. «i». tar... . 5 ,vv‘. Unzip... ..rl...r’tflc . .53.)... .3! .. ilhg t. . 03551:! ‘0 LV. h... .Vv .39.“. . .- VI... .) 1!. P. Ii' ‘ :I' E”. llclliivl. 1.1.0: .. .2 . T.v...o.l.r.. . 2..-... :- . 2:, .. . I..._....2‘\ . .II. . . ‘ ..'.t‘.b..‘. . l ‘ilfibt"| I . . . :1. .5}. i..- .07 z .2 is! i. 4. )5; 5.1.1.!- o......!. . . l .1 3 . .. iiidiiziz. 1.3L... 2... 1...... fl... 1)... .2... . . 22...... 2.92.... :2... . «2|! in... :9 ‘Oufit‘VD; ‘5')...“ 2. :2: . . 2.... $.55...“ I: r a. :6 51:227.... 511.27.... ‘1-=IV\A\f-\c‘ll A”... .. Jar... ...... 1 : s)...i.... (In? Lv. lyrurtafi. . A $1.. (r! .92.... :4...” .n. .2 1...: VII): . 6...»...- . L lv:rx. .LJ ”HIM“! .ongnlzbvjflanv 0 . 0- ail ‘l.u.'-o>.' 1“?! v‘l‘l..|-Iv¢ ii I': .. 3631.... .3257: .3 it ‘ll‘...§.€j:¢vl§:’l ‘9‘.“t .1 ‘0 [.7‘"... Y b. . I --!T I 2...... 5.; 93:..- . H. 5.2.3.9... .....5..u.rv..\l . . \f. . . .30.... . 6‘15 C. kill}; 0.. 5...: .‘r‘fvnvzi :\¢....2 2 ‘ .27.}... (I... . . . . b.1113. . . . . L , . -15....2'... . a 4.90 :39. .1}.5.._I 50...»): imis. ,3 l: .lr...x... 3.1.. lav)! .3 siftisr I tutti-5“ n v .2 "p .s.‘.-.f‘ .- 4. ivi' .. .111 3.. u 3.40 :x v . {at}... {3.51.1.1 3.»...Q0xzcglfi. it‘}fl.< tn. E‘il“ .y‘i‘l“ 'huut) $1.594? , D245Hix. : .. 0| .3... .. . Di‘il.r\"' ." 4 . . u p 2 o ' [is . (in... :5... II . , .. . . . . , . . r. , ... .2315 l. 21 .12: :295; t 0! .L . , :51}. 5.7... I: (5.1!? av_|§v.§llzé«.r 2...... .. ..:{.r..ix.:2\1u§§ftfil . . . . 15.. ‘1‘...)5353 . . . . . | .9 2.. (5.... I ‘1... :7. 1.2.1.3}... . . .[Ugfifilrtn-Ipr: 2.. .0}... . f! 2 Y. .0 I... . . . f I... 5:. . {5.5.1 5.1.... . .1) .QslouIllJ... I?! Itclvtvz .. .{tii’ff if... S): .. f... 4‘... ..‘|. 1:93;- K (3?)! .lio’l‘ffo Q.- ‘.v1..1.i1..10\!.l.u5 {RE I . . .. i. {"1} 51...! n. I. .2 . :.I 7...... t. \ols..uuh.£ ¢ . . .r .3 A. E a Ktr ll 1’. 2 . as 2.116: 1.1.... :\.:v.. .3! . .1. Atria. 7.23521)». .2. .1!.Ilp; $.\::; in. . 1.5... . ‘14:!) . #1.!!! .I...‘. 21):? . I; ID... [2:321 .II a). a!!! i i PXI‘I..L :A til...\.‘!l~ .1 {I310 . . . . . zn‘ilsfizzgcsly in: .Ellvl\‘7eii . . . .Il!‘€lb“.-t.l3.vi ‘21:!1353Dzi.‘ . . !1 6:33.... ..|i Clutzva) . . . . k. . :1). in. 7:1?2} L‘s Iii . . f ‘v . .. 1..9.v :- . :I I! . .. . . . .rvxtrle 3 23‘ n .. I'IQ .....7 u . .. . . . . . . . . 51.219 v.33!!! ll 21:2. . . 1 . . . :u 1.31xv.,.0\nnr .. . .. . I). .d...’ . . u . . , , . r 2 .2. I up»... V‘I‘Klv-i t - v :- Zuni \L ‘ [uanm . ... . .. \llnibl. {CIIVJ‘fV'lnn‘l't “0031'.” . . . .2. .. .2 2.. v..\§.9 2. I. . .T. .....J::xs.leb .: q. : .1. . . 1.3..L.I.. zEIiSIi .a..l_t!...-\. I .,2.1. Al 1" . » ......:......;.........-.o.....s..:.....,...13.....41......i¥.l _ .. .2 02. {JV-unil‘bll‘l.‘ '0 h [Iii will lllllmmmlgllim 3 1293 00790 LIBRARY Michigan State University This is to certify that the dissertation entitled LEGUME-POTA'I‘O ROTATIONS: NITROGEN MANAGEMENT, ECONOMICS, AND PATHOGENS presented by Timothy Scott Griffin has been accepted towards fulfillment of the requirements for Ph-D- degree in _Cr_o.p_and_5oil Sciences (Q; .1 5; Zéfizmx Major professor Date ?[3//60 MSU is an Affirmative Action/Equal Opportunity Institution 042771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE I m . J L} : EV . 7; it) £1! "’ W ‘- ‘ "—1 MSU Is An Affirmative Action/Equal Opportunity Institution emana-m LEGUME-POTATO ROTATIONS: NITROGEN MANAGEMENT, ECONOMICS, AND PATHOGENS By Timothy Scott Griffin A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 1990 .—-I 4-. 3/ 6% - o? ‘7‘: ABSTRACT LEGUME-POTATO ROTATIONS: NITROGEN MANAGEMENT, ECONOMICS AND PATHOGENS By Timothy Scott Griffin The implementation of crop rotations in potato production requires additional information on the potential agronomic benefits and the economic viability of rotational systems. The objectives of this research were: (i) to quantify legume nitrogen (N) accumulation in the first year of rotations; (ii) to evaluate the vine and tuber response of the subsequent potato crop to legume and fertilizer N; (iii) to compare net income and income variability of alfalfa- and corn-potato rotations and continuous potato using Monte Carlo simulations; and (iv) to evaluate the effect of first year rotation crop and second year N rate on potato pathogen populations and tuber quality parameters. Potato fertilized with 0 to 225 kg N ha‘1 was grown following; alfalfa, birdsfoot trefoil, red clover, sweetclover, hairy vetch, corn, fallow and potato, at the MSU Montcalm Research Farm and the Kellogg Biological Station (KBS), respectively. Seeding year harvest of alfalfa, birdsfoot trefoil and red clover generally did not significantly reduce legume N accumulation, compared to a green manure crop of the same species. The subsequent potato crop exhibited a vegetative response to both legume and fertilizer N sources at both locations. Tuber yield was unaffected by rotation or N rate at KBS. At Montcalm, tuber yield increased with N rate following red clover, corn, fallow, and potato. The lack of N response following the remaining legumes is attributed to a soil moisture deficit induced by spring regrowth of legumes. Root-lesion nematode populations increased under growing legumes crops in Year 1 (1987) of the rotations, but soil populations in May, 1988, were similar for all rotations. Similarly, soil inoculum levels of Verticillium dahliae were unaffected by rotations crop. Potato scab decreased with N rate in 1988, but black scurt‘ and growth cracking increased with N rate. Using state average yields and costs, continuous potatoes had a higher mean net income, with greater variability, than alfalfa- or corn-potato rotations. A 15% reduction in continuous potato yield resulted in net income being nearly identical to that from the alfalfa hay-potato rotation, but with 37% more variability in net income. The implementation of potato rotations depends most strongly on the assumption of declining potato yield in monoculture. ACKNOWLEDGMENTS Over the past four years at Michigan State, there have been many people who have helped me become a better scientist. I owe a debt to my guidance committee, including Drs. George Bird, Dick Chase, and Maury Vitosh, for teaching a forage agronomist about potatoes, and certainly to Dr. Oran Hesterman, my major professor, for being patient while I did so. Dick Kitchen, Jim Bronson, and Brian Graff, the farm managers at Montcalm, KBS, and East Lansing, took all of my odd requests in stride and I appreciate their help. I am also thankful to Fred Warner of Dr. Bird’s lab for performing all of the nematode and fungal assays, and to the Michigan Potato Industry Commission for funding a large portion of this project. I could not have done this research without the help of the undergraduate (Andy,Aaron, Renee and Leslie) and graduate students (Barry, John, and Dana) of the Forage Project. A special thanks go to Glen Harris, for all his help with my research, and all of the tapes he has made for me. To Janice, Justin, and Hannah; I could not have done this without you. iv PREFACE Chapters One and Three of this dissertation are written in the style required for publication in the Agronomy Journal. Chapter Two, written in English rather than metric units, is in the style required for publication in the Journal of Production Agriculture. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES CHAPTER ONE: RESPONSE OF POT‘ATO TO LEGUME AND FERTILIZER NITROGEN SOURCES ABSTRACT INTRODUCTION MATERIALS AND METHODS Legume management Corn and potato management, 1987 Potato management, 1988 Experimental design and statistical analysis RESULTS AND DISCUSSION Forage production by hay crop legumes Nitrogen accumulation by legumes Potato yield without N fertilizer Potato N response in rotation SUMMARY REFERENCES CHAPTER TWO: ECONOMIC COMPARISON OF ALF ALF A- POTATO, CORN-POTATO, AND CONTINUOUS POTATO SYSTEMS ABSTRACT INTRODUCTION MATERIALS AND METHODS Crop prices Crop yields Monte Carlo simulations vi PAGE viii 10 ll 12 13 13 15 18 22 30 32 35 35 37 40 42 45 45 RESULTS AND DISCUSSION Rotational versus continuous potato production Ef feet of decreased continuous potato yield on net income skewness in potato yield distributions CONCLUSIONS REFERENCES CHAPTER THREE: EFFECT OF FORAGE LEGUMES ON PATHOGEN POPULATIONS AND TUBER QUALITY IN LEGUME-POTATO ROTATIONS ABSTRACT INTRODUCTION MATERIALS AND METHODS Crop production practices Soil and plant sampling for pathogens Evaluation of tuber quality RESULTS AND DISCUSSION P. penetrans populations in rotations V. dahliae soil inoculum levels following different rotation crops Tuber evaluation of 1988 potato crop CONCLUSION REFERENCES vii 4‘7 47 50 53 54 57 60 60 62 65 65 66 67 68 68 70 71 75 78 LIST OF TABLES TABLE TITLE PAGE 1.1 Initial soils characteristics at the Michigan State 7 University Montcalm Research Farm and the Kellogg Biological Station. 1.2 Mean monthly precipitation and air temperature at the 8 Michigan State University Montcalm Research Farm and the Kellogg Biological Station (KBS). 1.3 Seeding year dry matter (DM) forage and weed yield 14 and nitrogen (N) nitrogen concentration of alfalfa, red clover, and birdsfoot trefoil harvested for hay at Montcalm and KBS in 1987. 1.4 Nitrogen (N) content in herbage and roots in fall, 1987 16 and spring, 1988, and total plowdown N (PDN) yield of green manure (GM) and hay crap legumes at Montcalm and KBS. 1.5 Effect of 1987 crop on vine nitrogen (N) concentration, 19 and dry matter (DM) and N production of 1988 potato crop, grown without N fertilizer, at Montcalm and KBS in 1988. 2.1 Enterprise budgets for potatoes, corn, and seeding year 41 alfalfa in Michigan. 2.2 Summary statistics of truncated normal yield 43 distributions for potato, corn, and alfalfa hay. 2.3 Simple correlation coefficients (r) between potatoes, 44 corn, and alfalfa hay. 2.4 Prices for potatoes, corn, and alfalfa hay in Michigan, 46 1981-1988 (adjusted to 19883). 2.5 Net income distribution for alfalfa green manure- 48 potato, alfalfa hay-potato, and corn-potato rotations (250 acres of rotation crop, 250 acres of potatoes) and continuous potatoes (500 acres) assuming normal distributions of prices and yields. viii 2.6 2.7 3.1 3.2 3.3 Mean and variability of net income from simulation (n-250) of continuous potato yield ranging from 225 to 300 cwt/a (potato yield normally distributed, coefficient of variation-124% for all yield levels). Effect of potato yield distribution on mean and variability of net income (mean yield-270 cwt/a for all distributions). Effect of rotation crops on population densities of Pratylenchus penetrans and Verticilliurn dahliae in 1987- 1988 at the Michigan State University Montcalm Research Farm in Entrican, MI. Influence of 1987 rotation crop on severe potato scab and black scurf, and on specific gravity of subsequent (1988) potato crop at the Michigan State University Montcalm Research Farm in Entrican, MI. Ranking of rotation crops based on population of root- lesion nematode (November, 1987), and severity of common potato scab and black scurf at the Michigan State University Montcalm Research Farm in Entrican, MI. ix 52 55 69 72 76 FIGURE 1.1 1.2 1.3 1.4 1.5 3.1 LIST OF FIGURES TITLE Effect of N fertilizer rate on late-season (August 25, 1988) potato vine content following: alfalfa and birdsfoot trefoil hay (mean response, ALF-BGM); red clover, birdsfoot trefoil hay, corn and potato (mean response, RC-BHY-C-P); and sweetclover (SWC) at Montcalm. Effect of N fertilizer rate on late season (August 30, 1988) potato vine N content following ’NITRO’ alfalfa, red clover and alfalfa green manure (mean response, RC-AGM), birdsfoot trefoil green manure (BFTGM), FALLOW, and corn and potato (mean response, CRN- POT) at KBS. Effect of N fertilizer rate on cull tuber yield following sweetclover (SWC), other legume craps (OTHER LEG; mean of alfalfa, birdsfoot trefoil, hairy vetch and red clover) and non-legume crops (NON-LEG; mean of corn, fallow, and potato) at Montcalm. Effect of N fertilizer rate on total potato tuber yield following corn (CRN), fallow (FAL), potato (POT), and red clover (RC) at Montcalm. Effect of N fertilizer rate on marketable tuber yield following corn (CRN) and all other crops (mean response, OTHER) at Montcalm. Effect of N fertilizer rate on (A) severe scab, (B) black scurf, (C) growth cracking, and (D) specific gravity of ’A’ grade tubers at Montcalm. PAGE 23 25 27 28 29 74 CHAPTER ONE RESPONSE OF POTATO TO LEGUME AND FERTILIZER NITROGEN SOURCES ABSTRACT The production of potato (Solanum tuberosum L.) in rotation with nitrogen (N)-fixing legumes requires additional information on the N contribution from legumes and on the management of potato N fertilization following legumes. The objectives of this research were to: (i) quantify N accumulation by green manure and hay crop legumes in Year 1 of a 2-yr rotation; (ii) determine the effect of legume species and management on the vine and tuber yield of the subsequent potato crop grown without N fertilizer; and (iii) evaluate the N fertilizer response of potato following legumes, corn (Zea mays L.), fallow, and potato. Rotations were estab- lished at two Michigan locations in 1987; the Montcalm Research Farm at Entrican and the Kellogg Biological Station (KBS) at Hickory Corners. Soil types were McBride sandy loam (coarse-loamy, mixed, frigid, Alf ic Fragiothods) at Montcalm and Oshtemo sandy loam (coarse-loamy, mixed, mesic, Typic Hapludalfs) at KBS. First year crops included: ’Saranac’ alfalfa (Medicago saliva L.), ’Viking’ birdsfoot trefoil (Lotus cornicularus L.), and ’Michigan Mammoth’ red clover (Trifolium pratense L.) as green manure (0 harvests) and hay crops (2 to 3 seeding year harvests); non- dormant ’Nitro’ alfalfa (KBS only) as hay; sweetclover (Melilotus spp.) and hairy vetch (Vicia villosa Roth) green manure; ’Pioneer 3737’ corn; fallow; and potato (’Shepody’ at Montcalm, ’Atlantic’ at KBS). Potato (Shepody) fertilized with O, 75, 150, or 225 kg N ha'1 was grown in Year 2 (1988). Alfalfa and red clover dry matter 2 yield was similar at Montcalm (6.2 and 5.7 Mg ha“, respectively), but ’Saranac’ alfalfa yielded more than ’Nitro’ alfalfa and red clover at KBS. Seeding year harvest of alfalfa, birdsfoot trefoil and red clover did not reduce plowdown N yield [PDN-fa11(herbage+root)N + spring herbage N] compared to a green manure crop of the same species, except for a 28% reduction for red clover at Montcalm. Periodic harvesting also reduced weed competition with alfalfa and red clover. Tuber yield (without N fertilizer) of the subsequent potato crap was not affected by rotation at either location. However, late season vine N content was 93 (Montcalm) and 111% (KBS) higher following legumes than following non-legumes and was significantly correlated with the PDN yield of the preceding legume. This indicated that N mineralized from legume material was available for vegetative growth late in the season, but may not have been available during tuber initiation to affect tuber yield. Tuber yield did not respond to N rate in any rotation at KBS, attributed to a soil moisture deficit (even with irrigation) during the first half of the growing season. At Montcalm, tuber yield was not affected by N rate following alfalfa, birdsfoot trefoil, sweetclover and hairy vetch, with an average yield of 30.9 Mg ha“. Based on this mean yield, these legumes replaced 69 kg N ha", compared to the response in the corn-potato rotation. Total tuber yield increased linearly with N rate following fallow and potato at Montcalm; following corn and red clover, maximum tuber yield required 150 kg N ha". INTRODUCTION Many crops, including potato, have been shown to benefit from crop rotation, due to both nitrogen (N) and non-N effects. Although recent research on producing agronomic crops like corn in rotation has focused on the N contribution from a preceding legume crop (Adams et a1., 1970; Hesterman et a1., 1986; Bruulsema and Christie, 1987; Harris and Hesterman, 1990), research on potato rotations has focused primarily on non-N effects, including pathogen or disease control (Curl, 1963; Brodie et a1., 1969) and improvement of soil physical properties (Odland andSheehan, 1957; Saloman, 1962). Little information is available on the potential for supplying all or a portion of the potato N requirement, or on the management of potato N f ertiliza- tion, when an N-fixing legume is included in a short-term rotation with potato. An accurate quantification of the N contribution from different legume species, and the effect of harvest management strategy (ie. green manure versus hay crops) on this contribution, is needed to more efficiently manage N fertilizer in legume-potato rotations. Nitrogen management is particularly important in legume-potato rotations because potatoes often require high rates of N (from 175 to 448 kg N ha“) to maximize marketable tuber yield (Dubetz and Bole, 1975; O’Sullivan, 1978), yet are adversely affected by insufficient or excess soil N during the growing season. Low levels of soil N early in the growing season reduce the number of tuber initiates (Roberts et a1., 1982), while high soil N levels delay tuber initiation and promote excessive vine growth (Kleinkopf et a1., 1981; Lauer, 1986; Ojala et a1., 1990), thereby reducing tuber yield and increasing the proportion of immature tubers. 4 Total N accumulated by legumes during the seeding year, from both biologi- cal fixation and uptake from soil, is affected by legume species and harvest manage- ment. Nitrogen content at plowdown can vary widely among legume green manure species, with previous estimates ranging from 9 (sweetclover; Sprague, 1939) to 357 kg N ha'1 (red clover; Groya and Sheaf f er, 1985). Recent estimates of N accumula- tion by alfalfa and red clover green manure, the forage legumes most widely used in rotations in the northern US. and southern Canada, have generally been between 125 and 225 kg N ha‘1 (Groya and Sheaf f er, 1985; Bruulsema and Christie, 1987). The effect of harvest frequency and timing on N accumulation by these two legume species has been variable, depending in part on the weed control methods used. Stickler and Johnson (1959) used no herbicides and found that alfalfa and red clover contained more N when clipped once or twice during the seeding year than if they were unharvested, attributed to a reduction in weed competition from periodic clipping. Groya and Sheaf f er (1985) used triflualin (a,¢,a-trif1uoro-2,6-dinitro-N,N- dipropl-p-toluidine) to control weed competition in seeding year alfalfa and red clover. They found that harvesting once during summer (herbage regrowth, roots, and crowns incorporated in the fall) or during both summer and fall (only roots and crowns incorporated in the fall) reduced the amount of N in alfalfa and red clover at plowdown by 20-70%, compared to green manure crops. Total N accumulation by biennial sweetclover, when used as a green manure or if harvested only once during the seeding year, generally exceeds that of alfalfa and red clover (Kroontje and Kehr, 1956; Bowren et a1., 1969; Groya and Sheaf f er, 1985; ), accumulating up to 348 kg N ha'1 (Groya and Sheaffer, 1985). The measurement of legume N accumulation, although useful for comparing legume species or management strategies, is not a direct estimation of the N con- tribution from a legume to a subsequent non-legume crop. In some cases, the amount 5 of N accumulated by legumes has little, if any, positive relationship to dry matter yield or total N uptake of the following crop (Kroontje and Kehr, 1956; Hoyt and Leitch, 1983; Hesterman et a1., 1986; Bruulsema and Christie, 1987). The value of legumes as an N source to a subsequent crop in rotation is more commonly estimated as an N fertilizer replacement value (FRV), defined as the "quantity of fertilizer N required to produce a yield in a crop that does not follow a legume that is identical to that produced by incorporation of the legume" (Hesterman, 1988). Estimates of FRV’s from extensive research on legume-corn systems range from 13 (red clover; Fribourg and Bartholomew, 1956) to approximately 200 kg N ha‘1 (’MN ROOT N’ alfalfa; Hesterman et a1., 1986). There are no FRV estimates available specifically for potato following a legume, although examples of potato yield increases of 10-50% following legumes, compared to potato monocultures or non-legume rotations, are not uncommon (Purvis and Blume, 1939; Wheeler, 1946; Terman, 1949; Murphy et a1., 1967; Emmond and Ledingham, 1972). In Michigan, the "N credit” (analogous to FRV) for legumes preceding potato is adapted from the estimate for corn, ranging from 60 to 100 kg N ha'1 in proportion to the percent stand of the legume crop (Vitosh, 1990). Research is needed to assess the contribution of N from legumes to a subse- quent potato crop and to evaluate potato N fertilizer management following legumes. The specific objectives of this research were: (i) to compare N accumulation by forage legumes used as either green manure (no harvest during the seeding year) or hay (harvested 2 or 3 times during seeding year); (ii) to evaluate the effect of first year legume species and management and N accumulation on tuber and vine yield parameters of a subsequent potato crop grown without N fertilizer; and (iii) to determine N fertilizer response of potato following legumes, corn, fallow, and potato. MATERIALS AND METHODS Two-year crop rotations were established in 1987 at two Michigan locations; the Michigan State University Montcalm Research Farm at Entrican, and the Kellogg Biological Station (KBS) at Hickory Corners. Soil types were McBride sandy loam (Alf ic Fragiothods) at Montcalm, and Oshtemo sandy loam (Typic Hapludalfs) at KBS. At Montcalm only , K20 was applied at a rate of 170 kg ha'1 to all plots prior to beginning the experiment according to soil test results (Table 1.1). Subsequent soil tests indicated that no additional P or K was required at either location for the duration of the experiment. First year crops included: ’Saranac’ alfalfa, ’Michigan Mammoth’ red clover, ’Viking’ birdsfoot trefoil, ’common’ sweetclover, spring seeded ’common’ hairy vetch, ’Pioneer 3737’ corn, potato (’Shepody’ and ’Atlantic’ at Montcalm and KBS, respectively), and potato followed by f all-seeded hairy vetch . ’Nitro’ alfalfa replaced sweetclover at KBS, and an uncropped fallow treatment was . included at both locations. Due to drought conditions early in the 1987 growing season (Table 1.2), irrigation (80 mm) was applied as a rescue treatment at both loca- tions in late June; crops were otherwise grown under non-irrigated conditions during 1987. W Legumes were seeded into 7.6 by 15.2 m plots, using a grain drill (11 rows, 0.18 m row spacing) with a legume seed box and packer wheels, on April 18 and 28, 1987, at Montcalm and KBS, respectively. Fall-seeded hairy vetch (following potato) was planted on September 20 and 21, 1987, at Montcalm and KBS, respectively. Seeding rates were: birdsfoot trefoil, 6 kg ha“; alfalfa, red clover, and sweetclover, 15 kg ha“; and hairy vetch, 31 kg ha“. Two plots per Table 1.1. 1nitiel soil eharaeterietiea'l' at the Itiehigan State University Normal. Research Fan and the Kellogg Biological Station ((88). Bray-1 Exehangeabl e Organi e Location P K Ca Total 11 matter CEC pH --------- kg ha"1 --------- ------ g lrg'1 ----- enol lrg'1 Montcalm 579 311 849 5.9 15 5 5.9 K83 315 569 2240 8.7 16 9 6.6 1’ supine in April, 1937. Table 1.2. Mean aonthly precipitation and air twerature at the Michigan State University liontealn Research Fern and the Kellogg Biological Station (KBS). Precipitation . Ai r Tamerature Long-tern Long-tern Location lionth 1987 was aeanT 1987 1933 neanT .......... m .----..--- ...--.----- °c .---.---.- Montcalm April 46 46 82 10 6 8 May 49 13 72 17 16 14 Jute 21 14 86 20 20 20 July 46 61 63 24 23 22 August 245 86 96 20 23 21 Septewer 83 134 78 17 16 17 TOTAL 490 354 477 KBS April 60 77 88 11 9 9 May 34 36 79 17 17 15 late 51 37 105 23 22 20 July 65 106 85 24 25 22 August 176 127 89 22 24 21 Septeaber 132 163 75 18 18 18 TOTAL 518 546 521 T Long-tern (1950-1981) name from National Heather Service Stations at Greenville, M1, and hickory Corners, 111, for Honteeln and KBS, respectively. 9 replication of alfalfa, red clover, and birdsfoot trefoil were planted. One was managed as a green manure crop not harvested during the seeding year. The other was managed as a hay crop, harvested two (birdsfoot trefoil) or three (alfalfa, red clover) times during the seeding year. ’Nitro' alfalfa at KBS was managed as a hay crop, and sweetclover (at Montcalm) and hairy vetch were managed as green manure. No pre-plant incorporated herbicide was used for legume seedings at either location. Broadleaf weeds in alfalfa, red clover, and birdsfoot trefoil were controlled with a single application of 2,4-DB [4-(2,4-dichlorophenoxy)butanoic acid] at 1.12 kg a.i. ha'l, applied when the legumes were at the first trifoliolate stage. Legumes managed as hay crops were harvested with a f lail-type plot harvester. Fresh forage yield was determined by harvesting a 1.2 by 15.2 m strip from the center of each plot, and weighing the total sample. Dry matter content of forage was determined by drying a subsample (approximately 500 g), for 4 days at 60C. At harvest, random grab samples were clipped at ground level and separated into forage legume and weed components, to determine the proportion of pure forage legume in the harvested sample. Because of the high proportion of weeds in some treatments, forage and weed yields are reported separately. Alfalfa and red clover were harvested on July 10, August 14 and October 16 at Montcalm, and on July 11, August 13, and November 5 at KBS. Birdsfoot trefoil was harvested on July 20 and October 16 at Montcalm, and July 11 and November 5 at KBS. Spring regrowth of hay crop legumes was not harvested as hay prior to incorporation via moldboard plowing in May, 1988. Dry matter and N content of legumes at the end of the seeding year was estimated by sampling four 0.33 by 0.33 m quadrats plot'1 (after a killing frost) on October 14, 1987, at Montcalm and October 29, 1987, at KBS. Growth during the interval between biomass sampling and hay harvest (2 days at Montcalm, 6 days at 10 KBS) was considered to be negligible, because a killing frost had occurred at both locations. The N yield of the hay removed was subtracted from the pre-harvest legume N content. Plants were excavated to 0.25 m (plow depth) and separated into herbage and roots plus crowns. Roots were washed thoroughly to remove soil, and all plant material was dried for 4 d at 60C and weighed. A11 plant material was ground to 1 mm in a cyclone mill. Total N content of herbage and roots plus crowns was determined by micro-Kjeldahl digestion of 0.10g plant material in 12M H,SO, (catalyzed by 1.53 K280, and 0.0753 Se), followed by colorimetric determination of NH4+ by Lachat Quik-Chem Method No. 10-107-06-2-E (Lachat Chemicals, Inc., Mequon, WI.). Legumes were sampled in the same way on May 3 (Montcalm) and May 5 (KBS), 1988, immediately prior to plowing, to determine the dry matter and N content of spring regrowth. Root N content declined an average of 15% at Montcalm and 24% at KBS through the winter. Thus, plowdown N (PDN) yield, defined as PDN-fall 1987 (herbage N + root N) + spring 1988 (herbage