEIGHT (9) -800 .J T j T I O o O 4 Figure 16. ln (leaf dry weight (9)) versus the total dry weight (g) of top and bulb predicted from re- gression (solid line), and observed (circles). Plant dry weights are in a range of 0.006-23.2 grams (A) and 0.006-2.0 grams (B). 131 Bulb weight measured in 1980 followed a similar relation- ship to total weight of top and bulb, as the onions collect- ed in 1981. However, bulb weight relative to total weight in the 1980 onions may have been slightly higher in the 1-4 gram total weight range than observed in the 1981 onions (Figure 17) (data-Appendix 6). Leaf weight, relative to the total weight of top and bulb, was higher in 1980 than in 1981 (Figure 18). Whether this resulted from different environmental conditions, or differ- ent inherent characteristics of the two cultivars studied is not known. The assumption of one equation describing leaf weight as a percent of total weight may not hold for all conditions. Separate relationships may need to be developed for different cultivars. Measurement of Leaf Surface Area In order to estimate photosynthetic capacity of onions per unit of leaf tissue, leaf surface area was measured on on- ions collected from the field in 1980. Since the process was long and tedious, regression equations estimating leaf surface area from easily measured leaf parameters were de- veloped, so that leaf surface area could be assessed quickly and accurately. The most precise method employed to measure leaf surface area was a paper density (Nelsen and Safir, 1982a), in which the area of flat onion leaves was deter- mined by dividing the weight of photocopied paper onions by Figure 17. Figure 18. 132 4-0 l In (BULB HEIGHT) -240 -I-G °. - I I I I Thri 2I 4 o I 10 12 TOTRL PLHHT HEIGHT (a) 1n (bulb dry weight (9)) versus the total dry weight (9) of top and bulb observed in 1980 (circles), and predicted from regression (solid line . 9 V Q 0d N .. '0 0 ' '5 .' ~°ou f “JO 3 IL E: .14. all .5 ‘3 Q I °. ? I I I I I r 0 2 4 B B 10 I” TOTRL PLHNT HEIGHT (9) ln (leaf dry weight (9)) versus the total dry weight (9) of top and bulb observed in 1980 (circles), and predicted from regression (solid line). 133 the density of the paper. This tended to be much more accu- rate than the measurement of leaf area with a leaf area me- ter (LI-COR Portable Leaf Area Meter, Model LI-3000, LAMBDA Instruments Corp.). This instrument worked well on flat leaves, but not on the cylindrical onion leaves, which tend- ed to retain their shape, even if cut lengthwise. The paper density method was used on 1980 onion leaves including a wide range of leaf surface areas, so that the surface area estimation methods could become widely applicable. Leaf lengths and diameter measurements were made on the same leaves for which surface area was measured. An attempt was made to predict leaf areas utilizing geometric formulas. Leaves of onions can be visualized as circular cones, or combinations of a circular cone and frustrums of circular cones. Measurements of leaf length and of at least one di- ameter are necessary for surface area estimation by this method. Nelsen and Safir (1982a) predicted surface area from leaf lengths and diameters measured at leaf midlength and also at 10% of the length from the leaf tip. Predic- tions based on leaf geometry correlated well with leaf sur- face areas measured by the paper density method. Another method previously used to calculate leaf surface area, was to estimate area from regression equations (Lampert, 1980), with the independent variables being leaf length and width. Both methods were applied to the 1980 data. High correla- tions were found between surface measured by the paper 134 density method and surface area estimated through geometry. Equally high correlations, with coefficients of determina- tion (R2) of 0.98, were found between leaf surface area mea- sured by the paper density method, and combinations of leaf length and diameters. Estimation of leaf area based on leaf length (Table 22) was the method chosen to estimate the leaf surface areas of field-grown onions, since it eliminated the problems involved in the measurement of leaf diameters, without sacrificing much accuracy. Measurements of diame- ters increased measuring time greatly, and had to be accom- plished soon after sampling, for leaves would lose their shape with time. Table 22. Surface area regressions 2 Number Equation 3_ of cases X-range (1) Y=2.250+0.04360*GLL2 0.95 85 1.7-5.25 (2) Y=2.269+0.04360*GLL*TLL 0.90 25 5.8-4l.4 (GLL) 8.5-42.1 (TLL) (3) Y=12.754+8.200*SLL 0.83 94 1.3-127.1 Y = surface area (cmz), GLL = green leaf length (cm), TLL = total leaf length (cm) and SSL = sum of sheath lengths (cm). Equation (1) was also utilized in the calculation of se- nescent leaf surface areas. Since senescence began at the leaf tips, the green leaf lengths and total leaf lengths could be measured. A second regression equation (2) (Table 22) was used to measure green surface area when leaves had 135 senesced. If the total leaf length was used as the indepen- dent variable of equation (1), then senescent leaf area could be estimated by the difference, (1) minus (2). If the leaf was totally dead, the senescent surface area was calculated solely by (1). By the final 1980 sampling, the leaves were dead due to foliar diseases. They were brittle, and lengths were impos- sible to measure. Since the length leaf sheaths above the bulb had been recorded for a number of previously collected onions, equation (3) (Table 22) was developed to estimate leaf surface area from the sum of sheath lengths. Equations (1) and (2) were applied to leaves collected during 1980 and 1981. Green leaf surface area (GSA) and senescent leaf sur- face area (SSA) were calculated. Since sampling was more extensive in 1981, this data was selected for further analy- sis. Relationships between GSA and the total leaf surface area (TSA) were investigated, as was the relationship be- tween TSA and leaf weight (LWT). GSA was linearly related to TSA, though the residual pat- terns of regressions showed that one equation did not fit well to all of the data. Senescence had not occurred on any leaves collected 3 weeks after planting. By 6 weeks after planting, cotyledons had largely senesced, and represented a fairly large percentage of the total surface area. After decay of the cotyledons, GSA remained a constant percentage of TSA. Rapid leaf senescence related to plant maturity 136 occurred after the fifteenth week after planting, and was not measured. Evidently, onion leaves senesced rapidly after collapse of the neck of the plant, and there was no evidence of a gradual increase in the rate of senescence throughout the length of the growing season. Piecewise linear regression was performed on overlapping data ranges to ascertain GSA-TSA relationships (Table 23). 2 When TSA was below 15.1 cm , equation (4) resulted. Equa- tions (5) and (6) resulted from regressions of GSA against 2 2 TSA, in TSA ranges of 25-180 cm and 108-1940 cm , respec- tively (data-Appendix 6). Table 23. Regressions describing the green leaf surface area to total leaf surface relationship . 2 Number Equation 3_ of cases TSA Range (4) GSA=-8.546+0.918*TSA 0.988 84 108-1940 (5) GSA: 4.244+0.939*TSA 0.962 38 25-108 (6) GSA=TSA 1.00 70 2.7-15.1 Equations (4) and (5) intersected when TSA equaled 26.4, and equations (5) and (6) intersected when TSA equaled 161.5. Below a TSA of 26.4 cm2 , equation (4) was assumed to apply, and senescence was calculated at zero. If TSA was between 26.4 and 161.5, equation (5) was assumed to apply, and senescence averaged approximately 12% of TSA. With TSA values greater than 161.5, equation (6) was used, with an 137 average senescence of approximately 9% of TSA (Figure 19). The relationship of total leaf surface (TSA) to leaf weight (LWT) was investigated with the aim of predicting TSA from 2 of known LWT. When plants were small, approximately 340 cm surface existed per gram of leaf tissue. This ratio fell to about 190 cmz/g for the largest observed plant weights. TSA was nonlinearly related to LWT, with the TSA/LWT ratio de- creasing sharply when the LWT increased, and then leveling off (Figure 20). The regression model, Y = a + be-Cx, was utilized for the analysis, where Y was equal to the natural logarithm of TSA/LWT, and x was equal to LWT. The natural logarithm transformation was made to insure homogeneity of variance. The best fitting equation was found by selecting c values, and finding the minimum error sums of squares for regression from repetitive linear regressions. One equation poorly fit the entire data set, as judged by residual patterns. The regression model was subsequently fit to two overlapping data subsets (Table 24). The equa- tions intersected at two Points. Equation (7) was consid- ered to apply when LWT was less than 0.404, one of the in- tersection points. When LWT was greater than 1.164, the other point of intersection, equation (8) was assumed to apply. An average of the two equations estimated 1n (TSA- LWT) values between the two points of intersection (Figure 20). Predicted ln (TSA) values agreed well with observed points (Figure 21) (data-Appendix 6). 138 .111 > 1 n 11.1.: I GREEN PLHNT SURFRCE HRER (cmz) 200 400 800 000 1000 1200 1400 1500 1000 2000 I A ' I ‘ l I I ' I I I ' 7 1 Tfi T I I r 200 400 800 800 1000 1200 1400 1800 1800 2000 TGTGL PLFINT SURFHCE GREG (cmz) up 300 4 w 250 L L o 200 l o GREEN PLGNT SURFRCE GREG (m2) 160 0 ' 50 v 100 ' 150 200 I 250 1 300 TGTGI. PLGNT SURFRCE GREG (m2) f Figure 19. Green leaf surface area (sq. cm) versus the total leaf surface area (sq. cm) with fitted linear regressions lines. The total leaf sur- face areas are in a range of 2-1940 square cm (A) and 2-300 square cm (B). 1n (LERF SURFRCE RRERI o 1.0 2.0 3.0 4.0 5.0 0.0 1.0 0.0 0.0 III.0 Figure 20. 139 1.0 I In (SURFBCE HREH/LEBF HEIGHT) 5.0 .. 1 f1 v I v T v 1 v I v I v ' v ' v. 0.0 140 240 340 440 540 8.0 740 .40 9.0 1040 LEHF DRY HEIGHT (9) ln (total leaf surface area (sq. cm) leaf dry weight (9)) versus the leaf dry weight (9) with fitted regression lines. h (LERF BURFRCE RREM 0.0 1.0 2.0 3.0 4.0 5.0 0.0 1.0 0.0 0.0 10.0 I I I I I 40 10° 20° 30° ‘00 LEGF DRY Figure 21. 5:0 010 710 0:0 410 (".0 0:0 r oizfi o'.4 0:4 014 1.0 HEIGHT 00 LERF DRY HEIGHT.09 1n (total leaf surface (sq. cm) versus the leaf dry weight predicted from regression (solid line), and observed (circles). Leaf weight data is in a range of 0.005-9.58 grams (A) and 0.005-1.0 grams (B). Figure 22. Figure 23. 140 In (LERF SURFRCE RRER) 0.0 110 210 31.0 410 5.0 LEHF HEIGHT (9) ln (total leaf surface area (sq. cm)) versus the leaf dry weight (9) observed in 1980 (circles), and predicted from regression (solid line). .00 m I 1 4M 1 L GREEN SURFBCE RRER (m2) 200 1 o ' £51174bgfi db ' db - mm TGTGI. GUREGcE GREG («13) Green leaf surface area (sq. cm) versus the total leaf surface area (sq. cm) observed in 1980 (circles), and predicted from regression (solid line). 141 Table 24. Regressions describing the leaf surface area to leaf weight relationship 2 Number Equation 3_ of cases LWT Range (7) Y=5.834+1.208*e'cx 0.821 93 0.004851.99 (8) Y=5.231+0.925*e-cx 0.548 77 0.3100-9.58 Y = ln (TSA/LWT) and x = LWT. TSA/LWT ratios were higher in 1981 than in 1980 (Figure 22), although the leaf weight to total weight ratio was low- er in 1981. This implies that the 1980 onion leaves were thicker and exhibited less surface area per unit weight. Also, the green leaf surface area (GSA) reached a slightly higher percentage of the total leaf surface area (TSA) in 1980 than in 1981 (Figure 23). Senescent leaf area, rela- tive to total leaf surface, was thus lower in 1980 than in 1981. Root Lengths The analysis of variance showed that plants weights did not differ between the two areas sampled in 1980, so data was combined. Multiple regression analysis was conducted in order to relate root length per volume of soil to plant sizes and soil depths. The root data was divided into sub- .sets, according to distances cores were taken from sample plants. Depth of a core segment was determined by the aver- age depth, so that a core segment taken 5-10 cm below the 142 soil surface had an average depth of 7.5 cm. The significance of regressions was greatly increased if some measure of the size of the two neighboring plants was in- cluded as an independent variable. Only the data collected on 25 June and 8 July with distances of 0 and 8 cm from the sample plant, were included for further analysis. At these dates, onion root systems of the neighboring plants evident- ly had an effect on the root systems of the sample plant. The average plant spacings within the row were 6.5 cm and 9.1 cm for the 25 June and 8 July dates, respectively. Weaver and Bruner (1927) and Drinkwater and Janes (1955) found the root system of onions to extend approximately 30 cm laterally from the bulb, and up to 60-70 cm deep at maxi- mum root growth, though the majority of onion roots were lo- cated in the top 20 cm within a 15 cm lateral distance of the bulb. Minimal branching and a lack of root hairs were noted for onion roots. The roots senesced quickly after plant maturity. 3 of soil. Root lengths ranged from 0 to 3.13 cm per cm Higher correlations were found when the size of neighboring plants was weighted by the distance between the sample plant and its neighbors, and when depths were transformed by their inverses, than when depths were not not transformed (data- Appendix 6). Fairly high coefficients of determination were found upon inclusion of a large number of variables (Table 25). 143 However, because variablility in plant sizes was high, and since the most significant regressions were found with data collected at two sampling dates, this study should be re- garded as preliminary. Further root sampling should proba- bly be conducted on plants without nearby neighbors, so that individual root systems can be modeled before advancing to the sampling of onions within a row. Table 25. Root length regressions 2 Number of cases 21 Distance Equation‘ 0 cm Y = 0.00679 2 -0.691*WSA /DEP lO.94*(l/DEP) 0.173*WSA/DEP 0.77 85 + + 0.336 0.69 48 0.0171*SA1 0.00157*SA2 0.00003984* 2.083*WSA/DEP 0.450*WSA/DEP I+I+III 2 Y = root length (cm) per soil volume (cm3), DEP = depth (cm) below the soil surface from 2.5-17.5, WSA a weighted surface = (surface area of sample plant + 1/2 * the surface area of the two neighboring plants) / distance (cm) between the sample plant and the two neighbors, SAl a surface area of the closest neighbor and SA2 = surface area of the 2nd closest neighbor. ' All regressions were conducted by a stepwise deletion technique. The surface area of neighboring plants at the 8 cm distance, was significant as a variable in the regres- sion, though not at the 0 cm distance (Table 25). Interest- ingly, root length per volume was negatively related to the 144 surface area of the closest neighbor (Table 25). This may be due to a density effect, and calculation of root lengths should probably take into account competition, as well as overlapping root systems. The Effect of Plant Density on Onion Yield Calculations of yield per area can be attained by summing the weights of individual plants within that area. However, the effect of plant size due to the number and spacing of plants must be accounted for. This effect is primarily one of competition for limited resources. Frappell (1979) men- tioned water, nutrients and light as the resources most com- peted for in plant communities. Generally, plants at low densities grow larger than plants at higher densities due to the lack of competition, though cooperative effects of neighboring plants of some crops can cause greater plant growth at slightly higher densities (Gillis, 1979). Individual plant weights usually decline with increasing plant density, though total yield may be in- creasing. Eventually, total yield per area will level off (asymptotic growth), or decline (parabolic growth) with in- creasing density. One pattern of growth is usually associ- ated with growth of a particular crop, though in some crops, yield may be either asymptotic or parabolic with increasing density, depending upon environmental conditions (Faraz- daghi, 1968). 145 The effect of plant density, or the number of plants per unit area, on yield, can be distinguished from the effect of plant arrangement, or the distribution of plants within the area. For row crops, plant arrangement can be measured by the rectangularity, or the ratio of inter-row distances to intra-row distances between plants. Plants growing at rec- tangularities of 1:1 tend to grow larger than plants growing at higher rectangularities, though the densities remain the same (Frappell, 1979). The decrease in yield with higher rectangularities is presumably due to earlier competition. Onions were grown at 16 inch row spacings in Michigan, un- til the late 1950's and early 1960's (Lucas, 1970). At this time, growers initiated use of 36 inch row spacings, or a combination of one wide row spacing of about 32 inches, and one narrow spacing of 4 to 8 inches (Lucas, 1970). This double-row practice could be expected to decrease yields because of higher rectangularities, but was adopted for the conveniance of late-season cultivation, and for a savings from a lower number of pesticide bands necessary for com- plete foliage coverage. There is little information availa- ble to judge the effect of this change in plant arrangement on yields. Willey and Heath (1969) reviewed quantitative functions relating yield to density. A group of equations, termed reciprocal equations, relating the inverse of mean weight per plant to density, seemed to offer the best methods in 146 relating yield to density. Shinozaki and Kira (1956) proposed the first reciprocal equation, of the form: l/w = a + bp (9) where w = the mean weight per plant, p = plant density and a and b are constants. This equation fit data of crops show- ing asymptotic relationships of yield to density, in which the reciprocal of w is linear with density. However, this equation gave poor fits to parabolic yield-density relationships. Holliday (1960a) pointed out that density-independent plant growth at low densities should not be included in use of equation (9). He suggested a modified form of (9): l/w = a + bm (10) where m = p-n, and n included the range of densities of densities at which density independent plant growth oc- curred. Holliday (1960b) proposed another equation to fit yield- density data. This was a quadratic equation: l/w = a + bp + cp2 _ (11) with a, b and c as constants. This equation added flexibil- ity in fitting regression lines to both asymptotic and para- bolic yield-density relationships. 147 Bleasdale (1966b) suggested use of a different equation to fit both asymptotic and parabolic yield-density relationships: 8 1/w = a + bp (12) where 8, a and b are constants. This equation is a simpli- fied form of an equation used by Bleasdale and Nelder (1960). When 8 = 1.0, an asymptotic relationship is implied, and parabolic relationships are described when 8 is found to be less than 1.0. Frappell (1973), in a study of the effect of onion spacing on yield, used equation (12) to describe yield-density data (Table 26). The value of 8 did not differ from 1.0, thus describing an asymptotic relationship. Rectangularities of 8:1 depressed yield approximately 10%, compared to a square arrangement. Frequency distributions of bulb sizes were positively skewed, with a greater number of small onions than larger bulbs. Rogers (1977) used this type of data to construct a nomogram relating yields of varying sizes to density. The biological significance of the constants fa‘ and 'b' from equation (12), has been discussed (Frappell, 1973, and Willey and Heath, 1969). As growth tends toward density- independence at low plant densities, the value of 'w' ap- proaches l/a. This value has been considered to be the genetic potential of the crop in a particular environment. 148 Table 26. Regression equations describing yield-density data -4 -4 2 Data Set a x 10 b x 10 11b 3_ Frappell 67/68 season 18.97 1.37 7299 - Frappell 68/69 season 18.97 1.43 6994 - Frappell 69/70 season 18.97 1.10 9091 - Lucas (cv. Spartan 25.75 1.29 7752 0.99 Banner) Lucas (cv. Downing 40.23 1.15 8696 0.99 Yellow Globe) 1980 dry wt.* 365.0 10.44 958 0.66 1981 high P field, 29.59 0.624 16026 0.72 week 21, fresh, wt. 1981 low P field, 38.47 0.630 15873 0.66 week 21, fresh wt. 1981 low P field, 542.7 3.25 3076 0.61 week 15, dry wt. 1981 low 8 field, * 149.2 3.63 2749 0.71 week 21, dry wt. * - Indicates bulb weight 1980 cultivar = 192, 1981 divided by 0.471 cultivar = Spartan Banner 149 As density reaches a high value, yield per unit area ap- proaches l/b, or the yield asymptote. This value has been used as a measure of the potential of the environment in producing yields. Few workers have studied density effects with time. Though density had greater effects on plant growth with time, 'b' values of asymptotic relationships declined with increased time, because of the nature of reciprocal numbers (Shinozaki and Kira, 1956). In other words, l/b, or the maximum yield increases with time, as Bleasdale (1966b) also proposed. Scaife and Jones (1976), however, found evidence that the maximum yields in lettuce were constant over time. Asympto- tic yields were acheived during all stages of the crop life- time, though this level was reached at lower densities with increased time. The effect of limiting resources may be greater at high densities than at low densities. Nichols (1967) found that onions exhibit different 'b' values with different fertili- zation levels, while the value of 'a' was not significantly different at various rates. In other words, at low densi- ties, fertilizer rates did not affect yield, but yield was increasingly depressed by suboptimal fertility with increa- sing plant density. Similar effects of irrigation on sever- al other crops have been reported (Frappell, 1979). 