N) [1] was used here as a measure of the minimum amount of N accumulated by legumes prior to incorporation. WM First-year corn and potato were planted on May 5 and 12, 1987, at Montcalm and KBS, respectively. Corn was planted at 60,000 seeds ha", without herbicide or starter fertilizer, and was cultivated twice to control weeds. Corn whole plots (7.6 by 15.2 m) were split into four subplots (3.8 by 7.6 m), and fertilized with 0, 75, 150, or 225 kg N ha'1 (as ammonium nitrate), broadcast after planting. In October, grain was harvested from a 7.0 m section of the center two rows in each four row subplot. Stover was removed from harvest area, weighed, and five plants were taken as a subsample. After drying and grinding ll subsamples, total N content of stover and grain was determined by micro-Kjeldahl digestion as described above. Potatoes were planted in 0.92 m rows, with 0.25 m spacing between seed pieces weighing approximately 1003. Aldicarb {2-methyl-2-(methylthio)propana1-0- [(methylamino) carbonyl]oxime) was applied at planting at 3.36 kg a.i. ha‘1 to control nematodes. Weeds were controlled with metolachlor [2-chloro-N-(2-ethyl-6- methylphenyl)-N~(2-methoxy-1-methylethyl)acetamide] and metribuzin [(4-amino-6- (l,l-dimethylethyl)-3-methylthio)-l,2,4-triazin-5(4H)-one], applied preemergent at 2.24 and 0.56 kg a.i. ha“, respectively. The same split N rates as applied to corn were broadcast on potatoes prior to emergence. Yield was determined by harvesting the center two rows of each subplot on September 1 and 4 at Montcalm and KBS, respec- tively. W Following moldboard plowing in early May, 1988, potato (’Shepody’) was planted in all whole plots on May 5 at KBS, and during May 12-14, at Montcalm. Row spacing, spacing of seed pieces, and pesticide applications (ie. aldicarb, metolachlor and metribuzin) were identical to those in 1987. All whole plots were split into subplots fertilized with the same N rates used in 1987. For subplots receiving N fertilizer, 75 kg N ha'1 was applied at planting, and the remainder, if any, was sidedressed prior to hilling. Irrigation (250 mm at KBS, 350 mm at Montcalm) was applied based on precipitation and evapo-transpiration (Vitosh, 1984). At Montcalm, irrigation began on June 3, and 18 to 20 mm was applied every 5 to 7 d until early August. At KBS, irrigation (25 to 35 mm application“) commenced on June 11, and was applied every 7 to 9 d, except for a 12 d period in late June when the irrigation well was broken. Foliar fungicide and insecticide were applied as needed during the growing season. Potato vine growth was sampled on August 25 and 30 (90 to 95 days after emergence) at Montcalm and 12 KBS, respectively, because of clear visual differences in vine growth following different rotation crops. In addition, previous research has shown that maximum vine DM and N content generally occurs between 80 and 100 d after emergence (Jackson and Haddock, 1959; Saf f igna and Keeney, 1977). Vines from four hills were cut at soil level and weighed: A subsample was dried, ground to 1 mm, and total N content was determined by micro-Kjeldahl digestion. Dry matter and N content of vines are reported based on mean stand counts of 42,500 and 40,700 hills ha'1 at Montcalm and KBS, respectively. The center two rows were harvested from each subplot during September 30-October 5, 1988 at Montcalm. Harvest was delayed until November 15, 1988, at KBS because of high rainfall (300 mm) during Septem- ber and October, 1988. Thus, one row per plot was harvested following only alfalfa green manure, alfalfa hay, red clover hay, spring seeded hairy vetch, corn, and potato. Tubers were graded by hand into; ’A’ grade (114 to 281 g), ’B’ grade (<1l4 g), oversize (>281 g), and cull (second growth, soft rot, etc.). W3. Experimental design was a randomized complete block with four replications. Treatments were arranged in a split-plot design; 1987 crops were whole plots, and N fertilizer rates were sub-plots. Significant differences (P<0.05) in N accumulation by legumes, as indicated by analysis of variance, were separated using Least Significant Difference (LSD; Steel and Torrie, 1980). Where possible, significant effects were further partitioned using single degree of freedom orthogonal contrasts. The cf f ect of fertilizer N rate on potato vine and tuber yield parameters was first evaluated within each rotation using single degree of freedom orthogonal contrasts (linear, quadratic, and cubic); significant responses to N rate were further evaluated by regression of potato yield (vine or tuber) on N rate. If the effect of rotation crop was not significant in the analysis of variance, mean N response across rotations was evaluated as described. RESULTS AND DISCUSSION W3, Forage dry matter (DM) yield and N concentration (total N production/tatal DM production) of alfalfa (’Saranac’ at Montcalm; ’Saranac’ and ’Nitro’ at KBS) and red clover were significantly higher than birdsfoot trefoil on common harvest dates and for total seeding year production (Table 3). Because of the poor competitive ability of birdsfoot trefoil in the seedling stage and the lack of chemical weed control, weed competition at both locations was severe; about 80% of the total biomass at the fall harvest on birdsfoot trefoil plots was composed of weeds. At Montcalm, the total DM production and weighted N concentration for seeding year alfalfa and red clover were not significantly differ- ent, although the DM yield of alfalfa was lower for the first harvest (July 9) and higher for the second harvest (August 13), compared to red clover (Table 1.3). The total DM yield of alfalfa (mean of ’Saranac’ and ’Nitro’) at KBS was significantly higher than red clover (6698 vs. 6154 kg ha‘l). Hoyt and Leitch (1983) and Groya and Sheaf f er (1985) also found that the total seeding year hay yield of alfalfa generally exceeds that of red clover. ’Saranac’ alfalfa yielded more than the nondormant ’Nitro’ alfalfa, in contrast to Hesterman et a1. (1986) and Groya and Sheaf f er (1985), who found that yields of dormant and non-dormant alfalfa varieties were similar. Both of these studies also reported higher forage yields for alfalfa than were observed here. In our research, soil moisture may have been limiting during the seeding year, even with supplemental irrigation in late June. Total precipitation plus irrigation from April through July, 1987, was only 80% of the long-term mean. 13 14 ...... .8... a. .5522. .8 . a. 59:08.8 :8... 2. .8... .3... c. 28:22. ........ m: as ea. co.uecaeeucou : oouco_o: .euop . . .5. Eu; .8 .33 m8. 2. ... 822.888 2 8220: .33 e. ... 3.; .3 .88 3.38.. ..-....................-.......... Exam ..................-..........--.. .oeu_z. .a> .oecaeam. eo>o.o one .¢> a..a..< ..oLocu uooenoc.n .a) eo>o.u not + e..a..< ..—~ 0.3 can Non new $3 :3 £3 «.59.: .36— :m o o o poo: 0.8 5.3 o.o~ 0.3 mt. E2... 33 m~n~ .303 n ... a: men one no...) --- m.~n o._.n Gen ... 3n~ S- «RN 959.. .4. ~63 Eb 3~ 08 too: p.- «.3 HQ... n.~n no E, 35 oon~ 8.63 — m8. p.94. inn ..- 9mm .3» Sea --- 33 .303 .33 o3 o --. o to»: o..- “.3 ... 0...... 3~ em- ..- no- 833.. n -.. s~o -.- nmo too: .-- 7mm ..- ~.S .-. e8— --- mon— ‘63 ~ 32.. S. ... 3m one: 5.3 7:” --- o.~n A3 e-~ .-- 38 «59.. P 5865: ....-......... w.ox o ...........-.... --.-...-...-.. —.ac ax ..--....-...-.- ..oeoc» cosodu ......e aedoe.a ._o~ocu eo>o.o 6...».a aLEaeda neoconsou uuo>ea= eo_uaooa uoowuoc_o nan .ocu_z. .ueeaeom. uooemuc_u one .oeu_x. .oeeaeem. eo_uaeueoueou z ensued v.0.» to Koo. c. man be. 3.35: we >2. ..3 voyage... :3...» 333...: u:- 826.9 was 33.3. *o 532E083 23 5995.: use 32> veer tee 39.3 2.3 .3qu Cu 53> 2:83 .n.— 63a» 15 WW Among green manure craps at Montcalm. sweetclover and red clover had similar PDN yields (238 and 225 kg N ha“, respec- tively), which were significantly higher than either alfalfa or birdsfoot trefoil (Table 1.4). Spring-seeded hairy vetch accumulated nearly as much N (211 kg ha‘l) as sweetclover or red clover, despite experiencing complete winterkill after the seeding year, with 93% of the N accumulated in the 1987 herbage. (Fall-seeded hairy vetch was not successfully established at Montcalm, and thus was eliminated from the analysis.) At KBS, red clover green manure accumulated more N than hairy vetch (spring- or f all-seeded) and birdsfoot trefoil. Fall-seeded hairy vetch, which is commonly used as a winter cover crop in the southern U.S. (eg. McVay et al., 1990), accumulated only 33 kg N ha“1 at KBS, mostly in the spring herbage. This minimal accumulation of N, and the unsuccessful establishment at Montcalm, were a result of the late seeding date for f all-seeded hairy vetch. Further research in Michigan has shown that hairy vetch should be seeded prior to August 15 to ensure survival through the winter (Hesterman and Harris, 1990; unpublished data). This early seeding requirement makes it difficult to use hairy vetch as a cover crop in a cropping system following potato or corn. Estimates of N accumulation by green manure legumes in short rotations have been based largely on seeding year N accumulation sampled prior to fall plowing (Kroontje and Kehr, 1956; Stickler et al., 1958; Bruulsema and Christie, 1987), or in some cases, on N content at incorporation for overwintering legume cover crops (Fleming et al., 1981; McVay et al., 1989). In our research, N content at either fall sampling or pre-plow sampling alone underestimates the amount of N accumulated by the legumes. So although it is difficult to directly compare the plowdown N yield of legumes, as defined here, with previous research, seeding year N accumulation (fall herbage plus fall roots; Table 1.4) can be compared with previous estimates. 16 Table 1.6. Hitrogen (H) content in herbage and roots in fall, 1987 and spring, 1988, and total plowdown M (POM) yield of green eenure (GM) and hey crop legunee at Montcale and KBS. M content Location Legume/management Fell herbage Fell roots Spring herbage Spring roots PDNT -------------------------- kg ha’1 °'------------'-~------~- Montcalm Alfalfa GM 68 75 66 51 189 Alfalfa hay 51 70 51 50 172 Red clover GM 127 58 60 69 225 Red clover hey 53 76 36 51 163 Birdsfoot trefoil GM 29 26 17 28 70 Birdsfoot trefoil hay 21 18 10 17 69 Sueetclover GM 93 96 68 90 Z38 Hairy vetch GM 197 16 O 0 211 K83 Alfalfa GM 39 71 68 68 158 Alfelfe hay 61 109 80 82 230 'Hitro' alfalfa hay 26 92 68 80 166 Red clover GM 105 67 50 68 202 Red clover hay 37 96 98 86 229 Birdsfoot trefoil GM 20 16 16 22 68 Birdsfoot trefoil hay 15 18 15 18 68 Hairy vetch (spring) GM 105 11 17 ~ 11 133 Hairy vetch (fall) GM 2 1 20 12 33 LSOO_05 29 19 29 33 56 ifiPDH I fell (herbage H + root H) + spring herbage H. l7 Nitrogen content of alfalfa green manure at the end of the seeding year was similar to that of ’lndian’, ’African’, and ’Ranger’ alf alf as established under similar conditions (Stickler ct al., 1958). It was considerably lower, however, than recent estimates by Groya and Sheaf f er (1985) and Bruulsema and Christie (1987), where pre-plant herbicides were used to control weeds. Nitrogen accumulation by red clover and sweetclover (at Montcalm) was higher than reported by Stickler and Johnson (1959) and Stickler et a1. (1958), but lower than from Groya and Sheaf f er (1985), who found 224 and 348 kg N ha", respectively, in these two species at the end of the seeding year. Seeding year harvest of birdsfoot trefoil (at both locations), alfalfa at Montcalm, and red clover at KBS did not significantly affect PDN yield, compared to the green manure crop of the same species. This information, coupled with the reduction in weed biomass at successive harvests of alfalfa and red clover (Table 1.3) shows that periodic clipping of green manure craps (herbage returned to soil) may reduce weed competition without reducing N accumulation (as shown by Groya and Shcaf f er, 1985). This type of clipping management of green manure crops is particularly applicable to short rotations where the expense of herbicides may not be justifiable. The weed biomass data in Table l.3 also show that the harvest in early August eliminated weed competition from the final fall harvest of alfalfa and red clover. Seeding year harvest of red clover at Montcalm did reduce the PDN yield from 225 to 163 kg ha‘l, due primarily to removal of herbage in the fall (1937) harvest. Bruulsema and Christie (1987) found that a single summer harvest of alfalfa or red clover during the seeding year did not significantly affect the amount of legume N incorporated in the fall (compared to the unharvested treatment), because adequate time was given for regrowth prior to plowdown. The removal of l8 alfalfa or red clover herbage in the fall, however, has been shown to reduce the amount of N incorporated (fall plowing) compared to green manure or summer harvest only (Groya and Sheaf f er, 1985; Hesterman et al., 1986). At KBS, ’Saranac’ alfalfa as hay had a higher PDN yield than either ’Nitro’ alfalfa as hay or ’Saranac’ as green manure (Table 1.4). All hay crop legumes at KBS had a higher root N content in the fall of the seeding year than the green manure crop of the same species, although ’Saranac’ alfalfa was the only species to exhibit a significant difference. Stickler and Johnson (1959) also saw an increase in fall root N content for alfalfa that had been clipped during the seeding year, which they attributed to a reduction in weed competition. Similarly, the increase in fall root N of alfalfa at KBS may be due, in part, to reduced weed pressure (although end-of- season weed biomass was not determined for green manure crops). W The effect of previous crop on the N concentration, DM content