150 Gillis and Ratkowsky (1978) found that estimators of equa- tion (12) could be markedly biased. They suggested use of equation (11) to estimate unbiased parameters of a recipro- cal equation. Equation (11) was applied to data from a study by Lucas (1970) conducted at the Michigan State University Muck Farm (Table 26). This equation was also used to analyze yield- density relationships in the 1980 and 1981 data. Since treatments given 30, 97 or 193 kg/ha P of the 1981 low P field apparently received adequate nutrition, equation (11) was applied to the pooled data from all dates (Table 26). Equation (11) was applied to the pooled data from the 1981 high P field (Table 26). Total dry weights at the fin- al sampling dates of 1980 and 1981 were estimated by divid- ing the bulb weights by 0.471, the value that bulb weight as a percentage of total weight of top and bulb, approached. The row spacing in the 1981 experiments was 0.4 meters. Since this was constant, variations in density signified differences in rectangularities. These ranged from 8:1 to 25:1 in the low P field, and from 6:1 to 19:1 in the high P field. Increasing rectangularity at higher densities could be expected to decrease yields, but the effect of density could not be distinguished from plant arrangement. The 1980 and 1981 densities were also unreplicated, unlike the Lucas or Frappell experiments, which probably explains the differences in coefficients of determination. 151 The onions from the 1981 experiments exhibited much higher asymptotes, or the reciprocal of the slope values, than did onions grown in 1980, or in the Lucas and Frappell experi- ments. The 1981 fresh weights may have been slightly heavi- er than the field-cured weights of the Lucas or Frappell ex- periments, though the difference could only be on the order of a few percentage points, and could not account for the disparity in ceiling yields. The higher calculated asymp- totes from the 1981 experiments possibly resulted from envi- ronmental conditions more favorable for onion growth. An assumption was made that density-independent onion growth occurred only below a density of 10 plants per m2. Growth of plants was assumed equivalent at densities below this value. A density of 10 plants per m2 would provide ap- proximately 32 cm of space between plants in a 1:1 rectangu- larity. With rows 0.4 m apart, this density would provide 25 cm between plants within the row, with a rectangularity of 1.6:1. At higher rectangularities, lower densities might be necessary for density-independent onion growth. The slopes of 1/w against density regressions were similar for onions grown in the 1981 low P field and collected 15 or 21 weeks after planting (Table 26, Figure 24). This signi- fies that a constant maximum yield with time may have gov- erned onion growth, similar to the report of Scaife and Jones (1976) for lettuce. A constant asymptote is approached with increasing density (Figure 25), though this level is Figure 24. Figure 25. 152 A I A A I A 1 (IBM HEIGHT PER PLRNT (1/9) 0.00 A L l A L 0.00 L V V I v V v V U V v v \h"' PLRNT DENSITY (pub/m3) v I Y v v V The reciprocal of the dry weight per plant (l/g) versus plant density (plants per square meter) with fitted regression lines for data collected 15 weeks (circles) and 21 weeks (triangles) af- ter planting. 2900 90003300 TGTGI. PLRNT DRY HEIGHT (g/m’) too 1 1000 l 1 l l ' V I V ‘ V V ' V ‘Y Hh"'}b' PLGHT DENSITY (plum/m3) . V I ' Total plant dry weight (grams per square meter) predicted from regression and observed data from 15 weeks (circles) and 21 weeks (triangles) af- ter planting. 153 acheived at lower densities at a earlier date (data-Appendix 6). The density-independent plant weight of an onion crop can be estimated through the equation: l/DDPW = l/DIPW +'b * (Density - D0) (13) where DDPW = density-dependent plant weight, DIPW = density- independent plant weight, b = slope of l/w against density regression, Density = plants per unit area, and D0 = the maximum density at which density-independent plant weights occur. The value for 'b' can be estimated if at least two densities are present in the crop sampled. If the asymptote of yield with increasing density can be estimated for the existing environmental conditions, then 'b' would be equiva- lent to the reciprocal of the asymptote. Plant Emergence In order to accurately quantify onion growth over a sea- son, estimation was made of the time required from seeding to plant emergence. Emergence was recorded during the ini- tial weeks of the mycorrhizal experiments conducted with sandy muck and Houghton muck soils (III C and III D), in order to calculate emergence-time relationships. Percent emergence was calculated as the number of plants emerged at any time divided by the final number of emerged onions (Table 27). Air temperatures were also recorded on a hygro- thermograph, since temperature was thought to control 154 emergence rates to a large extent. Diurnal temperature curves were approximated as sine curves from observed maxi- mum and minimum temperatures (Baskerville and Emin, 1969). Degree days after planting were calculated with a lower tem- perature threshold of 5.6 C, or 42 F (Table 27), a value in- dicated by data of Butt (1968) to cause cessation of onion growth. Table 27. Emergence data and regressions of two emergence trials Trial 1 Trial 2 Degree Days % Emergence Degree Days % Emergence 66.6 0.0 74.8 0.0 86.7 7.6 98.9 5.3 93.6 28.8 116.6 60.1 109.5 67.5 127.0 76.9 127.4 90.1 182.6 97.4 159.1 . 99.0 204.8 99.0 209.7 100.0 245.7 99.7 301.0 99.7 319.3 100.0 Regression Equation 3: Trial 1: Y = -9.986 + 0.0937*DD 0.973 Trial 2: Y = -7.399 + 0.0611*DD 0.879 Y a 1n (x/(l-x)), x after emergence. % emergence, and DD = degree days Since the emergence pattern with increasing degree days resembled logistic growth (Figure 26), percent emergence was regressed against degree days after planting upon a logit transformation of In (x/(l-x), where x = % emergence between 155 PERCENT ENERGENCE 0 ‘ 50 1:30 1&0 2:30 250 350 350 DEGREE DHYS RFTER PLHNTING Figure 26. Percent onion emergence versus degree days (>5.6 C) after planting for trial 1 (circles) and tri- al 2 (triangles) with fitted regression lines. 156 0.01 and 0.99. Emergence was slightly delayed in trial 2 compared to trial 1, and late emergence was more pronounced in trial 2 (Figure 26). The lower slope of the trial 2 re- gression reflects the slower emergence (Table 27). For purposes of a simulation model, the regression equa- tion from trial 1 was selected, though both curves were sim- ilar. Since the regression predicted slight emergence at unrealistically low and high degree days, it was assumed to apply in a 57.5 to 163 degree day range. Emergence was as- sumed to be 0 and 100% below and above this range, respec- tively. Plant Maturity Accurate prediction of plant maturation is essential for prediction of onion growth, since rapid vegetative growth apparently occurs through most of the growing season. The 1980 and 1981 studies of leaf senescence showed relatively low levels of leaf senescence up to 77 and 105 days after planting, in the respective years. Cessation of growth probably occurred after the collapse of neck tissue. By the 1981 harvest, at 147 days after planting, the tops had fal- len on most plants, and leaf tissues were nearly dead from a combination of senescence and foliar diseases. The harvest was delayed due to flooding, and plant maturity had probably occurred some weeks earlier. Maturity was also obscured by foliar diseases during the 1980 season, so that data was not available at the time of neck collapse. 157 The timing of onion maturity is related to environmental and cultural factors, and can be expected to differ among cultivars. Maturity is favored by long photoperiods and high temperatures (Magruder et al, 1941). Excessive nitro- gen can delay maturity, while high plant densities may limit maturation time (Magruder et al, 1941). Water stress may also hasten onion maturity (Drinkwater and Janes, 1955). Since photoperiods will be relatively constant for field- grown onions, plant maturity can probably be predicted from temperature and water stress data, given that other cultural practices are optimal for growth. Since no data was avail- able for modeling plant maturity, a function was devised to predict leaf senescence, based upon when maturity was thought to occur in 1981. The function predicted green leaf surface area (GSA)/total leaf surface area (TSA) values, from degree days accumulated from planting. The function describing senescence was GSA/TSA = 0.915*e-0’1705*(DD'1780) , where DD is equivalent to degree days after planting computed above a 5.6 C base. This rapid leaf senescence is initiated when a degree day of 1780 is reached. At this time in 1981, the GSA/TSA ratio was calculated from equation (6) to be 0.915. By a degree day of 1820, senescence is largely complete, according to this function. More study needs to be done to predict onion maturity. Quantitative relationships should also be found for the predominant onion cultivars, and should include the 158 variables of both accumulated temperature and accumulated water stress 0 Relative Growth Rates as Related to Temperature Plant growth and development has been related to tempera- ture accumulution for a number of crops (Arnold, 1974, Gu- tierrez et al, 1975 and Wang, 1960). Frequently, growth or development has been described as a linear function of temp- erature above a minimum point, and often with a maximum threshold above which growth ceases. A similar attempt was made to describe relative growth rates as function of ac- cumulated temperature. Butt (1968) studied the growth of onions as related to temperature. Goltz and Tanner (1972) reported the effect of plant temperature on stomatal resistances of field-grown onions planted in Wisconsin and in Idaho. Relative growth rates were calculated from Butt's data and compared to the patterns of stomatal resistance with changes in temperature observed by Goltz and Tanner. Relative growth rates from Butt's data exhibited a maximum near 23 C and dropped off sharply with temperatures above this point. A minimum rela- tive growth was calculated to occur at approximately 5.6 C or 42 F, which is comparable to that found for other crops. A maximum growth occurring at 23 C is not realistic in light of the information from the field-grown onions, which exhib- ited a maximum stomatal conductance near leaf temperatures of 28 to 33 C. Air temperatures were within a degree or two 159 of leaf temperatures, and can not account for the disparity in observations. A quantitatitive relationship describing the relative growth rates of onions as a function of air temperature was assumed, based on the published observations. A minimum air temperature of 5.6 C necessary for growth was chosen, based on Butt's report (Figure 27). A maximum relative growth rate occurring at 28 C was assumed, and any temperatures above this point were not considered to increase the growth rate (Figure 27). Temperatures above 35 C, which might in- hibit onion growth, were not thought to occur frequently in Michigan onion fields, so provisions were not made for high inhibitory temperatures. Calculated Growth Rates of 1980 and 1981 Onions An average temperature of 28 C or higher for a day, accumulated one 'degree unit' under the temperature curve (Figure 27). Growth rates per degree unit were calculated from 1980 and 1981 dry plant weights and air temperatures. At emergence, leaf surface area and dry weight of the top 3 and 0.0012 grams, and bulb were assumed to equal 1.0 cm respectively. Density-independent plant weights were cal- culated from observed plant weights, plant densities and slopes of 1/w against density regressions (Table 28). Growth rates per degree unit were calculated from slopes of the natural logarithm of plant weight versus degree units (Figure 28). These growth rates possessed units of the 160 l RELBTIVE OROHW RRIE oo‘ 0.! 1 o I I I I I I I In is to as so TEHPERRTURE (C) Figure 27. Relative growth rate function (0-1) of onions as related to temperature (C). 9 S F o. A A ‘ i‘ 3 51 _°... *7- he Va. 5 4 D- ' I— "' . u z xq_ 38' >-" we” 3 g ' o—a mo“ E“ 894' I 0.1 E t‘" a , c: . g 1 59. red ‘1" #7., 3" g 4 . °. 8< I T'l'l'l'l‘l I I .1 I' 1 I 1'1 1 ' O a to to 30 ‘0 ‘0 3° 7° °° "' o 10 2: so a so so 70 I DEGREE UNITS RFTER PLRNTINO DEGREE UNITS” Figure 28. An example of the ln (plant dry weight (9)) versus degree units after planting (A) from which growth rates were calculated. The calcu- lated growth rates (l/degree unit) versus degree unit at interval midpoint (B) are shown for 1980 data (circles) and 1981 data (triangles). 161 Table 28. 1980 and 1981 onion data Days Plant Plant x _ Growth After Dry Wt Density2 DIPW Degree Rate Planting (g) (plantslm ) (9) Unit (1 DU) 1980 Data 36 0.0745 39.8Y 0.0747 9.29 0.445 49 0.494 39.8 0.502 16.29 0.272 62 1.92 39.8 2.04 24.79 0.165 77 7.73 39.8 10.18 36.13 0.142 100 12.82 39.8 21.31 51.51 0.048 1981 Data 21 0.0104 94.5 0.0104 7.64 0.283 42 0.1429 107.5 0.1436 21.42 0.191 63 1.47 95.2 1.54 36.91 0.153 84 5.82 99.5 7.17 50.91 0.110 105 11.73 96.1 17.38 63.85 0.068 147 22.41 79.4 53.89 78.20 0.079 x - DIPW = density-independent plant weight y - Only measured in 1980 on the 100th day after planting 162 change in dry weight/dry weight/degree unit. The calculated growth rates were assumed to apply at the degree unit of the interval midpoint (D050) (Table 28, Figure 28). The growth rates declined with time in both years, partic- ularly in 1980. Part of this drop can be attributed to high plant respiration by the larger plants, compared to smaller plants (McCree, 1974). However, some of the decrease in growth rates may also be attributed to plant water stress. The drops in growth during the last 1980 interval and in the next to last 1981 interval coincided to times of low water availability. Millar et al (1971) found onion growth to be quite sensi- tive to small changes in available soil water. Onion sto- mates closed at a leaf water potential of -6.5 to -7.0 bars, which occurred in soils with moisture tensions below -0.5 bars. This soil water potential could be expected to occur often in onion fields, and this lack of water could serious- ly limit onion growth. An assumption of abundant plant- available water may be difficult to justify, because of the sensitivity of onions to water stress. Part of the drop in growth rates observed in the last part of the 1980 season may have been caused by the downy mildew epidemic. Without further information of the effect of wa- ter stress on onion growth, it is difficult to understand the effects of diseases. 163 Simulation Program A FORTRAN program was written in order to simulate onion growth2 (Appendix 7). Maximum and minimum air temperatures are read, and diurnal temperature curves are estimated by sine curves. Degree days after planting are also calculated with a base temperature of 5.6 C. The program reads the number of days for the simulation to run, the time increment per step, the observed plant density, the density-indepen- dent onion dry weights, the degree unit after 50% emergence corresponding with the plant weight observations, and the 'b' values of the yield-density regressions. Emergence is predicted and plant weights of bulbs and leaves are calcu- lated. Density-dependent plant weights are calculated from the density-independent plant weights and a correction fac- tor for density calculated from equation 13. Green and se- nescent leaf surface area are calculated, based on equations 4-8. When a degree day of 1780 is reached, the plant maturation process is assumed to begin. This program calculated sizes of onion plants, based on 1980 and 1981 data (Table 29). Since plant maturation was thought to occur before day 147 in 1981, the degree unit at a degree day of 1800 was used of the day 147 degree unit. The growth rate calculated to apply at the midpoint of the last interval was applied beyond that point so the final 2 - The program was written with the help of Julia Pet of the Entomology Department. Table 29. 164 Plant simulation results 1980 Trial 1 density=10 plants/m 1981 Trial 1 density=10 plants/m 2 Plant Dry Weight (9) 0.0742 0.497 2.08 10.43 42.24 2 Plant Dry Weight (9) 0.0111 0.166 1.64 7.42 18.87 45.67 1980 Trial 2 2 density=39.8 plants/m Plant Dry 221 Weight (g) 36 0.0740 49 0.490 62 1.95 77 7.88 100 18.25 1981 Trial 2 2 density=100 plants/m Plant Dry 2&1 Weight (g) 21 0.0111 42 0.165 63 1.56 84 5.97 105 11.67 147 18.32 165 weights are overestimated for the 1980 data and underesti- mated for the 1981 data. The growth rate trends were de- creasing and increasing for the 1980 and 1981 data sets, re- spectively. If the effect of water stress can be included in the model, and the decreasing trend of growth rates pre- dicted, final estimation of onion yields should be much improved. If the effect of highly variable factors, such as temper- ature and water stress, are included in the simulation mod- e1, then the effect of relatively unchanging factors, such as cultivar, soil characteristics, or soil nutrition can be assessed through the observations made throughout the sea- son. Use of correction factors for plant density (equation 13) can also serve to calibrate onion growth, since they will vary with environmental conditions. The lack of data compatible with simulation construction was noted. Yields of variety trials were available (Jon Fobes, personal communication), but are not adaptable for inclusion in a simulation model, because of the nonuniformi- ty of plant densities. Results of fertilization trials related yields to added nutrient amounts, rather than con- centrations of nutrients with the soil. This knowledge is necessary in order to model onion responses to fertilizer amendments. 166 This model can provide a framework for sampling and pre- dicting onion growth. The assumptions need to be tested and the effect of other variables included, before the model can become effective on a practical basis. 167 Bibliography Abbott, L.K. and A.D. Robson. 1977a. The distribution and abundance of vesicular arbuscular endophytes in some Western Australian soil. Aust. J. Bot. 25:515-522. Abbott, L.K. and A.D. Robson. l977b. Growth stimulation of subterranean clover with vesicular arbuscular mycorrhizas. Aust. J. Agr. Res. 28:639-649. Abbott, L.K. and A.D. Robson. 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. Arnold, C.Y. 1974. Predicting stages of sweet corn devel- opment. J. Am. Soc. Hort. Sci. 99:501-505. Asimi, S., V. Gianinazzi-Pearson and S. Gianinazzi. 1980. Influence of increasing soil phosphorus levels on inter- actions between vesicular arbuscular mycorrhizae and Rhizo- bium in soybeans. Can. J. Bot. 58:2200-2205. Azcon, R. and J.A. Ocampo. 1981. Factors affecting the vesicular-arbuscular infection and mycorrhizal dependency of thirteen wheat cultivars. New Phytol. 87:677-685. Azcon-Aguilar, C. and J.M. Barea. 1981. Field inoculation of Medica o with v-a mycorrhiza and Rhizobium in phosphate- fixing agricultural soil. Soil Biol. Biochem. 13:19-22. Barrow, N.J., N. Malajczuk and T.C. Shaw. 1977. A direct test of the ability of vesicular-arbuscular mycorrhiza to help plants take up fixed soil phosphate. New Phytol 78: 269-276. Baskerville, G.L. and P. Emin. 1969. Rapid estimation of heat accumulation from maximum and minimum temperatures. Ecology 50:514-517. Baylis, G.T.S. 1967. Experiments on the ecological signif- icance of phycomycetous mycorrhizas. New Phytol. 66: 231-243. Baylis, G.T.S. 1969. Host treatment and spore production by Endogone. N. 2. J. Bot. 7:173-174. Baylis, G.T.S. 1970. Root hairs and phycomycetous mycor- rhizas in phosphorus-deficient soil. Plant Soil 33:713-716. 168 Baylis, G.T.S. 1972a. Fungi, phosphorus, and the evolution of root systems. Search 3:257-259. Baylis, G.T.S. 1972b. Minimum levels of available phos- phorus for non-mycorrhizal plants. Plant Soil 36:233-234. Baylis, G.T.S. 1975. The magnolioid mycorrhiza and myco- trophy in root systems derived from it. Pages 373-389 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhi- zas. Academic Press, New York. Bethlenfalvay, G.J., M.S. Brown and R.S. Pacovsky. 1982. Relationships between host and endophyte development in mycorrhizal soybeans. New Phytol. 90:537-543. Bethlenfalvay, G.J., R.S. Pacovsky and M.S. Brown. 1981. Measurement of mycorrhizal infection in soybean. Soil Sci. Soc. Am. J. 45:871-875. Bevege, D.T., G.D. Bowen and M.F. Skinner. 1975. Compara- tive carbohydrate physiology of ecto- and endo- mycorrhizas. Pages 149-174 in: F.E. Sanders, B. Mosse and P.B. Tinker. Endomycorrhizas. Academic Press, New York. Biermann, B. and R.G. Linderman. 1981. Quantifying vesicu- lar-arbuscular mycorrhizae: a proposed method toward stan- dardization. New Phytol. 87:63-67. Black, R.L.B. and P.B. Tinker. 1977. Interaction between effects of vesicular-arbuscular mycorrhiza and fertiliser phosphorus on yields of potatoes in the field. Nature 267: 510-511. Black, R. and P.B. Tinker. 1979. The development of endo- mycorrhizal root systems II. Effect of agronomic factors and soil conditions on the development of vesicular-arbuscular mycorrhizal infection in barley and on the endophyte spore density. New Phytol. 83:401-413. Bleasdale, J.K.A. 1966a. The effect of plant spacing on the yield of bulb onions (Allium cepa L.) grown from seed. J. Hort. Soc. 41:145-153. Bleasdale, J.K.A. 1966b. Plant growth and crop yield. Ann. Appl. Biol. 57:173-182. Bleasdale, J.K.A. and J.A. Nelder. 1960. Plant population and crop yield. Nature 188:342. 