and N content of the vines of the subsequent potato crop was significant (P<0.05) at both locations. Single degree of freedom orthogonal contrasts were utilized to partition the general effects of rotation (’rotation vs. potato’; Table 1.5), legumes (’legume vs. non-legume’), and legume management (’GM vs. hay legumes’) on these vine yield parameters. Vines accumulated more DM in the rotations than in the potato-potato system at both locations, and more N in the rotations at KBS. This was due to the promotion of vine DM accumulation in the legume-based rotations. The M vine N concentration, DM content, and N content following legumes was significantly higher than the m values following non- legumes. In some cases (eg. sweetclover), N content was 2 to 3-f old higher following a legume than a non-legume. However, there were exceptions to this generalization. For example, vine N content following red clover green manure and hay and birdsfoot trefoil hay at Montcalm (N content-33, 28, and 23 kg ha“, 19 Table 1.5. Effect of 1987 crop on vine nitrogen (H) concentration, and dry eatt (0M) and H production of 1988 potato crop, green without H fertilizer, at Montcale and KBS in 1988. Vine H Location 1987 crop concentration Vine 0M yield Vine H yield g kg'1 ------------ kg ha‘1 -------------- Montcalm Alfalfa GM: 25.2 2307 57 Alfalfa hay 28.1 1767 69 Red clover GM 26.0 1632 33 Red clover hay 22.8 1229 28 Birdsfoot trefoil GM 23.3 1891 65 Birdsfoot trefoil hay 19.9 1182 23 Sueetclover GM 27.6 3186 95 Hairy vetch GM 27.9 3022 87 Corn 21.0 1061 21 Potato 20.7 1626 30 Fallow 20.2 1665 30 KBS Alfalfa GM 27.6 2379 65 Alfalfa hay 33.0 3616 116 'Hitro' alfalfa hay 25.7 2861 73 Red clover GM 28.5 2895 82 Red clover hay 30.0 3587 106 Birdsfoot trefoil GM 26.3 2031 50 Birdsfoot trefoil hay 25.2 3006 76 Hairy vetch GM (spring) 26.8 2606 70 Hairy vetch GM (fell) 26.6 1959 50 Corn 26.6 1795 65 Potato 26.5 1615 60 Fallow 25.0 2510 61 rth r t Vine H concentration Vine 0M yield Vine H yield ------------------------ pg>r -----------------------~- Montcale Rotation vs. cont. potato HS ' HS Leguaes vs. non-legumes * " *' GM v. hay legumes HS '*' '* KBS Rotations vs. cont. potato HS ' ** Legules vs. non-legumes ' *9 *** GM vs. hay legumes ' HS " "' *, **, *** Significant at 0.05, 0.01, and 0.001, respectively. HS 8 Hot significant at 0.05 level. I’ Sampled on August 25 and 30, 1988, at Montcalm and KBS, respectively. GM 8 green manure, not harvested in 1987. 20 respectively) was similar to that following non-legumes (mean N content- 27 kg ha“). The vegetative response of potato to legume N has not previously been documented. The promotion of vegetative growth late in the growing season following legumes such as seen here has been observed when excess fertilizer N was available during tuber bulking and maturation. Ojala et al. (1990), in a review of potato N and irrigation management, stated that ”excessive nitrogen fertilizer application during tuber bulking can promote late-season vegetative growth and delay tuber maturity.” Lauer (1985) examined N uptake by potato receiving from 150 (”deficient") to 610 (”excessive”) kg N ha'l, and found that the excessive N rate promoted both vine dry matter and N accumulation later in the growing season, although the effect on tuber yield was not discussed. Direct measurements of soil inorganic N level late in the growing season, or of N mineralization rates within different rotations, were not determined in our research. However, the vegetative response of potato to legume is clear. Potato vine N content and the PDN yield of the preceding legume crops were significantly correlated at both locations, with correlation coefficients (r) of 0.41 (Pa-0.01, n-32) at Montcalm and 0.55 (P-0.001, n-36) at KBS. This indicates that 1) sufficient residual N was available following legumes to sustain vine growth later into the growing season, and 2) the magnitude of the response was affected by the amount of N in the previous legume crop. This relationship between vine N and legume N may be partially responsible for the difference in the effect of legume management at the two locations. At Montcalm, where green manure legumes accumulated more N than hay crops of the same species, vines accumulated more DM (Pa0.00l; ’GM vs. hay . legumes’) and more N (P-0.007) following green manure than following hay craps. At KBS, the opposite effect was observed; vine DM and N content was higher 21 following hay crop legumes, which had higher PDN yields than the respective green manure crops. Although N accumulation by legumes during 1987 and early 1988 ranged from 33 to 238 kg ha“, no significant differences in tuber yield between rotations were observed at either location when no N fertilizer was applied to the second year potato crop. Mean total, marketable (’A’ grade plus oversize), ’B’ grade, and cull yields were 28.5, 25.7, 1.8, and 1.0 Mg ha'l, respectively, at Montcalm, and 27.4, 22.8, 1.5, and 3.0 Mg ha", respectively, at KBS. The lack of tuber yield response to rotation in our research is in contrast to the results of Odland and Sheehan (1957) and Emmond and Ledingham (1972). In these two studies, increases in total tuber yield ranged from 13% following red clover (Odland and Sheehan, 1957) to 25% following sweetclover (Emmond and Ledingham, 1972), compared to continuous potato. Because each of these studies consisted of numerous rotation cycles, it is unclear whether the yield increases were due a contribution of N from the preceding legume crop or to non-N effects associated with the legumes. Using no N fertilizer, O’Sullivan and Reyes (1980) demonstrated that non-N effects may also be important in potato rotations, as potato yield increased from 17.6 (potato-mtg) to 26.4 Mg ha‘1 (corn-mam), although an N contribution from corn would not be expected. Preceding a non-legume crop with an N-fixing legume does not always lead to higher non-legume yield (cg. Kroontje and Kehr, 1956; Fleming et al., 1981; Hesterman et al., 1986), as shown by our results. In our research, there are two potential causes for the lack of tuber yield response to rotation, given the marked vegetative response discussed above. First, although a considerable amount of N is contained in the incorporated legume material, mineralized legume N may not accumulate in sufficient quantity during early vegetative growth to affect the number and survival of tuber initiates. Potato is most sensitive to insufficient or 22 excess soil N during the critical period of tuber initiation, and differences in N fertility during this period ultimately inf luencc the yield and quality of the potato crop (Roberts et al., 1982; Ojala et al., 1990). Alternatively, the lack of tuber yield response to legume N may indicate that some factor other than soil N level was limiting during tuber initiation and/or early tuber bulking. The dry spring in 1988 may have led to a soil moisture deficit, especially following legumes that utilized available soil moisture in the spring, that was not overcome until tuber initiation had taken place. Egtagg N response in rotation, The availability of residual soil N late in the growing season, from legumes or fertilizer, is apparent from differences in vine DM and N content and vine N concentration. Because these three parameters responded similarly to both N sources, only vine N content is presented here. The effect of fertilizer N rate on vine N content was not significant following fallow and hairy vetch at Montcalm (mean N content - 75 and 120 kg ha'l, respectively) and following alfalfa hay, birdsfoot trefoil hay, and spring- and f all-seeded hairy vetch at KBS (mean N content - 85, 77, 92, and 68 kg ha'l, respectively). At Montcalm, the response of vines to N fertilizer was similar following alfalfa (green manure or hay) and birdsfoot trefoil, so the mean response of these rotations is shown (ALF-BGM) in Figure 1.1. Given the low PDN yield of birdsfoot trefoil green manure (70 kg N ha"), the similarity in response to that following alfalfa may be due to the mineral- ization of N from weed biomass in the trefoil plots. Potato vines following sweetclover contained nearly 50% more N than following alfalfa or birdsfoot trefoil green manure (at all N rates), and twice as much N as following red clover, birdsfoot trefoil hay, corn, or potato ( mean response as RC-BHY-C-P in Figure 1.1). Vine N content following ’Nitro’ alfalfa at KBS was greater than following any other crop at N rates from 75 to 225 kg ha'1 (Figure 1.2). Significant vegetative 23 24o- zoo- \ see A .8 E 180- .. Air-am “ g .2... 8 W—G—P Z 0 ea- C S ‘0‘ o o 73 150 255 N applied ln 1988 (kg/ha) Figure 1.1. Effect of N fertilizer rate on late-season (August 25, 1988) potato vine N content following: alfalfa and birdsfoot trefoil green manure (mean response, ALF- BGM); red clover, birdsfoot trefoil hay, corn, and potato (mean response, RC-BI-IY-C- P); and sweetclover (SWC) at Montcalm. [Regression equations for ALF-BGM, RC- BHY-C-P, and swc, respectively: Y-51.08+0.826N-0.00176N’, 113-099; Y-32.4+0.305N, R’s0.98; and Y=87.17+l.364N-0.00308N’, 18.0.88, where Y—vine N content (kg ha'l) and N=fertilizer N applied (kg ha“).] 24 responses to N fertilizer among other legumes (red clover, and alfalfa and birdsfoot trefoil green manurcs) were nearly identical, except when no N fertilizer was applied. Vine N content was lower following non-legumes (fallow, corn, and potato) than following legumes, as was the case at Montcalm, suggesting that there is an additive effect of legume and fertilizer N on vegetative response. Total, marketable, ’B’ grade, and cull tuber yields of potato at KBS in 1988 were not affected by the previous crop or the 1988 N fertilizer rate, averaging 27.6, 22.7, 1.8, and 3.1 Mg ha", respectively. The lack of tuber yield response to both of these factors strongly suggests that, despite scheduled irrigation applications, soil moisture was limiting during at least the first half the growing season. The 12 d period in late June, when irrigation was not available, may have been particularly harmful, since this is the period of tuber initiation. The susceptibility of potato to drought is well documented (eg. van Loon, 1981). The results of Ojala et a1. (1990) show that as ”total seasonal water“ increased from 161 to 586 mm, total tuber yield increased approximately f ive-f old, and the yield response to residual plus fertilizer N also increased. Miller and Martin (1985) found that deficit irrigation prior to June 14 did not reduce total tuber yield measured at the end of the season (mid- September). However, tuber weight was significantly lower throughout most of the growing season (to August 4) as a result of the early season deficit, and vine dry matter was higher in plots receiving insufficient irrigation. There were few significant responses to previous crop and N fertilizer rate at Montcalm. The yield of ’B’ grade tubers was similar for all treatments (mean ’8’ yield - 1.6 Mg ha'l), and no significant correlation existed between ’B’ grade yield and any other tuber or vine yield parameter. Gardner and Jones (1975) and Tyler et al. (1983) also found that even as total tuber yield increased with N rate, ’8’ grade yield remained relatively constant across N rates. Malformed tubers (primarily 25 180- 150‘ A a 3 v 120... a .5. 5 u 90... mo 2 é tic-m 50- mm mm oat-ear so 7’: 130 225 N applied in 1988 (kg/ho) Figure 1.2. Effect of N fertilizer rate on late-season (August 30, 1988) potato vine N content following ’NITRO’ alfalfa, red clover and alfalfa green manure (mean response, RC-AGM), birdsfoot trefoil green manure (BFTGM), FALLOW, and corn and potato (mean response, CRN-POT) at KBS. [Regression equations for NITRO, RC-AGM, BFTGM, FALLOW, and CRN-POT, respectively: Y-68.42+l.46N- 0.00592N’, R3-0.92; Y-76.4+0.65N-0.00303N’, R’-0.37; Y-47.5+1.16N-0.00418N’, 113-093; Y-47.4+0.84N-0.00372N’, 111-090; and Y-37.7+0.69N-0.00283N3, R’-0.75, where Y-vine N content (kg ha") and N-fertilizer N applied (kg ha").] 26 tubers exhibiting second growth) indicate either intermittent nutrient supply and/or the sudden relief of plant stress during tuber bulking (Robins and Domingo, 1956; Hiller et al., 1985). Cull yield increased with N rate in all systems except hairy vetch-potato (cull yield-3.0 Mg ha"), and was significantly higher following sweetclover (SWC; Figure 1.3) than any other crop. Hiller et a1. (1985) discussed the potential causes of tuber malformation, which included excessive vegetative growth. We found that cull yield for all rotations and N rates was significantly correlated with vine DM and N content and N concentration (r-0.419, 0.519, and 0.481, respec- tively; P<0.001)). However, this correlative relationship may arise simply because excessive vine growth and tuber malformation are symptomatic of excess N, not because they are (necessarily) related. Total tuber yield following alfalfa, birdsfoot trefoil, sweetclover and spring- seeded hairy vetch exhibited no significant response to fertilizer N, averaging 30.9 Mg ha". Using the corn-potato rotation as a control (regression equation in Figure 1.4), these legumes replaced 69 kg ha‘1 of N fertilizer, according to the FRV tech- nique described by Hesterman (1988). Total yield of potato following fallow or potato increased linearly with N rate (Figure 1.4). Maximum yield following red clover (mean of green manure and hay crops) and corn required approximately 150 kg N ha", slightly below the 170 kg ha'1 rate recommended for potato in Michigan (Vitosh, 1990). The effect of N rate on marketable tuber yield was most pronounced in the corn-potato rotation (Figure 1.5), with maximum marketable yield requiring approximately 150 kg N ha", similar to optimum N rates reported by Murphy ct a1. (1967), Gardner and Jones (1975), and O’Sullivan (1978), and the recommended rate in Michigan (Vitosh, 1990). The reduction in marketable yield at the highest N rate was due primarily to an increased proportion of malformed tubers. The effect of rotation on marketable yield was not significant for the other rotations; thus the l 3 Cull tuber yield (Mg/ho) f is 160 do u Applied ln 1m (kg/ho) Figure 1.3. Effect of N fertilizer rate on cull tuber yield following sweetclover (SWC), other legume crops (OTHER LEG; mean of alfalfa, birdsfoot trefoil, hairy vetch, and red clover) and non-legume crops (NON-LEG; mean of corn, fallow, and potato) at Montcalm. [Regression equations for SWC, OTHER LEG, and NON-LEG, respectively: Y-l.32+0.051N-0.00013N3, 111-099; Y-0.746+0.015N, 113-039; and Y-0.338+0.014N, R3-0.94, where Y-tuber yield (Mg ha") and Nafertilizer N applied (kg ha' ).] 