169 Bray, R. H. 1948. Correlation of soil tests with crop response to added fertilizer and with fertilizer require- ments. Pages 53- 86 in: Diagnostic techniques for soils and crops. American Potash Institute. Washington, D.C. Bray, R.H. and L.T. Kurtz. 1945. Determination of total, organic, and available forms of phosphorus in soils. Soil Sci. 59:39-45. Butt, A.M. 1968. Vegetative growth, morphogenesis and carbohydrate content of the onion plant as a function of light and temperature under field- and controlled condi- tions. Meded. Lanbouwhogeschool Wageningen 68-10. Carling, D.E., M.F. Brown and R.A. Brown. 1979. Coloniza- tion rates and growth responses of soybean plants infected by vesicular-arbuscular mycorrhizal fungi. Can. J. Bot. 57:1769-1772. Carling, D.E., W.G. Riehle, M.F. Brown and D.R. Johnson. 1978. Effects of a vesicular-arbuscular mycorrhizal fungus on nitrate reductase and nitrogenase activities in nodulat- ing and non-nodulating soybeans. Phytopathol. 68:1590-1596. Christie, P., E.I. Newman and R. Campbell. 1978. The in- fluence of of neighboring grassland plants on each other's endomycorrhizas and root surface microorganisms. Soil Biol. Biochem. 10:521-527. Clarke, C. and B. Mosse. 1981. Plant growth responses to vesicular-arbuscular mycorrhiza XII. Field inoculation responses of barley at two soil P levels. New Phytol. 87:695-703. Cooper, K. M. 1975. Growth responses to the formation of endotrophic mycorrhizae in Solanum, Leptospermum, and New Zealand ferns. Pages 391- 407 in: F. E. Sanders, B. Mosse and P.B. Tinker. Endomycorrhizas. Academic Press, New York. Crush, J.R. 1973a. The effect of Rhizophagus tenius mycorrhizas on ryegrass, cocksfoot and sweet vernal. New Phytol. 72:965-973. Crush, J. R. 1973b. Significance of endomycorrhizas in tussock grassland in Otago, New Zealand. N. 2. J. Bot. 11:645-660. Crush, J.R. 1974. Plant growth responses to vesicular- arbuscular mycorrhiza VII. Growth and nodulation of some herbage legumes. New Phytol. 73:743-749. 170 Crush, J.R. 1976. Endomycorrhizas and legume growth in some soils of the Mackenzie Basin, Canterbury, New Zealand. N. Z. J. Agr. Res. 19:473-476. Daft, M.J. and A.A. El-Giahmi. 1974. Effect of Endo one mycorrhiza on plant growth VII. Influence of infect1on on the growth and nodulation in French bean (Phaseolus vulgaris). New Phytol. 73:1139-1147. Daft, M.J. and A.A. El-Giahmi. 1978. Effect of arbuscular mycorrhiza on plant growth VIII. Effects of defoliation and light on selected hosts. New Phytol. 65:343-350. Daft, M.J. and T.H. Nicolson. 1966. Effect of Endogone mycorrhiza on plant growth. New Phytol. 65:343-350. Daft, M.J. and T.H. Nicolson. 1969a. Effect of Endogone mycorrhiza on plant growth II. Influence of soluble phos- phate on endophtye and host in maize. New Phytol. 68:945-952. Daft, M.J. and T.H. Nicolson. 1969b. Effect of Endogone mycorrhiza on plant growth III. Influence of inoculum con- centration on growth and infection in tomato. New Phytol. 68:953-967. Daft, M.J. and T.H. Nicolson. 1972. Effect of Endo one mycorrhiza on plant growth IV. Quantitative relat1onsh1ps between the growth of the host and the develoopment of the endophyte in tomato and maize. New Phytol. 71:287-295. Drinkwater, W.O. and 8.8. Janes. 1955. Effects of irriga- tion and soil moisture on maturity, yield and storage of two onion hybrids. Amer. Soc. Hort. Sci. 66:267-278. Epstein, E. 1972. Mineral nutrition of plants: principles and perspectives. John Wiley and Sons, Inc., New York. Farazdaghi, H. 1968. Some aspects of the interaction be- tween irrigation and plant density in sugar beet. Ph.D. Thesis. Reading University, England. Frappell, B.D. 1973. Plant spacing of onions. J. Hort. Sci. 48:19-28. Frappell, B.D. 1979. Competition in vegetable crop communities. J. Aust. Inst. Agr. Sci. 45:211-217. Furlan, V. and J.-A. Fortin. 1973. Formation of endomycor- rhizae by Endogone calospora on Allium cepa under three tem- perature regimes. Nat. Can. 100:467-477. 171 Furlan, V. and J. A. Fortin. 1977. Effects of light inten- sity on the formation of vesicular-arbuscular endomycorrhi- zas on Allium cepa by Gigaspora calospora. New Phytol. 79. 335- 340. Gerdemann, J.W. 1968. Vesicular-arbuscular mycorrhiza and plant growth. Ann. Rev. Phytopathol. 6:397-418. Gianinazzi-Pearson, V. and S. Gianinazzi. 1978. Enzymatic studies on the metabolism of vesicular-arbuscular mycorrhiza II. Soluble alkaline phosphatase specific to mycorrhizal infection in onion roots. Physiol. Plant Path. 12:45-53. Gillis, P.R. 1979. The relationship between plant density and crop yield when depressions in yield per plant are ob- served at low densities. J. Hort. Sci. 54:305-307. Gillis, P.R. and D.A. Ratkowsky. 1978. The behavior of es- timators of the parameters of various yield-density rela- tionships. Biometrics 34:191-198. Gilmore, A.E. 1971. The influence of endotrophic mycorrhi- zae on the growth of peach seedlings. J. Amer. Soc. Hort. Sci. 96:35-37. Giovanetti, M. and B. Mosse. 1980. An evaluation of tech- niques for measuring vesicular arbuscular mycorrhizal infec- tion in roots. New Phytol. 84:489-500. Goltz, S.M. and C.B. Tanner. 1972. Seed onion temperature and their effect on stomata. Hort. Sci. 7:180-181. Grunes, D.L. 1959. Effect of nitrogen on the availability of soil and fertilizer phosphorus to plants. Advance Agron. 11:369-396. Gutierrez, A.P., L.A. Falcon, W. Loew, P.A. Leipzig and R. van den Bosch. 1975. An analysis of cotton production in California: A model of cotton production and the effects of defoliators on yield. Environ. Entomol. 4:125-136. Hall, I.R. 1976. Response of Cgprosma robusta to different forms of endomycorrhizal inoculum. Trans. Br. Mycol. Soc. 67:409-411. Hall, I. R. 1977. Effect of applied nutrients and endo- mycorrhizas on Metrosideros umbellata and Leptospermum scoparium. N. Z. J. Bot. 15: 481- 484. 172 Hall, I.R. 1978. Effect of vesicular-arbuscular mycorrhi- zas on two varieties of maize and one of sweetcorn. N. 2. J. Agr. Res. 21:517-519. Hall, I.R., R.S. Scott and P.D. Johnstone. 1977. Effect of vesicular-arbuscular mycorrhizas on response of 'Grasslands Huia' and 'Tamar' white clovers to phosphorus. N. 2. J. Agr. Res. 20:349-355. Hayman, D.S. 1970. Endogone spore numbers in soil and vesicular-arbuscular mycorrh1za in wheat as influenced by season and soil treatment. Trans. Br. Mycol. Soc. 54:53-63. Hayman, D.S. 1974. Plant growth responses to vesicular- arbuscular mycorrhiza VI. Effect of light and temperature. New Phytol. 73:71-80. Hayman, D.S. 1975. The occurrence of mycorrhiza in crops as affected by soil fertility. Pages 495-509 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Hayman, D.S., A.M. Johnson and I. Ruddlesdin. 1975. The influence of phosphate and crop species on Endogone spores and vesicular-arbuscular mycorrhiza under f1eld conditions. Plant Soil. 43:489-495. Hayman, D.S. and B. Mosse. 1971. Plant growth responses to vesicular-arbuscular mycorrhiza I. Growth of Endogone-inoc- ulated plants in phosphate-deficient soils. New Phytol. 70:19-27. Hayman, D.S. and B. Mosse. 1972. Plant growth responses to vesicular-arbuscular mycorrhiza III. Increased uptake of la- bile P from soil. New Phytol. 71:41-47. Hayman, D.S. and B. Mosse. 1979. Improved growth of white clover in hill grasslands by mycorrhizal inoculation. Ann. Appl. Biol. 93:141-148. Hepper, C.M. 1977. A colorimetric method for estimating vesicular-arbuscular mycorrhizal infection in roots. Soil' Biol. Biochem. 9:15-18. Holevas, C.D. 1966. The effect of a vesicular-arbuscular mycorrhiza on the uptake of soil phosphorus by strawberry (Fragaria sp. var. Cambridge Favourite). J. Hort. Sci. 41:57-64. 173 Holliday, R. 1960a. Plant population and crop yield. Nature 186:22-24. Holliday, R. 1960b. Plant population and crop yield. Field Crop Abstr. 13:159-167 and 247-254. Jaspar, D.A., A.D. Robson and L.K. Abbott. 1979. Phosphor- us and the formation of vesicular-arbuscular mycorrhizas. Soil Biol. Biochem. 11:501-505. Jensen, A. 1982. Influence of four vesicular-arbuscular mycorrhizal fungi on nutrient uptake and growth in barley (Hordeum vulgare). New Phytol. 90:45-50. Jensen, A. and I. Jakobsen. 1980. The occurrence of vesic- ular-arbuscular mycorrhiza in barley and wheat grown in some Danish soils with different fertilizer treatments. Plant Soil 55:403-414. Jones, H.A. and L.K. Mann. 1963. Onions and their allies. Leonard Hill Ltd. Interscience Publishers Inc., New York. Khan, A.G. 1972. The effects of vesicular-arbuscular my- corrhizal accociations on growth of cereals I. Effects on maize growth. New Phytol. 71:613-619. Khan, A.G. 1975a. The effect of vesicular arbuscular my- corrhizal associations on growth of cereals II. Effects on wheat growth. Ann. Appl. Biol. 80:27-36. Khan, A.G. 1975b. Growth effects of va mycorrhiza on crops in the field. Pages 419-435 in: F.E. Sanders, B. Mosse, and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Khan, A.G. 1981. Growth responses of endomycorrhizal on- ions in unsterilized coal waste. New Phytol. 87:363-370. Kleinschmidt, G.D. and J.W. Gerdemann. 1972. Stunting of citrus seedlings in fumigated nursery soils related to the absence of endomycorrhizae. Phytopathol. 62:1447-1453. Kormanik, PP. 1981. A forester's view of vesicular-arbus- cular mycorrhizae and their role in plant development and productivity. Pages 33-37 in: R.F. Myers ed. Mycorrhizal associations and crop production. N. J. Ag. Exper. Stat. Res. Rep. No. RO4400-01-81. Kormanik, P.P., W.C. Bryan and R.C. Schultz. 1977. Influ- ence of endomycorrhizae on growth of sweetgum seedlings from eight mother trees. Forest Sci. 23:500-506. 174 Kotcon, J. 1979. Studies on the ecology of nematodes associated with vegetables in organic soils. M.S. Thesis. Mich. State Univ., E. Lansing. Koucheki, J.K. and D.J. Read. 1976. Vesicular-arbuscular mycorrhiza in natural vegetation systems II. The relation- ship between infection and growth in Festuca ovina L. New Phytol. 77:655-666. Kruckelmann, H.W. 1975. Effects of fertilizers, soils, soil tillage, and plant species on the frequency of Endogone chlamydospores and mycorrhizal infection in arable so1ls. Pages 511-525 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Lambert, D.H., H. Cole, Jr. and D.E. Baker. 1980. Adapta- tion of vesicular-arbuscular mycorrhizae to edaphic factors. New Phytol 85:513-520. Lampert, E.P. 1980. Endemic ecology of the cereal leaf beetle (Oulema melanopus(L.)). Ph.D. Thesis. Mich. State Univ., E. Lansing. 528pp. Lucas, R.E. 1970. Onion spacing and population studies. Mich. Agr. Exper. Stat. Res. Rep. 115. Magruder, R., R.E. Webster, H.A. Jones, T.E. Randall, G.B. Synder, H.D. Brown, L.R. Hawthorn and A.L. Wilson. 1941. Description of types of principal American varieties of onions. USDA Misc. Publ. 435. Manjunath, A. and D.J. Bagyaraj. 1981. Components of va mycorrhizal inoculum and their effects on growth of onion. New Phytol. 87:355-361. McCool, P.M. 1980. Induced pathogenicity of the vesicu- lar-arbuscular mycorrhizal fungus, Glomus fasciculatus on tomato subjected to environmental stress. Phytopathol. 71:240 (abstract). McCree, K.J. 1974. Equations for the rate of dark respi- ration of white clover and grain sorghum, as functions of dry weight, photosynthetic rate, and temperature. Crop Sci. 14:509-514. Menge, J.A. 1982. Effects of soil fumigants and fungicides on endomycorrhizal fungi. Phytopathol. 72:1125-1132. Menge, J.A., C.K. Labanauskas, E.L.V. Johnson and R.G. Platt. 1978a. Partial substitution of mycorrhizal fungi for phosphorus fertilization in the greenhouse culture of 175 citrus. Soil Sci. Soc. Am. J. 42:926-930. Menge, J.A., D. Steirle, D.J. Bagyaraj, E.L.V. Johnson and R.T. Leonard. 1978b. Phosphorus concentrations in plants responsible for inhibition of mycorrhizal infection. New Phytol. 80:575-578. Miller, A.A., W.R. Gardner and S.M. Goltz. 1971. Internal water status and water transport in seed onion plants. Agron. J. 63:779-784. Miller, O.K., Jr. 1981. Taxonomy, morphology, and distri- bution of mycorrhizae. Pages 5-13 in: R.F. Myers ed. Mycor- rhizal associations and crop production. N. J. Ag. Exper. Stat. Res. Rep. No. RO4400-01-81. Mosse, B. 1972a. Effects of different Endogone strains on the growth of Papsalum notatum. Nature 239:221-223. Mosse, B. 1972b. The influence of soil type and Endogone strain on the growth of mycorrhizal plants in phosphate deficient soils. Rev. Ecol. Biol. Sol. 9:529-537. Mosse, B. 1973. Plant growth responses to vesicular-arbus- cular mycorrhiza IV. In soil given additional phosphate. New Phytol. 72:127-136. Mosse, B. 1975. Specificity in vesicular-arbuscular mycor- rhizas. Pages 469-484 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Mosse, B. 1977. Plant growth response to vesicular-arbus- cular mycorrhizas X. Responses of Stylosantheg and maize to inoculation in unsterile soils. New Phytol. 78:277-288. Mosse, B. and D.S. Hayman. 1971. Plant growth responses to vesicular-arbuscular mycorrhiza II. In unsterilized field soils. New Phytol. 70:29-34. Mosse, B., D.S. Hayman and D.J. Arnold. 1973. Plant growth responses to vesicular-arbuscular mycorrhiza V. Phosphate uptake by thgse plant species from P-deficient soils la- belled with P. New Phytol. 72:809-815. Mosse, B., D.S. Hayman and G.J. Ide. 1969. Growth respons- es of plants in unsterilized soil to inoculation with vesic- ular-arbuscular mycorrhiza. Nature 224:1031-1032. Mosse, B. and J.M. Phillips. 1971. The influence of phos- hate and other nutrients on the development of vesicular- arbuscular mycorrhiza in culture. J. Gen. Microbiol. 69: 157-166. 176 Mosse, B., A. Warner and C.A. Clarke. 1982. Plant growth responses to vesicular-arbuscular mycorrhiza XIII. Spread of an introduced va endophyte in the field and residual growth effects of inoculation in the second Year. New Phytol. 90: 521-528. Nelsen, C.E. and G.R. Safir. 1982a. The water relations of well-watered, mycorrhizal and nonmycorrhizal onion plants. J. Amer. Soc. Hort. Sci. 107:271-274. Nelsen, C.E. and G.R. Safir. 1982b. Increased drought tol- erance of mycorrhizal onion plants caused by improved phos- phorus nutrition. Planta 154:407-413. Nemec, S. 1978. Response of six citrus rootstocks to three species of Glomus, a mycorrhizal fungus. Proc. Fla. State Hort. Soc. 91:10-14. Nichols, M.A. 1967. A note on a plant density and fertili- ser experiment with onions in New Zealand. Hortic. Res. 7: 144-147. Ocampo, J.A. and D.S. Hayman. 1981. Influence of plant interactions on vesicular-arbuscular mycorrhizal infections. II. Crop rotations and residual effects of non-host plants. New Phytol. 87:333-343. Ocampo, J.A., J. Martin and D.S. Hayman. 1980. Influence of plant interactions on vesicular-arbuscular mycorrhizal infections. I. Host and non-host plants grown together. New Phytol. 84:27-35. Ohms, R.E. 1957. A flotation method for collecting spores of a phycomycetous mycorrhizal parasite from soil. Phyto- pathol. 47:751-752. Olsen, S.R., F.S. Watanabe and R.E. Danielson. 1961. Phos- phorus absorption by corn roots as affected by moisture and phosphorus concentration. Soil Sci. Soc. Proc. 25:289-294. Owusu-Bennoah, E. and B. Mosse. 1979. Plant growth responses to vesicular-arbuscular mycorrhiza XI. Field inoculation responses in barley, lucerne and onion. New Phytol. 83:671-679. Owusu-Bennoah, E. and A. Wild. 1979. Autoradiography of the depletion zone around onion roots in the presence of vesicular-arbuscular mycorrhiza. New Phytol. 82:133-140. 177 Pairunan, A.K., A.D. Robson and L.K. Abbott. 1980. The effectiveness of vesicular-arbuscular mycorrhizas in in- creasing growth and phosphorus uptake of subterranean clo- ver from phosphorus sources of different solubilities. New Phytol. 84:327-338. Phillips, J.M. and D.S. Hayman. 1970. Improved procedure for clearing roots and staining parasites, and vesicular- arbuscular mycorrhizal fungi for rapid assessment of infec- tion. Trans. Br. Mycol. Soc. 55:158-161. Porter, D.M. and M.K. Beute. 1972. Endogone species in roots of Virginia type peanuts. Phytopathol. 62:783 ~(abstract). Porter, W.M. 1979. The 'most probable number' method for enumerating infected propagules of vesicular arbuscular mycorrhizal fungi in soil. Aust. J. Soil Res. 17:515-519. Porter, W.M., L.K. Abbott and A.D. Robson. 1978. Effect of rate of application of superphosphate on populations of vesicular arbuscular endophytes. Aust. J. Exp. Agr. Anim. Husb. 18:573-578. Powell, C.Ll. 1975a. Potassium uptake by endotrophic mycorrhizas. Pages 461-468 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Powell, C.Ll. 1975b. Plant growth responses to vesicular- arbuscular mycorrhiza VIII. Uptake of P by onionBQnd clover infected with different Endogone spore types in P labelled soils. New Phytol. 75:563-566. Powell, C.L1. 1976. Development of mycorrhizal infections from Endogone spores and infected root segments. Trans. Br. Mycol. Soc. 66:439-445. Powell, C.Ll. l977a. Mycorrhizas in hill-country soils II. Effect of several mycorrhizal fungi on clover growth in sterilised soils. N. 2. J. Agr. Res. 20:59-62. Powell, C.L1. l977b. Mycorrhizas in hill country soils III. Effect of inoculation on clover growth in unsterile soils. N. 2. J. Agr. Res. 20:343-348. Powell, C.Ll. 1977c. Mycorrhizas in hill country soils. V. Growth responses in ryegrass. N. 2. J. Agr. Res. 20:495-502. 178 Powell, C.Ll. 1979. Spread of mycorrhizal fungi through soil. N. 2. J. Agr. Res. 22:335-339. Powell, C.L1. and J. Daniel. 1978. Growth of white clover in undisturbed soils after inoculation with efficient mycor- rhizal fungi. N. 2. J. Agr. Res. 21:675-681. Powell, C.L1. and J. Sithamparanathan. 1977. Mycorrhizas in hill country soils IV. Infection rate in grass and legume species by indigenous mycorrhizal fungi under field condi- tions. N. Z. J. Agr. Res. 20:489-494. Ratnayake, M., R.T. Leonard and J.A. Menge. 1978. Root exudation in relation to supply of phosphorus and its pos- sible relevance to mycorrhizal formation. New Phytol. 81: 543-552. Redhead, J.F. 1975. Endotrophic mycorrhizas in Nigeria: Some aspects of the ecology of the endotrophic mycorrhizal association of Khaya grandifolia C. DC. Pages 447-459 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhi- zas. Academic Press, New York. Reid, C.P. and G.D. Bowen. 1979. Effects of soil moisture on v/a mycorrhiza formation and root development in Medicago. Pages 211-219 in: J.H. Harley and R.S. Russell eds. The soil-root interface. Academic Press, New York. Rhodes, L.H. and J.W. Gerdemann. 1975. Phosphate uptake zones of mycorrhizal and nonmycorrhizal onions. New Phytol. 75:555-561. Rhodes, L.H. and J.W. Gerdemann. 1980. Nutrient translo- cation in vesicular-arbuscular mycorrhizae. Pages 173-195 in: C.B. Cook, P.W. Pappas and E.D. Rudolph eds. Cellular interactions in symbiosis and parasitism. Ohio State Uni- versity Press, Columbus, Ohio. (Rhodes, L.H., M.C. Hirrell and J.W. Gerdemang5 1978.31nf1u- ence of soil phosphorus on translocation of Zn and P by external hyphae of vesicular- arbuscular mycorrhizae of onion. Phytopathol. News 12: 197 (abstract). Rich, J.R. and G.W. Bird. 1974. Association of early-season vesicular-arbuscular mycorrhizae with increased growth and development of cotton. Phytopathol. 64:1421-1425. Rogers, 1.5. 1977. The influence of plant spacing on the frequency distribution of bulb weight and marketable yield of onions. J. Hort. Sci. 53:153-161. 179 Ross, J.P. 1971. Effect of phosphate fertilization on yield of mycorrhizal and nonmycorrhizal soybeans. Phyto- pathol. 61:1400-1403. Ross, J.P. 1980. Effect of nontreated field soil on sporu- lation of vesicular-arbuscular mycorrhizal fungi associated with soybean. Phytopathol. 70:1200-1205. Ross, J.P. and J.A. Harper. 1970. Effect of Endogone mycorrhiza on soybean yields. Phytopathol. 60:1552-1556. Ross, J.P. and R. Ruttencutter. 1977. Population dynamics of two vesicular-arbuscular endomycorrhizal fungi and the role of hyperparasitic fungi. Phytopathol. 67:490-496. Safir, G.R. 1980. Vesicular-arbuscular mycorrhizae and crop productivity. Pages 231-252 in: P.S. Carlson ed. The biology of crop productivity. Academic Press, New York. Safir, G.R., J.S. Boyer and J.W. Gerdemann. 1972. Nutrient status and mycorrhizal enhancement of water transport in soybean. Plant PhysiOl. 49:700-707. Safir, G.R. and C.E. Nelsen. 1981. Water and nutrient uptake by vesicular-arbuscular mycorrhizal plants. Pages 25-31 in: R.F. Myers ed. Mycorrhizal associations and crop production. N. J. Agr. Exper. Stat. Res. Rep. No. RO4400- 01-81. Saif, S.R. 1977. The influence of stage of host develop- ment on vesicular-arbuscular mycorrhizae and endogonaceous spore populations in field-grown vegetable crops I. Summer- grown crops. New Phytol. 79:341-348. Saif, S.R. and A.G. Khan. 1977. The effect of vesicular- arbuscular mycorrhizal association on growth of cereals III. Effects on barley growth. Plant Soil 47:17-26. Sanders, F.E. 1975. The effect of foliar-applied phosphate on the mycorrhizal infections of onion roots. Pages 261-276 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycor- rhizas. Academic Press, New York. Sanders, F.E. and P.B. Tinker. 1973. Phosphate flow into mycorrhizal roots. Pestic. Sci. 4:385-395. Sanders, F.E., P.B. Tinker, R.L.B. Black and S.M. Palmerley. 1977. The development of endomycorrhizal root systems: 1. Spread of infection and growth-promoting effects with four species of vesicular-arbuscular endophyte. New Phytol. 78:- 257-268. 180 Scaife, M.A. and D. Jones. 1976. The relationship between crop yield (or mean plant weight) of lettuce and plant den- sity, length of growing period, and initial plant weight. Agric. Sci. 86:83-91. Schenck, N.C. 1981. Relationships of mycorrhizal fungi to other microorganisms in the root and rhizosphere. Pages 19-23 in: R.F. Myers ed. Mycorrhizal associations and crop production. N. J. Agr. Exper. Stat. Res. Rep. No. RO4400- 01-81. Schenck, N.C. and M.K. Kellam. 