28 40.. FM. é “' ... 3 i 32" / \ 3‘ a .. ... 24 0 1’5 130 225 N Applied In 1988 (kg/ho) Figure 1.4. Effect of N fertilizer rate on total potato tuber yield following corn (CRN), fallow (FAL), potato (PCT), and red clover (RC), at Montcalm. [Regression equations for CRN, FAL, POT, RC, respectively: Y-24.33+0.12N-0.00036N2, 3.3-0.97; Y-28.87+0.0413N, R3-0.98; Y-28.93+0.0292N, 113-034; Y-28.08+0.567N-0.000176N’, R3-0.87, where Y-tuber yield (Mg ha") and N-fertilizer N applied (kg ha").] 29 \:.. Marketoble tuber yield (Mg/ho) g l 7'5 150 225 N Applled In 1988 (kg/ha) Figure 1.5. Effect of N fertilizer rate on marketable tuber yield following corn (CRN) and all other crops (mean response, OTHER) at Montcalm. [Regression equations for CRN and OTHER, respectively: Y-22.2l+0.122N-0.000463N’, R’s0.90; and Y=26.15+0.28N-0.00012N3, R3-0.81, where Y-tuber yield (Mg ha“) and N-fertilizer N applied (kg ha'l).] 30 mean N response is shown (OTHER; Figure 1.5). Murphy et al. (1967) examined potato yield at each of six N rates (67 to 235 kg ha") following potato, red clover, and red clover plus oat (Avena saliva 1.). They found that within each N rate, previous crOp had no significant effect on tuber yield. O’Sullivan (1978) observed that potato N response following corn or potato was nearly identical, except that yield was slightly higher following corn. SUMMARY In our research, we found that seeding year harvest of legumes was not necessarily detrimental to N accumulation, compared to a green manure crop of the same species. For alfalfa, birdsfoot trefoil, and red clover at two locations, only red clover at Montcalm exhibited a significantly lower PDN yield due to seeding year harvest. Furthermore, periodic harvests served to reduce weed competition with alfalfa and red clover (but not birdsfoot trefoil). This type of management should also be considered for weed control in green manure crops, where clipping has been shown to increase total N production (Stickler and Johnson, 1958; Groya and Sheaf f er, 1985). We observed a pronounced vegetative response to legume N by the subsequent potato crop, grown without N fertilizer. The higher vine DM and N content (f ollow- ing legumes) late in the growing season is evidence that N from incorporated legumes eventually becomes available to the potato crop. However, the lack of tuber yield response to legume N indicates that either there is a lack of synchrony between legume N availability and potato N demand or that available N was not the limiting factor in potato yield. The response of the potato crap to fertilizer N in different rotations at Montcalm helps to distinguish between these alternatives. A significant tuber yield response to fertilizer N was observed in only one legume-based rotation 31 at Montcalm; red clover-potato. If the lack of tuber response to legume N was due to asynchrony in supply and demand, the addition of N fertilizer would be expected to elicit an increase in tuber yield (as in the corn-potato rotation, for example). This not being the case, the lack of tuber response to legume N, and to fertilizer N following legumes, is attributed to a soil moisture deficit induced by legume growth during the dry spring in 1988. Even with supplemental irrigation, this deficit was apparently not overcome until after tuber initiation had occurred. REFERENCES Adams, W.E., H.D. Morris, and RN. Dawson. 1970. Effect of cropping systems and nitrogen levels on corn (Zea mays 1..) yields in the southern Piedmont region. Agron. J. 62:665-669. Bowren, K.E., D.A. Cook, and R.K. Downey. 1969. Yield of dry matter and nitrogen from tops and roots of sweetclover, alfalfa and red clover at five stages of growth. Can. J. Plant Sci. 49:61-68. Brodie, B.B., J.M. Good, and WE Adams. 1969. Population dynamics in cultivated soil: effect of sod-based rotations in Cecil sandy loam. J. Nema. 1:309-312. Bruulsema, T.W., and ER. Christie. 1987. Nitrogen contribution to succeeding corn from alfalfa and red clover. Agron. J. 79:96-100. Curl, E.A. 1963. Control of plant diseases by crop rotations. Bot. Rev. 29:413-479. Dubetz, 8., and LB. Bole. 1975. Effect of nitrogen, phosphorous, and potassium fertilizers on yield components and specific gravity of potatoes. Amer. Potato J. 52:399-405. Emmond, G.S., and RJ. Ledingham. 1972. Effects of crop rotation on some soil-borne pathogens of potato. Can. J. Plant Sci. 52:605-611. Fleming, A.A., J.E. Giddens, and E.R. Beaty. 1981. Corn yields as related to legumes and inorganic nitrogen. Crop Sci. 21:977-980. Fribourg, H.A., and W.V. Bartholomew. 1956. Availability of nitrogen from crop residues during the first and second season after application. Soil Sci. Soc. Amer. Proc. 20:505-508. Gardner, B.R., and J.P. Jones. 1975. Petiole analysis and the nitrogen fertil- ization of Russett Burbank potatoes. Amer. Potato J. 52:195-200. Groya, F.L., and CC. Sheaf f er. 1985. Nitrogen from forage legumes: harvest and tillage effects. Agron. J. 77:105-109. Harris, G.1-1., and CB. Hesterman. 1990. Quantifying the nitrogen contribution from alfalfa to soil and two succeeding crops using nitrogen-15. Agron. J. 82:129-134. Hesterman, QB. 1988. Exploiting forage legumes for nitrogen contribution in cropping systems. pp. 155-166 111 J.P. Powers (ed.). Cropping strategies for efficient use of water and nitrogen. ASA Spec. Publ. 51. 32 33 Hesterman, O.B., C.C. Sheaf f er, D.K. Barnes, W.E. Leuschen, and J.H. Ford. 1986. Alfalfa dry matter and nitrogen production, and f ertilizcr nitrogen response in legume-corn rotations. Agron. J. 78:19-23. Hiller, L.l(., D.C. Keller, and RE. Thornton. 1985. Physiological disorders of potato tubers. pp. 389-455 in P.H. Li (cd.) Potato physiology. Academic Press Inc., Orlando. Hoyt, PB, and R.H. Leitch. 1983. Effects of forage legume species on soil moisture, nitrogen, and yield of succeeding barley crops. Can. J. Soil Sci. 63:125-136. Jackson, R.D., and J.L. Haddock. 1959. Growth and nutrient uptake of Russett Burbank potatoes. Amer. Potato J. 36:22-28. KleinkOpf, G.E., D.T. Westerman, and RB. Dwelle. 1981. Dry matter produc- tion and nitrogen utilization by six potato cultivars. Agron. J. 73:799-802. Kroontje, W., and W.R. Kehr. 1956. Legume top and roots yields in the year of seeding and subsequent barley yield. Agron. J. 48:127-131. Lauer, D.A. 1985. Nitrogen uptake patterns of potatoes with high-frequency sprinkler-applied N fertilizer. Agron. J. 77:193-197. Lauer, D.A. 1986. Response of Nooksack potatoes to nitrogen fertilizer. Amer. Potato. J. 63:251-262. McVay, K.A., D.E. Radcliffe, and W.L. Hargrove. 1989. Winter legume effects on soil properties and nitrogen fertilizer requirements. Soil Sci. Soc. Amer. Proc. 53:1856-1862. Murphy, H.J., P.N. Carpenter, and MJ. Goven. 1967. Potato f ertilization- rotation studies on Aroostock Farm permanent fertility plots 1951-1965. Maine Agr. Expt. Stn. Bull. 645. Odland, T.E., and J.E. Sheehan. 1957. The effect of redtop and red clover on yields of following cr0ps of potatoes. Amer. Potato J. 34:282-284. Ojala, J.C., J.C. Stark, and GE. Kleinkopf. 1990. Influence of irrigation and nitrogen management on potato yield and quality. Amer. Potato J. 67:29-44. O’Sullivan, 1.0. 1978. Effects of rotation and nitrogen on yield and quality of potatoes. Can. J. Plant Sci. 58:475-483. O’Sullivan, 1.0., and A.A. Reyes. 1980. Effects of soil fumigation, rotation, and nitrogen on yield, petiole NOS-N, and Verticillium wilt of potatoes. J. Amer. Soc. Hort. Sci. 105:809-812. Purvis, B.R., and J.M. Blume. 1939. The role of green manures in potato production. Amer. Potato J. 16:32-36. 34 Roberts, 8., W.R. Weaver, and J.P. Phelps. 1982. Effect of rate and time of fertilization in nitrogen and yield of Russett Burbank potatoes under center pivot irrigation. Amer. Potato J. 59:77-86. Robins, J.S., and CE. Domingo. 1956. Potato yield and tuber shape as affected by severe soil moisture deficits and plant spacing. Agron. J. 48:488-492. Saf f igna, P.G., and D.R. Keeney. 1977. Nitrogen and chloride uptake by irrigated Russett Burbank potatoes. Agron. J. 69:258-264. Saf f igna, P.G., D.R. Keeney, and CB. Tanner. 1977. Nitrogen, chloride, and water balance with irrigated Russett Burbank potatoes in a sandy soil. Agron. J. 69:251-257. Saloman, M. 1962. Soil aggregation-organic matter relationships in redtop- potato rotations. Soil Sci. Soc. Amer. Proc. 26:51-54. Sprague, H.B. 1939. The value of winter green manure crops. New Jersey Agr. Expt. Stn. Bull. 609. Steel,R.G.D., and J.H. Torrie. 1980. Principles and procedures of statistics: a biometrical approach. 2nd Edition. McGraw-Hill Book Co., New York. Stickler, F.C., and 1.1. Johnson. 1959. The influence of clipping on dry matter and nitrogen production of legume green manures. Agron. J. 51:137-138. Stickler, F.C., W.D. Shrader, and 1.]. Johnson. 1958. Comparative value of legume and fertilizer nitrogen for corn production. Agron. J. 50:157-160. Terman, G.L. 1949. Green manure crops and rotations for Maine potato soils. Maine Agr. Expt. Stn. Bull. 474. Tyler, K.B., F.E. Broadbent, and J.C. Bishop. 1983. Efficiency of nitrogen uptake by potatoes. Amer. Potato J. 60:261-269. van Loon, CD. 1981. The effect of water stress on potato growth, develop- ment, and yield. Amer. Potato J. 58:51-69. Vitosh, ML. 1984. Irrigation scheduling for potatoes in Michigan. Amer. Potato J. 61:205-213 Vitosh, ML. 1990. Potato fertilizer recommendations. Michigan State Univ. Coop. Ext. Serv. Bull. E-2220. Wheeler, EL 1946. The residual cf f ect of crop rotations on potato yield and the presence of potato scab. Michigan Agr. Expt. Stn. Quart. Bull. 38:326-332. CHAPTER TWO ECONOMIC COMPARISON OF ALF ALF A-POTATO, CORN-POTATO, AND CONTINUOUS POTATO SYSTEMS ABSTRACT Crop rotations for high-value crops like potato (Solanum tuberosum L.) will be implemented only when the economic viability of such systems is demonstrated. Monte Carlo computer simulations were used here: (i) to compare the magnitude and variability of net income from alfalfa (Medicago sativa L.) green manure-potato, alfalfa hay-potato, and corn (Zea mays L.)-potato rotations and continuous potatoes, assuming normal distributions for prices and yields; and (ii) to evaluate the effect of reduced mean yield and skewness (positive and negative) on the net income from continuous potatoes. In a baseline simulation to compare rotational and continuous potato production systems, continuous potatoes had a higher mean annual net income (S498/a), with more variability (standard deviation, 0-3372/3), than any rotational system. The alfalfa hay-potato rotation generated more net income ($312/a) than the alfalfa green manure-potato or corn-potato rotations (S226 and $276/a, respectively), although absolute variability was similar for all rotations (a-SZOO/a). A 15% reduction in continuous potato yield (to 255 cwt/a), simulating the long-term yield decline commonly seen in continuous potatoes, resulted in a mean net income ($322/a) nearly identical to that from the alfalfa hay-potato rotation. However, net income variability was 37% higher for continuous potatoes, confirming that monocultural production carries greater economic risk than rotations, even at the 35 36 same mean net income. The introduction of positive skewness (gamma distribution) into potato yield did not affect mean net income or income variability, compared to a normal distribution, when both distributions had a mean of 270 cwt/a and F34 cwt/a. However, if potato yield exhibited negative skewness (beta distribution), mean net income and variability both increased, as did potential loss (minimum net income- -200, -158, and -$250/a for normal, gamma and beta distributions, respectively). The implementation of potato rotations for agronomic reasons depends most strongly on the assumptions that potato yield in monoculture declines with time. The simulations performed here demonstrate that if continuous potato yield declines more than 15%, rotations (with alfalfa or corn) generate more net income, with less variability, than a continuous potato system. INTRODUCTION The potential agronomic benefits from growing potatoes or other crops in rotation with nitrogen (N)-fixing legumes have been demonstrated, and include: N contribution from legumes to subsequent potato crops (Emmond and Ledingham, 1972); improved soil structure and water infiltration (Odland and Sheehan, 1957; Saloman, 1962); and disruption of pest and weed cycles (Brodie et a1., 1969; Weinhold et al., 1964). Evidence that agricultural practices lead to environmental pollution, particularly nitrate and pesticide contamination of groundwater, coupled with an interest in reducing non-renewable, petroleum-based inputs in potato production, have increased interest in crop rotations for environmental reasons. The adoption of rotational production systems by producers on a large scale, however, will occur only when the economic viability of these systems is demonstrated. Two criteria that have been useful in comparing the economics of different cropping systems are the magnitude and variability of net income (gross income minus variable cash costs). A primary goal of agricultural production, like other business enterprises, is to provide an acceptable level of return to the producer for capital, labor, and management inputs. Linear programming and enterprise budgets have both been used to select the optimal production system from a set of alternatives (Heady et a1., 1956; Helmers et a1., 1986; Rowberry and Anderson, 1983). In some cases, higher net income has been demonstrated from legume-corn rotations than continuous monocultures of corn (Heady and Jensen, 1952; Hesterman et a1., 1986). The profitability of legume-corn rotations resulted from either increased corn yields in rotation (eg. Heady and Jensen, 1951) and/or reduced production costs (eg. 37 38 Hesterman et al., 1986b) for the subsequent corn crop. Few economic analyses have been conducted evaluating rotations with intensively managed crops like potato, requiring high inputs of pesticide, labor, and capital. Rowberry and Anderson (1983) examined net returns of potato rotations with barley (Hordeum vulgare L.) and corn compared to continuous potatoes, and found that continuous potatoes provided the highest mean net income over a 7 yr period. A selection of optimal potato rotations under variable risk aversion was conducted by El-Nazar and McCarl (1986). They found that as risk aversion increased, the time period between potato crops in rotation also increased. Income variability or risk plays an important, and often overlooked, role in economic decision-making by producers (Stovall, 1966). Expected income can be simply defined by: E (income) - (P(Y))-C [l] where P, Y, and C are the price, yield and production costs, respectively, of a specific crop. Risk is introduced into [1] because commodity prices and crop yields cannot be known with certainty until the crop is harvested and sold. Thus, both P and Y must be considered stochastic components, each with a characteristic (but unknown) probability distribution. This assumes that the variability in production costs is minimal compared to price or yield variability (Musser et a1., 1981; Buccola, 1986). If only one crop is produced, variability in net income is defined as 012, following a multivariate distribution dependent on the variability in price and yield. However, the production of two (or more) crops, either through crop rotation or diversification, requires that variability in income be expanded to: 39 at: - 012 + Oi: + Zpeicj, for i and j - l to n [2] (from Heady, 1952; p is the correlation of income between crop i and j) to include income covariance between different crops. The use of either rotational or diversified systems as an effective method of reducing income variability is based on the premise that if a strong positive correlation does not exist between the net income generated by different crops in rotation (ideally, p s 0), then reduced income from one crop is more likely to be accompanied by increased income from another crop (Heady, 1952; Stovall, 1966). This results in less drastic losses when low prices prevail and lower profits under high price conditions, compared to continuous monocultures. Computer simulations, such as Monte Carlo techniques, are useful in establishing the probability distribution of net income from different enterprise combinations (eg. Buccola, 1986) when sufficient agronomic or production cost data are not available. For example, if the underlying price and yield distributions are estimated from detrended historical price and yield series (Pope and Ziemer, 1984; Tew and Reid, 1988), numerous combinations of price and yield can be used to generate a distribution of potential net incomes. Concerns have been raised, however, over the widely held assumption that variability in crop prices and yields follow normal distributions (King and Robison, 1981; Collender and Zilberman, 1985), and over the impact of this assumption on net income distribution (Tew and Reid, 1988). Research by Day (1962) demonstrated that long-term yield series commonly exhibit some degree of non-normality or skewness, which could ultimately affect the probability of attaining a certain level of income and the identification of optimal production systems. 40 Further investigation is needed on the economic viability of legume- or corn- potato rotations, compared to continuous potato production, and on the identi- f ication of the factors influencing the relative profitability of these systems. This research uses computer simulations to address these two broad areas. The economic evaluation is specifically for alternative potato production systems (ie. potatoes are grown at least every other year). The objectives of this research were: i) to determine the mean net income and variability in net income from alfalfa-potato, corn-potato and continuous potato systems assuming normal distributions of prices and yields; and ii) to examine the effect of reduced mean yield level and skewness on net income from continuous potato production. MATERIALS AND METHODS Monte Carlo simulations were performed to compare the expected mean net income and the variability in income of four production systems: alfalfa green manure-potato, alfalfa hay-potato and corn-potato rotations; and continuous potatoes. Enterprise budgets (Table 2.1) were established for each crop using information from Shapley and Williams (1988) for potatoes and Nott et a1. (1988) for alfalfa and corn, with minor modifications. These budgets include variable production costs, but not fixed costs for land or machinery ownership. Nitrogen fertilizer rates are; 175 1b/a for potatoes following potatoes or corn, 100 lb/a for potatoes following alfalfa (Vitosh, 1990), and 110 lb/a for corn (Nott et a1., 1988). All costs, revenues, and net incomes are calculated based on a farm size of 500 acres, slightly smaller than the average farm size for potato producers in Michigan (Shapley, 1987). All 500 acres are in potatoes for the POT system. The rotational systems have 250 acres in potatoes and 250 acres in either alfalfa or corn. 41 Table 2.1. Enterprise budgets for potatoes, corn, and seeding year alfalfa in Niehigan. Crop ' Seeding-year Expense Potatoesf Cornt alfalfa: Yield goal 300 cut/a 100 bu/a 3 ton/a ------------------------------------ Slacre ----------------------°---- Seed 150.00 19.60 26.25 Non-l fertilizer 49.92 10.80 30.00 n fertilizer 35.00 (20.00)§ 22.00 0.0 Chemical costs Fungicide 64.10 0.0 0.0 Insecticide 73.43 2.20 3.00 Herbicide 63.44 13.00 8.75 Cultural labor1 30.08 32.67 39.60 Fuel, oil! 70.00 9.90 8.00 Machinery repair 75.00 18.00 5.00 Utilities 20.00 2.50 1.50 Harvest, storage 99.86 20.10:: 75.00:: Packing, marketing 442.36 1.00 0.00 Freight 0.0 12.00 22.40 Drying 0. 12.00 0.0 TOTAL 1133.19 (1123.19); 137.77 219.50 ’ Adapted from Shapley and Uilliane (1988). Adapted from Nott et a1. (1988). 5 Cost of N fertilizer with reduced application rate following alfalfa (Vitosh, 1990). 1 Labor for: planting, cultivation, irrigation and spraying, 85.50/hr. Includes fuel for potato irrigation. fl Fixed harvest costs a $99.86/acre; variable harvest costs I $1.47/cut. Custom harvest cost for corn. Sehuab and Norgaard (1988). :1 Custom harvest cost for alfalfa; $25.00/aere/harvest. 42 9.132211% Mean yields of corn and potato were initially set at 100 bu/a and 300 cwt/a, respectively (Table 2.2). These yields levels were used in the cost publications cited above, but also approximate mean yields (1981 to 1988) from Montcalm county, M1, the primary potato producing county in the state (Michigan Agricultural Statistics). The variability in potato and corn yields was calculated as the standard deviation of the yield series (1981 to 1988) from Montcalm county (0.38.1 cwt/a for potatoes and 12.9 bu/a for corn). Seeding year alfalfa hay yield from Nott et a1. (1988) was estimated to be only 2.0 ton/acre. Recent research by Groya and Sheaffer (1985) and Hesterman et al. (1986a) in Minnesota suggested that such yields underestimate the seeding year yield potential of alfalfa. This was confirmed by Griffin and Hesterman (1990; unpublished data‘), who achieved yields of 3.5 to 4.0 ton/acre on potato soils in Montcalm county. Thus the mean yield of alfalfa was set at 3.0 ton/acre («r-0.3 ton/acre). All yield distributions summarized in Table 2.2 were assumed to be normal. Yield distributions were truncated at 210 and 360 cwt/a for potatoes and at 50 and 150 bu/a for corn, approximately 10% below and above the extremes observed in the yield time series, to simulate a realistic yield range for these crops in Michigan. Correlation coefficients between yields of potatoes, corn, and alfalfa hay (Table 2.3), required to construct variance- covariance matrices for simulations, were determined from statewide average yields of each crop (1981 to 1988) because alfalfa yields specific to Montcalm county were not available. As expected, corn and alfalfa hay yields were more highly correlated with each other than with potato yields, because corn and alfalfa are generally grown under dryland conditions, while potatoes are grown under irrigation. 1 See Chapter One, Table 1.3. 43 Table 2.2. Sunnary statistics of truncated normal yield distributions for potato, corn, and alfalfa hay. Crop Potato Corn Alfalfa Yield mits cut/a bu/a ton/a Mean 300 100 3.0 Standard deviation 38.1 12.9 0.3 Miniuua yield estimate 210 50 1.5 Maxiuua yield estimate 360 150 4.5 44 Table 2.3. Sieple correlation coefficients (C) between potatoes, corn, and alfalfa hay. grog yields Potato Corn Alfalfa .................. L ...---------..--.-. Potato 1.00 0.28 0.43 Corn 1.00 0.83 Alfalfa 1.00 Qrgp priceg Potato Corn Alfalfa .................. £ .-------------.-..- Potato 1.00 0.69 0.57 Corn 1.00 0.88 Alfalfa 1.00 45 mm Nominal crop prices were compiled from the Michigan Agricultural Statistics reports (1981-1988), and adjusted to constant 1988 dollars (19885) using the "Prices paid to producers" indices (USDA, 1989). Reliable estimates of the market price of alfalfa were not available. Thus, the alfalfa prices were calculated based on the cost of replacing the energy and protein in alfalfa with corn and soybean meal, respectively, according to Hesterman et al. (1988). Real crop prices (in 19888; Table 2.4) were then used to calculate the mean and standard deviation in crop prices. Price distributions were assumed to be normal, and the null hypotheses of normality could not be rejected, based on the Shapiro-Wilks test (Shapiro and Wilk, 1965), for potato, corn or alfalfa hay prices. Correlations between the prices of the three products (potatoes, corn, and hay) are also shown in Table 2.4. The correlation between the price and yield of each crop was assumed to be zero, based on the assumption that total production from the farm was not sufficient to influence the market price (ie. perfect competition; El-Nazar and McCarl, 1986). MW Monte Carlo simulations were performed using the Agricultural Risk Management Simulator (ARMS; King et a1., 1989). Following the input of production costs, price and yield distributions, and covariance between craps (for prices and yields), 250 random estimates were generated from the price and yield distributions of each crop. Net income (gross income minus cash costs) was calculated from paired price and yield estimates according to equation [1], yielding a probability distribution of potential net incomes for each production system. The initial (”baseline") simulation comparing rotations (AGM, AHY, and CRN) with continuous potatoes (POT) was performed using normal distributions for all crop prices and yields (described in Tables 2.2 and 2.3). 46 Table 2.4. Prices for potatoes, corn, and alfalfa hay in Michigan, 1981-1988 (adjusted to 19883). Crop Year Potato Corn Alfalfa Slcwt Slbu Slton 1981 4.66 1.90 79.7 1982 4.61 2.42 85.4 1983 6.65 2.56 99.1 1984 4.50 2.05 74.5 1985 4.40 2.05 74.8 1986 6.36 2.44 86.1 1987 5.47 2.69 110.2 1988 8.00 2.60 94.9 Mean 5.58 2.34 88.1 Standard deviation 1.31 0.29 12.5 47 There is ample agronomic evidence that yield levels in continuous potato monocultures cannot be maintained with constant inputs of fertilizer and pesticide,with the decline in potato yield over time generally being attributed to increased disease severity in monoculture (McDole and Dallimore, 1978; O’Sullivan and Reyes, 1980). Long-term yield series for continuous potatoes are not available for Michigan to examine the economic impact of such a trend. Thus, a simulation (rt-250) was conducted to asses the effect of declining mean potato yield on net income. Holding pre-harvest and fixed harvest costs constant, potato yields were reduced in 5% increments (15 cwt/a) to 75% of the baseline level of 300 cwt/a. The coefficient of variation (a/mean) was held constant at 12.7% as potato yield fell, and each yield distribution was truncated at 70 and 130% of the mean yield. A third simulation was conducted to evaluate the effect of potato yield distribution on net income level and variability (potato price followed a normal distribution with mean=$5.58/cwt and a=$l.31/cwt). Research by Day (1962) demonstrated that both positive and negative skewness was evident in crop yields over time, with the direction of skewness varying with crop species and management. Thus, a normal yield distribution was compared to two skewed distributions: a positively skewed gamma distribution approximated by «#2 (coefficient of skewness=0.60); and a negatively skewed beta distribution approximated by «5 and 5-2 (coefficient of skewness- -0.51). The parent distributions from which estimates were generated each had a mean of 270 cwt/a and «:34. RESULTS AND DISCUSSION W The mean and variability of net income for the four production systems is summarized in Table 2.5. The 95% confidence interval limits associated with the mean and standard deviation of each 48 Table 2.5. Net income distribution for alfalfa green manure-potato, alfalfa hay-potato, and corn- potato rotations (250 acres of rotation crop, 250 acres of potatoes), and continuous potatoes (500 acres) asst-ing normal distributions of prices and yields. Production syStem Alfalfa green Alfalfa hay- Corn- Continuous manure-potato potato potato potato -------------------------- Net income, $1,000's -----------'------------- Mean 113 156 138 249 Lower c1 limi :1“ 100 11.2 126 226 Upper Cl limit 125 169 151 272 Standard deviation 99 106 100 186 Lower CI limit 92 98 92 172 Upper C1 limit 109 117 110 205 CV, X 88 68 72 75 Met income range Minimum -52 -32 -27 -55 Maximum 373 434 419 762 T C1 8 95% confidence interval. 