1978. The influence of vesicular arbuscular mycorrhizae on disease development. Fla. Agr. Exper. Stat. Bull. 798. Schenck, N.C. and R.A. Kinloch. 1980. Incidence of mycor- rhizal fungi on six field crops in monoculture on a newly cleared woodland site. Mycologia 72:445-456. Schenck, N.C. and V.M. Schroder.' 1974. Temperature re- sponse of Endogone mycorrhiza on soybean roots. Mycologia 66:600-605. Schultz, R.C., P.P. Kormanik, W.C. Bryan and G.H. Brister. 1979. Vesicular-arbuscular mycorrhiza influence growth but not mineral concentrations in seedlings of eight sweetgum families. Can. J. For. Res. 9:218-223. Shinozaki, K. and T. Kira.' 1956. Intraspecific competition among higher plants. VII. Logistic theory of the C-D effect. J. of Institute of Polytechnics, Osaka City Univer., Ser. D7:35-72. Smith, F.A. and S.E. Smith. 1981. Mycorrhizal infection and growth of Trifolium subterranean: comparison of natural and artificial inocula. New Phytol. 88:311-325. Sparling, G.R. and P.B. Tinker. 1978a. Mycorrhizal infec- tion in Pennine grassland 1. Levels of infection in the field. J. Appl. Ecol. 15:943-950. Sparling, C.P. and P.B. Tinker. 1978b. Mycorrhizal infec- tion in Pennine grassland II. Effects of mycorrhizal infec- tion on the growth of some upland grasses on -irradiated soils. J. Appl. Ecol. 15:951-958. St. John, T.V. 1980. Root size, root hairs and mycorrhizal infection: a re-examination of Baylis's hypothesis with tropical trees. New Phytol. 84:483-487. 181 Stribley, D.P., P.B. Tinker and J.H. Rayner. 1980a. Rela- tion of internal phosphorus concentration and plant weight in plants infected by vesicular-arbuscular mycorrhizas. New Phytol. 86:261-266. Stribley, D.P., P.B. Tinker and R.C. Snellgrove. 1980b. Effect of vesicular-arbuscular mycorrhizal fungi on the re- lations of plant growth, internal phosphorus concentration and soil phosphate analysis. J. Soil Sci. 31:655-672. Strzemski, J. 1975. Mycorrhiza in farm crops grown in ~ monoculture. Pages 527-535 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Sutton, J.C. 1973. Development of vesicular-arbuscular mycorrhizae in crop plants. Can. J. Bot. 51:2487-2493. Sutton, J.C. and G.L. Barron. 1972. Population dynamics of Endogone spores in soil. Can. J. Bot. 50:1909-1914. Thien, S.J. and W.W. McFee. 1970. Influence of nitrogen on phosphorus absorption and translocation in Zea may . Soil Sci. Soc. Amer. Proc. 34:87-90. Timmer, L.W. and R.F. Leyden. 1978. Relationship of seedbed fertilization and fumigation to infection of sour orange seedlings by mycorrhizal fungi and Phytophthora parasitica. J. Amer. Soc. Hort. Sci. 103:537-541. Timmer, L.W. and R.F. Leyden. 1980. The relationship of mycorrhizal infection to phosphorus-induced copper deficien- cy in sour orange seedlings. New Phytol. 85:15-23. Tinker, P.B. 1975a. Soil chemistry of phosphorus and my- corrhizal effects on plant growth. Pages 353-371 in: F.E. Sanders, B. Mosse and P.B. Tinker eds. Endomycorrhizas. Academic Press, New York. Tinker, P.B. 1975b. Effects of vesicular-arbuscular mycorrhizas on higher plants. Symp. Soc. Expt. Biol. 29:325-349. Tommerup, I.C. and L.K. Abbott. 1981. Prolonged survival and viability of va mycorrhizal hyphae after root death. Soil Biol. Biochem. 13:431-433. Trinick, M.J. 1977. Vesicular-arbuscular infection and soil phosphorus utilization in Lupinus spp. New Phytol. 78:297-304. 182 Wang, J.Y. 1960. A critique of the heat unit approach to plant response studies. Ecology 41:785-790. Warner, A. and B. Mosse. 1980. Independent spread of ve- sicular-arbuscular mycorrhizal fungi in soil. Trans. Br. Mycol. Soc. 74:407-410. Warner, A. and B. Mosse. 1982. Factors affecting the spread of vesicular mycorrhizal fungi in soil. I. Root den- sity. New Phytol. 90:529-536. Warnock, A.J, A.H. Fitter and M.B. Usher. 1982. The influ- ence of springtail Folsomia candida (Insecta, Collembola) on the mycorrhizal association of leek Allium porrum and the vesicular-arbuscular mycorrhizal endophyte Glomus fascicu- latus. New Phytol. 90:285-292. Weaver, J.E. and W.E. Bruner. 1927. Root development of vegetable crops. McGraw-Hill, New York. Willey, R.L.V. and S.B. Heath. 1969. The quantitative re- lationships between plant population and crop yield. Adv. Agron. 21:281-321. Appendix 1. 183 Field locations of 1979 commercial field survey Field GO-l GO-2 Cl 62 G3 G4 GS G6 ERl ERZ ER3 MF Location Grant, research field: across Spruce St. from Entomology trailer-Dyfonate plot Grant, non-Dyfonate plot in above field Grant, Peter Brink's field: back field Grant, Peter Brink's field: front field Grant, P. Plakmeyer's field: on north end of swamp off dirt road Grant, J. Plaisier's field: 120th Ave. Grant, P. Brink's field: north of 120th Ave. and red shed Grant, J. Plaisier's field: s.e. corner of 120th Ave. and Spruce St. Eaton Rapids, D. Kunkel's field: research plot Eaton Rapids, D. Kunkel's field: yellow onions Eaton Rapids, D. Kunkel's field: red onions Lainsburg, lst planting in Entomology plot at Muck Farm 184 Appendix 2. Procedure for Root-staining and Spore Isolation Root-staining Procedure Cover roots in large test tubes with 10% KOH. Heat tubes at 85 C in water bath for 50 minutes. Wash roots on screen to remove KOH. Acidify roots by soaking for 1 hour in 0.1 M HCl. Stain overnight in 0.1% acid fuchsin in lactophenol. Lactophenol is made by mixing by weight: one part phenol, 1 part lactic acid, 1 part distilled water and 2 parts glycerol. Destain roots in clear lactophenol twice, with roots remaining in each solution overnight. Spore Isolation Procedure 1. Sieve soil through 425 um and 38 um screens with 38 um screen on bottom. Spores will be trapped on this screen. Decant spores, soil and debris into centrifuge, and spin for 3 minutes at 2500 rpm in a tipping-bucket centri- fuge. Decant water and light organic matter. Add 50% sucrose (45.5 g/100 ml H20) to centrifuge tube and mix to resuspend spores. Centrifuge for 125 minutes. Decant sucrose into 38 um screen and wash spores with water. Transfer spores to a gridded petri plate for counting. Note: Repeated extraction of spores from the same samples showed that an average of 88% of the total amounts of spores were extracted during the first isolation. A11 spore counts reported were first extractions divided by 0.88, with the exception of the 1979 commercial field survey which reports the first extraction only. 185 Appendix 3. Commercial onion fields sampled in 1980 Fieldx Location Plot Soily GO Grant, J. Plaisier's field: 1-west end SM across Spruce St. from Ento- mology trailer-western half in onions 61 Grant, P. Brink's field: 1-north end SM back 80 acres-western edge 2-south end OM sampled G3 Grant, P. Plakmeyer's field: 1-north end OM on north end of swamp off dirt road G4 Grant, Bolthouse Farm's field: l-north end M n.e. corner of Oak St. and 2-midd1e 120th Ave. 3-south end M GS Grant, P. Brink's field: l-south end M north of 120th Ave. and red 2-middle M/OM shed - eastern half in onions 3-north end OM ERl Eaton Rapids, D. Kunkel's l-near end OM field: 1/4 mile behind 2-far end OM house Field Planting Date Cultivar 61 5/7 192 G3 4/28 Sentinel G4 4/26 Spartan Banner GS 5/7 Spartan Banner x - Numbers may not be those of the 1979 survey. y - M = marl, OM c organic muck, SM 2 sandy muck 186 Appendix 4. Bray P1 Soil Phosphorus Extraction Procedure I. Reagents Needed 1. 2. Bray P1 Extracting Solution: Add 5 ml of 5.0 N HCl and 12 ml of NH4F per liter of distilled water Ammonium Molybdate: Dissolve 25 g of ammonium molybdate in 212 ml of distilled water and filter. Add this to a solution of 40 ml H O and 425 m1 concentrated HCl. Then add 27.5 of boric acid to suppress flouride interference with color development. Fiske-Subbarow powder: Mix and grind 10 g of l-amino-Z-napthol-4 sulfonic acid (Eastman 360), 20 g of sodium sulfite (NaSO3), and 584 g of sodium meta-bi-sulfite (NaZSZOS). II. Phosphorus Determination Procedure 1 Run standard curve, which is linear from 0-6 ppm P. Weigh 2.5 g of air-dried mineral soil or 1.7 g of air-dried muck soil to acid-washed 125 m1 flasks. Add approximately 1/8 of a teaspoon of low phos- phorus carbon (Darco G-60 from Eastman) to each flask. Add the same amount of carbon to empty flasks for controls. Add 20 ml of Bray P1 extracting solution to each flask. Shake for 4 minutes at 200 cycles/minute. Filter soil solution through Whatman #1 filter paper into large acid-washed test tubes. Refilter if necessary. Add 5 drops of 1 N HCl. Place 7.6 ml of each filtrate into small acid-washed test tubes. Place 7.6 ml of Bray Pl extracting solution into tubes for standards. Add 4 ppm of phosphorus to standards (32 ul of 1000 ppm P solution of KH2P04). Add 200 ul of ammonium molybdate solution to small tubes and vortex mix. Add 0.77 g of the Fiske-Subbarow powder to 5 m1 of warm H20. Vortex mix and add 200 ul to each small tube. Vortex mix again. 10. 187 Read absorbance at 620 nm after 10 minutes. Use 1.0 ml cuvette tubes. If absorbance is greater than about 0.500, repeat steps 5-8 after dilution. Calculate ppm P. a. Calculate correction factor from standard curve. b. Subtract blank value form absorbance and read ppm P from standard curve. c. Multiply by 20 (from 20 m1 of Bray P1 extracting solution), divide by weight of soil used (2.5 or 1.7 g) and multiply by dilution factor. Following is a FORTRAN program used to calculate ppm P from absorbances and dilution factors. PROGRAM STANCV(INPUT,OUTPUT,TAPE1,TAPEZ) INTEGER DF REWIND 1 PRINT*, 'ENTER BLANK VALUE' READ*, BV PRINT*, 'ENTER AVERAGE 4 PPM P STANDARD VALUE' SV=SV-BV C CALCULATE CORRECTION FACTOR FROM STANDARD C CURVE CF=0.377SV 50 READ(1,100,END=200) ABS,DF C SUBTRACT BLANK FROM ABSORBANCE ABS=ABS-BV C APPLY CORRECTION FACTOR ABS=ABS*CF C APPLY REGRESSION EQUATION FROM STANDARD CURVE PPMP=(ABS-0.00253)/0.094 PPMP=PPMP*11.76*DF C PRINT RESULT WRITE(2,101) PPMP GO TO 50 100 FORMAT(F4.3,12) 101 FORMAT(F8.1) 200 CONTINUE STOP END Appendix 5. Commercial onion fields sampled in 1981 188 Field Location Plot Soilx GO Grant, J. Plaisier's field: 1-west end SM across Spruce St. from Ento- l-middle SM mology trailer-eastern half 3-east end SM in onions 61 Grant, P. Brink's field: l-north end SM just west of Entomology 2-middle OM trailer 3-south end OM GB Grant, P. Plakmeyer's field: l-north end OM on north end of swamp off 2-middle OM dirt road 3-south end OM GS Grant, P. Brink's field: 1-south end M north of 120th Ave. and 2-middle M/OM red shed-western half 3-north end OM in onions G7 Grant, P. Brink's field: 1-north end S across Spruce St. from 2-middle SM fieldhouse and south 3-south end SM of drying shed GB Grant, P. Brink's field: l-south end SM adjacent to and southeast 2-middle SM of fieldhouse 3-north end 5 G9 Grant, J. Plaisier's field: l-south end OM n.e. of Spruce St.