49 system, calculated according to Equations (3) and (4) in Collender (1989), are also shown. Based on this initial simulation, the alfalfa hay-potato rotation provided the highest mean net income ($156,000, for farm size-500 acres) among rotational systems, 12 and 28% higher than the corn-potato and alfalfa green manure-potato rotations, respectively. Although mean net income was significantly different for these three systems (AHY > CRN > AGM), absolute variability was similar for all rotational systems, with «$100,000. The higher relative variability of the alfalfa green manure system resulted because only one saleable product (potatoes) was produced. Thus, this system does not actually represent economic diversification of the farm operation, only a reduction in potato acreage. In addition, the potential contribution of N from the green manure cr0p to the subsequent potato crop (circa 75 1b/a) has a monetary value of $3750 (for 250 acres), compared to alfalfa establishment costs of about $30,000. The comparison made here between the two alfalfa-based rotations supports Hesterman et al. (1986b) and Allison and Ott (1987), who found that the value of alfalfa as a cash crop far exceeds its value as an N source for subsequent crops under current price and cost conditions. The mean net income from continuous potatoes ($249,000) exceeded that from any rotational system. Rowberry and Anderson (1983) also found that continuous potatoes provided a higher mean income than rotations with barley, corn or soybeans (Glycine max Merr.). They did not, however, address the issue of income variability in continuous and rotational potato production systems. We found that in addition to a higher mean, net income from continuous potatoes was more variable (ie. higher standard deviation), with a greater potential for both profit and loss, than any of the rotations. Although our research does not directly address risk aversion, the selection of an appropriate potato production system would represent a compromise between these two factors, net income and variability, based on the preferences of 50 the producer. El-Nazar and McCarl (1986) included risk aversion in their selection of optimal potato rotations in Oregon. They found that as producer risk aversion increased, 1) mean net income and variability decreased, and 2) the time span between potato crops increased, indicating that the potato phase of the rotations carried the greatest economic risk. Although we found that absolute variability increased with net income, as it usually does, relative variability was similar for the AHY, CRN, and POT systems, with a CV of approximately 70%. This conflicts with the results of Schurle and Erven (1979), who found that as the production of specialty vegetable crops increased, net income, absolute variability and relative variability all increased. The increased variability associated with monocultures that we see here is not specific to potato production. The analysis of monocultural, rotational and diversified grain production systems by Helmers et a1. (1986) showed that continuous corn had the highest standard deviation and the greatest range in net income, despite having the lowest mean net income. Likewise, continuous grain sorghum [Sorghum bicolor (L.) Moench.] had a higher standard deviation and CV than rotational production system. Schurle and Erven (1979) also found that net income from continuous corn production was more variable than from diversified systems including soybeans or wheat (Triticum aestivum L.). We. From the simulation discussed above, we found that continuous potatoes were more profitable, with more variability, than alfalfa- or corn-potato rotations. However, it must be kept in mind that the crop yields utilized were based on statewide averages. On any particular farm, it is unlikely that continuous potato yield can be maintained over time at 300 cwt/a without substantial increases in chemical inputs (pesticides and/or fertilizer). Conversely, continuous potato yields are expected to decline if inputs 51 remain constant (Wheeler, 1946; Emmond and Ledingham, 1972; McDole and Dallimore, 1978; O’Sullivan and Reyes, 1980). A second simulation was thus conducted to examine the specific effect of decreasing continuous potato yield, varying only mean potato yield and costs directly related to yield level (ie. $1.47/cwt). As expected, mean net income fell from $258,000 at 300 cwt/a to $116,000 at 225 cwt/a (Table 2.6), roughly following a linear trend. Thus a 25% reduction in mean potato yield resulted in a 55% reduction in mean net income. In comparison to the rotational systems shown in Table 2.5, continuous potatoes with a mean yield of 255 cwt/a (a 15% reduction from the baseline) generated a net income ($161,000) nearly identical to that from the alfalfa hay-potato rotation ($156,000). Agronomically, a 15% lower yield in a continuous versus rotational potato system is in agreement with the 12, 15, and 18% reductions for continuous potatoes (versus potatoes in rotation) reported by O’Sullivan and Reyes ( 1980), McDole and Dallimore (1978), and Emmond and Ledingham (1972), respectively. The variability in net income was considerably higher for continuous potatoes (tr-$168,000), with a mean yield of 255 cwt/a, compared to the alfalfa hay-potato rotation (”$106,000). This demonstrates that even at similar levels of net income, monocultural production inherently carries more economic risk than does a rotational system. The mean and variance components in economic analyses, once calculated, are commonly considered non-stochastic in nature; ie. that the mean is constant over time with observations being randomly distributed around the mean value. In reality, it is likely that these components change over time (McSweeney et a1., 1987). This concept seems particularly applicable to potato yield in monoculture, where the yield has been shown to decrease over time. The effect of such a change on mean net income is clearly shown in Table 2.6. However, the effect on income variability 52 Table 2.6. Mean and variability of net income from simulation (9:250) of continuous potato yield ranging from 225 to 300 cwt/a (potato yield normally distributed, coefficient of variationa12.9% for all yield levels). Mean potato yield (cwt/a) 225 240 255 270 285 300 ------------------------------- Net income, $1,000's ----------------~°------°°-- Mean 116 136 161 207 224 258 Lower :1 1m :1 96 116 140 186 203 233 Upper C1 limit 135 157 182 228 246 282 Standard deviation 154 163 168 168 177 195 Lower C1 limit 145 150 155 155 163 180 Upper Cl limit 173 179 184 184 195 215 CV, X 133 120 104 82 87 76 Net income range Minieun -162 -134 -96 -83 -66 -100 Maximum 578 766 762 720 806 860 T’Cl 8 95% confidence interval. 53 is much less dramatic; the same 25% reduction in the mean yield reduces the standard deviation of net income only 22%. Thus if potato yields decline in monoculture, the producer may be faced with lower net income levels without a proportional reduction in variability, thus increasing relative variability. Because producers are less likely to accept risk (ie. variability) at lower income levels (King and Robison, 1981), rotational systems are superior under these conditions. WW Net income from monocultural potato production is directly related to the yield of potatoes. As simplified in Equation [1], if yield, Y, falls without a proportional reduction in costs, C, net income also falls. The effect of skewed yields or prices on the distribution of net income is less certain and tends to be highly case specific. For example, Buccola (1986) demonstrated that net income may be skewed even though both price and yield are normally distributed. (This arises when the price and yield of a specific crop are correlated, an assumption not imposed here.) Alternatively, Tew and Reid (1988) found that income distribution may be normal when both price and yield exhibit significant skewness. Skewness was introduced into monocultural potato yields here because of the likelihood of skewness in yield distributions (Day, 1962) and because producer perception of risk is influenced by skewness (Collender and Zilberman, 1985). For each distribution, mean potato yield was 270 cwt/a and «34 cwt/a. The imposition of positive skewness (following a gamma distribution) in potato yield did not significantly change mean net income or income variability compared to normally distributed potato yields (Table 2.7). The normally distributed yields did lead to greater potential loss than with positively skewed yields (minimum net income- -$100,000 and -$79,000, respectively). As discussed by Collender and Zilberman (1985), highly risk-averse producers favor positively skewed distributions over negative skewness or normality for this reason; low values are less likely while 54 maintaining a long tail to the right (ie. high values from the distribution are not constrained). Mean net income and the standard deviation of net income were higher for the negatively skewed potato yields (mean-$228,000; «$183,000) than normal or positively skewed yields. This distribution may be more appropriate for monocultural crop production because yields above the mean value are restricted (in this case, maximum yield was 320 cwt/a, compared to 350 cwt/a for the gamma distribution), while there is a greater potential for catastrophic yields to occur (minimum yield=l80 cwt/a). However, there is no reliable data that indicate that this distribution more realistically describes monocultural potato yield over time in Michigan, compared to normal or positively skewed distributions. As a result of the long tail to the left of this beta distribution, the negatively skewed distribution also exhibited a greater potential loss (-$125,000) than the other two yield distributions. For the three potential distributions simulated here, we observed no consistent effect on relative income variability (CV-80 to 88%), and the imposition of skewed yield distributions did not result in skewed net income distributions. From this simulation, it is apparent that the effect of skewed yields on net income is less pronounced than the effects of the reduced mean yield levels examined in the previous simulation. CONCLUSIONS From the simulations performed here, we found that the profitability of rotational versus continuous potato production depends primarily on the potato yield level utilized. 1f the assumption is made, as it was in the initial simulation, that potato yield is unaffected by the preceding crop (ie. 300 cwt/a for all systems), continuous potato production generates more net income than any rotation. 55 Table 2.7. Effect of potato yield distribution on mean and variability of net income (mean yield 8 270 cut/a for all distributions). Yield distribution Normal Callas, Beta, positively skewed negatively skewed ----------------- Net income, 81,000's --------------" Mean 183 , 199 228 Lower c1 limitT 163 179 206 Upper CI limit 203 220 251 Standard deviation 161 165 183 Lower Cl limit 149 152 169 Upper Cl limit 177 181 201 CV, x 88 83 80 Met income range Mininml ~100 -79 -125 Maxi-us 662 726 833 T C1 I 95% confidence interval. 56 As continuous potato yield was reduced from 15 to 20%, rotations were superior to continuous potatoes; they generated net income at a level similar to continuous potatoes, but exhibited less variability in net income. The comparison of the two alfalfa-based rotations shows that the use of alfalfa solely as an N source for potatoes (rather than as a cash crop and an N source) reduces net income more than 25%, without reducing income variability. REFERENCES Brodie, B.B., J.M. Good, and W.E. Adams. 1969. Population dynamics of plant nematodes in cultivated soil: effect of sod-based rotation in Cecil sandy loam. J. Nema. 1:309-312. Buccola, S.T. 1986. Testing for nonnormality in farm net returns. Amer. J. Agr. Econ. 68:334-343. Collender, RN. 1989. Estimation risk in farm planning under uncertainty. Amer. J. Agr. Econ. 69:996-1002. Collender, R.N., and D. Zilberman. 1985. Land allocation under uncertainty for alternative specifications of return distributions. Amer. J. Agr. Econ. 67:779-786. Day, R.H. 1962. Probability distributions of field crop yields. J. Farm Econ. 47:713-741. El-Nazar, T., and BA. McCarl. 