-120th 2-middle OM Ave. intersection, 2nd field 3-north end OM east of Spruce St. Field Planting Date Cultivar 61 4/20 Krummery Special GS 4/21,4/22 Sentinel G7 4/27 Spartan Banner 68 4/28,4/29 192 x - M = marl, OM = organic muck, S = sandy, SM = sandy muck 189 Appendix 6. Computer files containing onion data Computer files containing onion data can be loaded from the UP tapes, UP1647 or UP3042. These two tapes will be maintained by Dr. G.R. Safir of the Botany & Plant Pathology Department of Michigan State University. Documentation of the data files can be found on the files: CDDAGRANT1980DOCUMENT and CDDAMF1981DOCUMENT. The computer files containing data collected in 1980 and 1981 are : ' 1980 1981 CDDAGRANTONIONDATA1980 CDDAMFONIONDATAIQBI CDDAGRANTONIONDATA19802 CDDAMFONIONDATA19812 CDDAGRANTONIONDATA19803 CDDAMFONIONDATA19813 CDDAGRANTBOMAXMINTEMPS CDDAMFBlMAXMINTEMPS CDDAGRANTSADATA1980 CDDAMFBIDENSITYDATA CDDAROOTDATA1980 CDDAOPONIONDATA CDDABPONIONDATA 190 Appendix 7. Onion simulation program GOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO PROGRAM PDRIVE(INPUT,OUTPUT,TAPE1,TAPEG) ********************************************** VARIABLES USED: PDRIVE BASE — BASE TEMPERATURE BELOW WHICH NO D-DAYS ACCUMULATE CUMDD - CUMMULATIVE DEGREE DAYS DT - TIME INCREMENT DU - DEGREE UNITS FAREA - FIELD AREA (M2) FRMDD - DEGREE DAYS HOUR - (PI)/12, USED IN TEMPERATURE SINE WAVE HRANG - DAILY TEMPERATURE RANGE HTIME — TIME OF DAY (0-24) IDAY - JULIAN DAY IDT - NUMBER OF DT'S IN ONE DAY LEAFAR - LEAF AREA (CM2) Lx - PARAMETER, o=Go THROUGH INITIALIzATION l=SKIP INITIALIZATION NDAYS — NUMBER OF DAYS IN SIMULATION RUN Nx - PARAMETER (SEE Lx) PDATE - PLANTING DATE PDEN - PLANTING DENSITY (PLANTS/M2) PLANTS - NUMBER OF PLANTS EMERGED STEMP - SOIL TEMPERATURE ARRAY STMP - INSTANTANEOUS SOIL TEMPERATURE SUMDU - SUM OF DEGREE UNITS TBGRAM - TOTAL DRY WEIGHT OF BULB (GRAMS/PLANT) TEMP - INSTANTANEOUS AIR TEMPERATURE THETA - ARGUMENT OF SINE FUNCTION TLGRAM - TOTAL DRY WEIGHT OF LEAVES (GRAMS/PLANT) TMAx - DAILY MAXIMUM TEMPERATURE TMEAN - DAILY AVERAGE TEMPERATURE TMIN - DAILY MINIMUM TEMPERATURE TPLANT - TOTAL NUMBER OF PLANTS PLANTED 21 — NUMBER OF HOURS IN ONE DT SUBROUTINES CALLED: PDRIVE ATEMP - READS IN AIR TEMPERATURE FROM TAPE 1 DEGDAY - COMPUTES NUMBER OF DEGREE DAYS ACCUMULATED PER DAY FNL - TABLE LOOK-UP FUNCTION ONION - CUMPUTES ONION PLANT GROWTH SSTEMP - READS IN SOIL TEMPERATURE FROM TAPE 2 *************************************************** 191 INTEGER DAY(20),UPDATE,DC,OLD INTEGER PDATE DIMENSION Dw(20),SD(20),GR(20),DUH(20) REAL LEAFAR(2),TGRAM(2),TLGRAM(2),TBGRAM(2) COMMON/PASSlZ/TEMP,STMP,TPLANT,PDEN,DT,FAREA,IDAY,LEAFAR, +TLGRAM,TBGRAM,PLANTS,Lx,PDATE,CUMDD,TGRAM,SUMDU +,CWSADU,GR,DUH,NPT,DF DATA BASE/5.6/,LX/0/,CUMDD/0./ C *************************************************** REWIND 1 REWIND 6 C *************************************************** PRINT*,'NUMBER OF DAYS IN SIMULATION?’ READ*,NDAYS PRINT*,'WHAT IS THE TIME INCREMENT, DT=?' READ*,DT PRINT*,'WHAT IS THE FIELD AREA (M2)?' READ*,FAREA PRINT*,'PLANTING DENSITY (SEEDS/M2)=?' READ*,PDEN PRINT*,'ENTER NUMBER OF DATES' READ*,NPT PRINT*,'ENTER DEGREE UNITS AT SAMPLING' READ*,(SD(K),K=2,NPT+1) PRINT*,'ENTER DRY WEIGHTS CORRESPONDING TO DATES' READ*,(DW(K),K=2,NPT+1) PRINT*,'ENTER B VALUE FOR DENSITY CORRECTION' READ*,DF C *************************************************** TPLANT=FAREA*PDEN IDT=INT(1./DT+.005) Zl=DT*24. Nx=o HOUR=(22./7.)/12. 00 CALCULATE GROWTH RATE AT 1/2 OF DEGREE UNIT INTERVAL SD(1)=0.0 DW(1)=-6.725 INITIALIZE FIRST DATE PDATE=1 Do 20 I=2,NPT+1 CALCULATE DELTA DEGREE UNITS DELDU=SD(I)-SD(I-1) CALCULATE DEGREE UNITS AT 1/2 INTERVAL DUH(I)=SD(I-1)+0.5*DELDU CALCULATE LN(OBSERVED WTS) Dw(I)=ALOG(Dw(I)) DELDW=DW(I)-DW(I-1) c CALCULATE GROWTH RATE GR(I)=DELDW/DELDU 000000 192 20 CONTINUE B=(GR(3)-GR(2))/(DUH(3)—DUH(2)) GR(1)=GR(2)-B*DUH(2) DUH(1)=0.0 WRITE(6,301) (DUH(JJ),JJ=1,NPT+1) WRITE(6,301) (GR(JJ),JJ=1,NPT+1) 301 FORMAT(10F8.3) C MAIN DAY LOOP Do 1000 L=1,NDAYS HTIME=0. C INPUT TEMPERATURES AND CALC. DDAYS. 10 CALL ATEMP(IDAY,TMIN,TMAX,TMEAN,HRANG) Nx=1 CALL DEGDAY(TMAX,TMIN,BASE,FRMDD) C WITHIN DAY LOOP Do 900 J=l,IDT c COMPUTE INSTANTANEOUS TEMPERATURES CUMDD=CUMDD+FRMDD*DT HTIME=HTIME+21 THETA=(HTIME-9.)*HOUR TEMP=TMEAN+HRANG*SIN(THETA) C COMPUTE ONION GROWTH CALL ONION Lx=1 900 CONTINUE WRITE(6,120) IDAY,CUMDD,LEAFAR(2),TGRAM(2),TLGRAM(2) +,TBGRAM(2),CWSADU,SUMDU,PLANTS 120 FORMAT(I4,F10.2,F9.2,3F8.4,F7.5,F9.3,F9.1) IDAY=IDAY+1 1000 CONTINUE END *************************************************** ******************SUBROUTINES********************** *************************************************** SUBROUTINE ONION ************************************************** VARIABLES USED: ONION BGRAM - CHANGE IN BULB DRY WEIGHT DURING 1 DT DELL - DAYS TO EMERGENCE DF - PLANT DENSITY FACTOR DT - TIME INCREMENT FAREA - FIELD AREA GRAMS - NUMBER OF GRAMS OF DRY WEIGHT ACCUMULATED DURING ONE DT IDAY - JULIAN DAY 0000000000000 00000 193 LEAFAR - LEAF AREA (CM2) LGRAM — CHANGE IN LEAF DRY WEIGHT DURING ONE DT Lx - PARAMETER, 0=GO THROUGH INITIALIZATION l=SKIP INITIALIZATION Mx - PARAMETER (SEE Lx) NSEED - NUMBER OF NON-EMERGED SEEDS PBULB - PERCENT OF DRY WEIGHT GOING To THE BULB PDEN - DENSITY OF PLANTS (PLANTS/M2) PLANTS - NUMBER OF PLANTS PLEAF - PERCENT OF DRY WEIGHT GOING TO THE LEAVES PPDEL — DAYS To EMERGENCE IN THE PREVIOUS DT PRATE - RATE OF PLANT EMERGENCE PSTORE - ARRAY OF NON-EMERGED SEEDS RPLANT - NUMBER OF PLANTS LEAVING THE DELAY TBGRAM — TOTAL DRY WEIGHT OF BULB (GRAMS/PLANT TEMP - AIR TEMPERATURE TLGRAM - TOTAL DRY WEIGHT OF LEAVES (GRAMS/PLANT TPLANT - NUMBER OF PLANTS ENTERING THE DELAY 000000000000000000000 ************************************************** REAL PRATE,PSTORE,PDEN,PLANTS,PPDEL,PBULB,PLEAF REAL LEAFAR(2),Ll,L2,LGRAM,NSEED,LNY,PROOT DIMENSION TGRAM(2),TLGRAM(2),TBGRAM(2) DIMENSION SUMSA(2),DUH(20),GR(20),SEN(2) INTEGER PDATE LOGICAL FLAG DIMENSION PRATE(20),PSTORE(20) COMMON/PASSlz/TEMP,STMP,TPLANT,PDEN,DT,FAREA,IDAY,LEAFAR, +TLGRAM,TBGRAM,PLANTS,Lx,PDATE,CUMDD,TGRAM,SUMDU +,CWSADU,GR,DUH,NPT,DF c*************************************************** C PROGRAM INITIALIZATION IF(Lx.GT.0) GO To 20 PLANTS=0. TGRAM(1)=0. PPDEL=100. DELL=0. RPLANT=0. NSEED=0. xx=2o DO 54 JJ=1,2 TGRAM(JJ)=0.0 TLGRAM(JJ)=0. TBGRAM(JJ)=0. LEAFAR(JJ)=0. 54 CONTINUE SUMDU=0. 0 Mx=o IDAY=0 DO 10 1:1,20 PRATE(I)=0. 10 194 PSTORE(I)=0. CONTINUE C*************************************************** C COMPUTE INSTANTANEOUS EMERGENCE 20 IF(IDAY.LT.PDATE) Go TO 100 CALL EMERGE(EM,DEM,CUMDD) PLANTS=EM*TPLANT HPLANT=TPLANT*0.5 IF(PLANTS.LT.HPLANT) GO To 100 PDEN=PLANTS/FAREA IF(Mx.GT.0) GO To 25 LEAFAR(1)=1.0 TBGRAM(1)=0.0 TGRAM(1)=0.0012 FLAG=.FALSE. Mx=1 C COMPUTE DRY MATTER ACCUMULATION 25 6 CONTINUE OLDWT=TGRAM(1) OLDSA=LEAFAR(1) IF(TEMP.GE.5.6.AND.TEMP.LE.28.0) THEN DU=-0.25+0.04464*TEMP ELSEIF(TEMP.LT.5.6) THEN DU=0.0 ' ELSE DU=1.0 ENDIF DU=DU*DT SUMDU=SUMDU+DU USE LAST DELTA WT/SA AS GROWTH RATE IN THE CASE OF DECLINING LEAF AREA IF(FLAG) THEN GRAMS=CWSADU*LEAFAR(1)*DU TGRAM(1)=TGRAM(1)+GRAMS GO TO 36 ENDIF CALCULATE GROWTH RATES FROM OBSERVED DATA IF(SUMDU.GE.DUH(NPT+1)) THEN TGDUDU5=GR(NPT+1) ELSE TGDUDUS=TABLI(GR,DUH,SUMDU,NPT) ENDIF GRAMS=TGDUDU5*DU*TGRAM(1) TGRAM(1)=TGRAM(1)+GRAMS ADJUST FOR DENSITY OF PLANTS IF(TGRAM(1).GE.0.00.AND.PDEN.GT.10.0) THEN RW=1./TGRAM(1)+DF*(PDEN-10.) TGRAM(2)=1./RW 195 ELSE TGRAM(2)=TGRAM(1) ENDIF C COMPUTE DISTRIBUTION OF DRY MATTER C COMPUTE LEAF, BULB DRY WEIGHTS DO 31 I=1,2 xxx=-o.2425*TGRAM(I) XXX=EXP(XXX) PBTW=-0.7520-2.0383*XXX XXX=-0.l420*TGRAM(I) XXX=EXP(XXX) PLTW=-1.0961+0.7914*Xxx PBULB=EXP(PBTW) PLEAF=EXP(PLTW) TLGRAM(I)=TGRAM(I)*PLEAF TBGRAM(I)=TGRAM(I)*PBULB 31 CONTINUE C DRY MATTER To PLANT PART SIZE CONVERSION C COMPUTE SURFACE AREA DO 32 I=1,2 IF(TLGRAM(I).LT.0.4044) THEN xx1=TLGRAM(I)*(-4.4861) XX1=EXP(XX1) LNY=5.8344+1.2077*XX1 ELSE xx2=TLGRAM(I)*(-o.3574) XX2=EXP(XX2) Y2=5.2307+0.9250*xx2 IF(TLGRAM(I).LT.1.1638) THEN XXl=TLGRAM(I)*(-4.4861) XX1=EKP(XX1) Yl=5.8344+1.2077*XX1 LNY=(Y1+Y2)/2. ELSE LNY=Y2 ENDIF ENDIF Y=EXP(LNY) SUMSA(I)=Y*TLGRAM(I) 32 CONTINUE C COMPUTE SENESCENCE DO 33 I=1,2 IF(SUMSA(I).LT.26.39) THEN LEAFAR(I)=SUMSA(I) ELSEIF(SUMSA(I).LT.161.48) THEN LEAFAR(I)=4.2437+0.8392*SUMSA(I) ELSE LEAFAR(I)=-8.5456+0.9184*SUMSA(I) c COMPUTE SURFACE AREA AFTER TOP FALL IF(CUMDD.GE.1780.) THEN PSA=-0.0889-0.l705*(CUMDD-1780.) PSA=EXP(PSA) 196 LEAFAR(I)=PSA*SUMSA(I) ENDIF ENDIF SEN(I)=SUMSA(I)-LEAFAR(I) 33 CONTINUE C C COMPUTE CHANGE IN TOTAL WT PER LEAF AREA PER DEGREE UNIT IF(LEAFAR(1).LT.OLDSA) THEN FLAG=.TRUE. GO TO 37 ELSE DWT=TGRAM(1)-OLDWT SA5=(LEAFAR(1)-OLDSA)/2.+OLDSA IF(DU.GT.0.0) THEN CWSADU=DWT/SA5/DU ENDIF ENDIF 37 CONTINUE C 100 RETURN END c*************************************************** SUBROUTINE EMERGE(EM,DEM,CUMDD) C*************************************************** IF (CUMDD.LT.57.5) GO To 3 IF (CUMDD.GT.163.) GO To 5 Y=-9.986+(0.0937*CUMDD) EM=EXP(Y) EM=EM/(EM+1.) DEM=EM-PEM PEM=EM RETURN 3 PEM=0.0 DEM=0.0 EM=0.0 RETURN 5 DEM=0.0 EM=1.0 RETURN END *************************************************** *************************************************** SUBROUTINE DEGDAY(XMAX,XMIN,BASE,XHEAT) ************************************************** VARIABLES USED: DEGDAY 00000 000000 00000000 10 30 20 0000000000000 000000 3 01 C) 197 BASE - BASE TEMPERATURE BELOW WHICH NO D-DAYS ACCUMULATE XHEAT - DEGREE-DAYS XMAX - MAXIMUM TEMPERATURE XMIN - MINIMUM TEMPERATURE ************************************************** DATA TPIE/6.283185308/,HPIE/1.570796327/ IF (XMAX.GT.BASE) GO To 10 XHEAT=.00001 RETURN CONTINUE Z=XMAX-XMIN XM=XMAX+XMIN IF (XMIN.LT.BASE) Go To 20 XHEAT=XM/2.-BASE IF (XHEAT.GT.0.) GO TO 30 XHEAT=.00001 RETURN CONTINUE TBASE=BASE*2. A=ASIN((TBASE-XM)/Z) XHEAT=(2.*COS(A)-(TBASE-XM)*(HPIE-A))/TPIE IF (XHEAT.GT.0.) GO To 40 XHEAT=.00001 RETURN END *************************************************** *************************************************** SUBROUTINE ATEMP(IDAY,TMIN,TMAX,TMEAN,HRANG) ************************************************** VARIABLES USED: ATEMP HRANG - TEMPERATURE RANGE IDAY - JULIAN DAY TMAX - MAXIMUM TEMPERATURE TMEAN - TEMPERATURE MEAN TMIN - MINIMUM TEMPERATURE ************************************************** READ(1,50) TMAx,TMIN FORMAT(8x,2F6.0) TMAX=(TMAX-32.)*5./9. TMIN=(TMIN-32.)*5./9. 198 TMEAN=(TMAX+TMIN)/2. HRANG=(TMAx-TMIN)/2. RETURN END C C C C ************************************************ C ************************************************ C FUNCTION TABLI(VAL,ARG,DUMMY,K) C **********ink************************************ C C VARIABLES USED: TABLI c C VAL — Y INPUT VALUES C ARG - x INPUT VALUES C DUMMY - x INPUT C x - N-l SLOPES VALUES C TABLI - SLOPE OUTPUT VALUE g skid”:**~k***it************************************* DIMENSION VAL(1),ARG(1) DUM=AMAX1(AMIN1(DUMMY,ARG(K)),ARG(1)) DO 34 I=2,K IF(DUM.GT.ARG(I)) GO To 34 TABLI=(DUM-ARG(I-l))*(VAL(I)-VAL(I-l))/ +(ARG(I)-ARG(I-1))+VAL(I-1) RETURN 34 CONTINUE RETURN END This program can be accessed as CDPGONIONPLANTMODEL2 (see Appendix 6).