1986. The choice of crop rotation: a modeling approach and case study. Amer. J. Agric. Econ. 68:127-136. Emmond, G.S., and R.J. Ledingham. 1972. Effects of crops rotation on some soil-borne pathogens of potato. Can. J. Plant Sci. 52:605-611. Groya, F.L. and CC. Sheaf f er. 1985. Nitrogen from forage legumes: harvest and tillage effects. Agron. J. 77:105-109. Hazell, P.B.R. 1971. A linear alternative to quadratic and semivariance programming for farm planning under uncertainty. Amer. J. Agric. Econ. 53:53-62. Heady, EC. 1952. Diversification in resource allocation and minimization of income variability. J. Farm Econ. 54:482-496. Heady, E.O., and HR. Jensen. 1951. The economics of crop rotations and land use. Iowa Agric. Expt. Stn. Res. Bull. 383. Helmers, G.A., M.R. Langemeier, and J. Atwood. 1986. An economic analysis of alternative cropping systems for east-central Nebraska. Amer. J. Alter. Ag. 1(4):153-158. Hesterman, O.B., C.C. Sheaffer, D.K. Barnes, W.E. Leuschen, and J.H. Ford. 1986a. Alfalfa dry matter and nitrogen production, and fertilizer nitrogen response in legume-corn rotations. Agron. J. 78:19-23. Hesterman, O.B., C.C. Shaeffer, and El. Fuller. 1986b. Economic comparisons of crop rotations including alfalfa, soybean, and corn. Agron. J. 78:24-28. 57 58 Hesterman, O.B., MB. Tesar, J. Hilker, J.R. Black, E. DeVyst, and G. Schwab. 1989. RESEED: an economic evaluation for alfalfa re-establishment. Mich. State Univ. Coop. Ext. Serv. Integrated Management Software Program CP-023. King, R.P., and LJ. Robison. 1981. An interval approach to measuring decision maker preferences. Amer. J. Agr. Econ. 61:510-520. King, R.P., F.J. Benson, J.R. Black, and P.A. Held. 1989. Agricultural risk management simulator. Minnesota Ext. Serv. Software AG-CS-2577, Version 3.x. McDole, R.E., and CE. Dallimore. 1978. Continuation of cropping sequence studies on coarse textured soils in southeastern Idaho. Amer. Potato J. 55:221-226. McSweeney, W.T., D.E. Kenyon, and R.A. Kramer. 1987. Toward an appropriate measure of uncertainty in a risk programming model. Amer. J. Agr. Econ. 69:87-96. Musser, W.N., J. Ohannesian, and FR. Benson. 1981. A safety first model of risk management for use in extension programs. No. Cent. J. Agr. Econ. 3:41-46. Nott, S.B. G.D. Schwab, A.E. Shapley, MP. Kelsey, J.H. Hilker, and L.O. Copeland. 1988. 1988 crops and livestock budget estimates for Michigan. Dept. Ag. Econ., Ag. Econ. Report 508. O’Sullivan, J.O., and A.A. Reyes. 1980. Effects of soil fumigation, rotation, and nitrogen on yield, petiole Nos-N, and Verticillium wilt of potatoes. J. Amer. Soc. Hort. Sci. 105:809-812. Pope, R.D., and RF Ziemer. 1984. Stochastic effects, normality, and sampling errors in agricultural risk analysis. Amer. J. Agr. Econ. 66:31-40. Rowberry, R.G., and CW. Anderson. 1983. The profitability of continuous potatoes versus rotations including potatoes and other cash crops. Amer. Potato J. 60:503-510. ' Saloman, M 1962. Soil aggregation-organic matter relationships in redtop- potato rotations. Soil Sci. Soc. Amer. J. 26:51-54. Schurle, B.W., and B.L. Erven. 1979. The tradeoff between return and risk in farm enterprise choice. No. Cent. J. Agr. Econ. 1:15-21. Schwab, G.D., and K. Norgaard. 1988. Custom work rates for Michigan. Mich. State. Univ. Coop. Ext. Bull. E-2131. Shapley, A.E. 1987. Business analysis summary for potato farms: 1987 Telfarm data. Michigan State Univ. Agr. Econ. Report 522. Shapley, A., and A. Williams. 1988. Cost of producing tablestock potatoes in Michigan. Mich. State Univ. Coop. Ext. Bull E-2114. Shapiro, 8.5., and MB. Wilk. 1965. An analysis of variance test for normality (complete samples). Biometrika 52:591-611. 59 Stovall, J.G. 1966. Income variation and selection of enterprises. J. Farm Econ. 48:1575-1579. Tew, B.V., and D.W. Reid. 1988. Probability distributions of crop prices, yields, and gross revenue. Northeastern J. Agr. Econ. Res. 17:118-124. USDA. 1989. Agricultural prices: summary. PR1-3(89). Weinhold, A.R., J.W. Oswald, T. Bowman, J. Bishop, and D. Wright. 1964. Influence of green manures and crop rotations on common scab of potatoes. Amer. Potato J. 41:265-273. Wheeler, EJ. 1946. The residual cf f ect of crop rotations on potato yield and the presence of potato scab. Michigan Agr. Expt. Stn. Quart. Bull. 38:326-332. CHAPTER THREE EFFECT OF FORAGE LEGUMES ON PATHOGEN POPULATIONS AND TUBER QUALITY IN LEGUME-POTATO ROTATIONS ABSTRACT Research was conducted in Michigan in 1987-88 to evaluate the effect of forage legumes and corn on populations of Pratylenchus penetrans and Verticillium dahliae, and the influence legume-potato and corn-potato rotations and N fertilizer rate (0, 75, 150, 225 kg M“) on potato scab, black scurf, growth cracking and specific gravity of tubers. First year rotation crops included alfalfa, red clover, and birdsfoot trefoil as green manure and hay crops, sweetclover and hairy vetch as green manure, and corn. Potatoes were grown on all plots in 1988. All legumes (seeded in April, 1987) except sweetclover exhibited elevated p0pu1ations of P. penetrans (143-638 P. penetrans per g root + 100 cm3 soil) by July, 1987, compared to a corn-potato control. Population trends at the end of the growing season (November, 1987) were similar to those at mid-season. Despite high nematode populations under legume crops in November, 1987, no significant differences existed in soil P. penetrans population prior to planting potatoes in May, 1988. None of the first year rotation crops influenced soil inoculum levels of V. dahliae, which averaged 4 colony forming units (cfu) 3'1 soil in May, 1988. Red clover used as a hay crop in 1987 was associated with increased scab severity of the subsequent potato crop, with at least 50% more severely scabbed tubers than in other rotations. Incidence of black scurf following hairy vetch was 2 60 61 to 3-f old higher than following any other rotation crop. Growth cracking and specific gravity of tubers in 1988 were not affected by the previous rotation crops. Scab severity decreased in all rotations with increasing rates of N fertilizer (0 to 225 kg N ha"), while black scurf and growth cracking increased with N fertilizer rate to 150 kg ha". None of the rotations examined eliminated soil-borne pathogens or completely curtailed potato disease development, compared to corn-potato control. However, rotations including alfalfa or sweetclover did not increase the severity of potato scab or black scurf, and sweetclover exhibited lower nematode infestation than other legumes. In contrast, rotations including red clover hay or hairy vetch increased the severity of scab and black scurf, respectively, along with supporting higher nema- tode populations. INTRODUCTION The development of effective chemical pesticides, inexpensive nitrogen (N) fertilizer, and improved potato varieties have led to continuous potato (Solanum tuberosum L.) production on the same land in many places. Current concerns over the environmental and economic implications of high pesticide and N inputs in potato production have renewed interest in crop rotations, especially those including N- fixing legumes. However, there is little information available about the effect of alternative rotation crops based on either pathogen population trends during years when potatoes are not grown or tuber quality of the subsequent potato crop. Pratylenchus penetrans (Cobb) F ilip jev & Schuurmans-Stakh. is the most important nematode pathogen of potatoes on mineral soils in Michigan and is found on approximately 50% of the land used for potato production (Bird, 1981). The proliferation of P. penetrans during the growth of forage legumes in rotations with potatoes has justifiably caused concern over the potential damage to subsequent potato crops in rotation (Kimpinski, 1979). Nematode infection levels’in excess of 10,000 P. penetrans 3‘1 root dry matter have been observed in legumes like alfalfa (Medicago sativa L.) and red clover (Tri/olium pratense L.). These infection levels have been associated with forage yield reductions of more than 50 percent (Willis and Thompson, 1969; Thompson and Willis, 1975). Although forage legumes are susceptible to, and alternative hosts for, P. penetrans, little information is available comparing population trends of this pathogen for different legume rotation crops. There are several strategies for non-chemical control of P. penetrans. Like most soil-borne pathogens, populations of P. penetrans can be reduced by natural 62 63 attrition if sufficient time elapses between susceptible crops (Emmond and Ledingham, 1972). However, the wide host range of P. penetrans (Brodie et al., 1969; Francl et al., 1988) makes the selection of non-host rotation crops difficult. P. penetrans p0pu1ations may also be reduced by products released from decomposing plant residues, mainly N Hf (Heald and Burton, 1968; Walker, 1971), and more diverse soil microbial communities associated with rotations (Linford et al., 1938; Jonhson et al., 1969). It is also difficult to control Verticillium dahliae Kleb. and other fungal pathogens of potato using crop rotations. The long-term viability of V. dahliae microsclerotia in soil (Davis and McDole, 1978; Rowe et al., 1987) and the infection and colonization of weeds and other crops that do not develop typical wilt symptoms (Heale and Isaac, 1963) contributes to this difficulty. Inf ectivity and disease caused by V. dahliae is also influenced by P. penetrans. A synergistic interaction has been documented between V. dahliae and P. penetrans, where the presence of both patho- gens causes more damage than the sum of pathogens separately (Martin et al., 1982; Riedel and Rowe, 1985; Rowe et a1., 1987). Francl et a1. (1988) found that V. dahliae and P. penetrans alone reduced tuber yield up to 31 and 16%, respectively, while yield was reduced up to 63% when both pathogens were present. Streptomyces scabies (Thaxt.) Waks. & Henrici, the actinomycete that causes common potato scab, is also capable of long-term survival in the absence of a susceptible potato crop. Menzies (1963) classified S. scabies as a ”normal soil inhabit- ant” because of its ability to feed and reproduce saprophytically when host plants were not present. The severity of common potato scab is influenced by the time period between potato crops and, in some cases, the cr0p rotation, although the effect of these two factors has been inconsistent. Increases in scab severity have been shown many times as either the time between potato crops is shortened or as more 64 successive potato crops are grown (Goss and Afanasiev, 1938; Werner et al., 1944; Hooker, 1959). Wheeler (1946) also observed that scab increased with continuous potatoes, but potatoes following alfalfa or rye (Secale cereale L.) had significantly less scab than those following sweetclover (Melilotus spp.) or potatoes. Hooker (1959), however, found that the only crop to influence scab severity was potato. Because of the erratic control of scab with crop rotations, the most prevalent method of control- ling scab has been to maintain soil pH at levels unfavorable to S. scabies (<5.0;Ellison et a1., 1951) using ammonium-based N fertilizers. Because of the diverse host range and saprophytic ability of Rhizoctom'a solani Kuhn., research on cultural control practices has focused primarily on the composi- tion of plant material incorporated immediately prior to a susceptible crop. R. solani has been controlled in many crops by incorporating plant material with a high ca'rbon-to-nitrogen (C/N) ratio like barley (Hordeum vulgare L.) straw to immobilize soil N. Control has been observed at C/N ratios greater than 6 (compared to approxi- mately 15 for alfalfa), but more commonly at C/N ratios of 40-200 (Davey and Papavizas, 1959, 1963). The inhibitory effect of high C/N ratio plant material has been attributed to the poor competitiveness of R. solani for soil N, which can be reversed by the application inorganic N fertilizer (Papavizas and Davey, 1960; Papavizas, 1970), and to increased populations of antagonistic fungi (Papavizas and Davey, 1960). The research reported herein examines differences in pathogen populations and disease severity in two-year rotational potato systems, and was conducted within the framework of a field experiment on nitrogen management in legume-potato rotations. No treatments were imposed to specifically influence p0pu1ations of soil- borne pathogens, as short rotations are unlikely to adequately control any of the pathogens discussed here. Rather, rotation systems were evaluated for their poten- 65 tial to increase pathogen populations and disease severity. Specifically, the ob jee- tives of this research were: 1) to monitor P. penetrans and V. dahliae populations during and immediately following legumes in two-year legume-potato rotations; and 2) to evaluate the effect of first year rotation crop and second year N fertilizer rate on common scab severity, incidence of Rhizoctonia, and growth cracking and specific gravity of tubers in the subsequent potato crop. MATERIALS AND METHODS Crop production pragtiggs, Research was initiated in 1987 at the Michigan State University Montcalm Research Farm, near Entrican, MI, to examine pathogen populations trends and disease severity in legume-potato rotations. The experiment was conducted on a McBride sandy loam soil (coarse-loamy, mixed, frigid Alfie Fragiothods). Initial soil test results indicated: pH 5.9; 311, 579, and 849 kg ha'1 of K, P, and Ca, respectively; organic matter, 15 g kg"; and CEC of 5 cmol kg'l soil. Alfalfa (’Saranac’), red clover (’Michigan Mammoth’), birdsfoot trefoil (Lotus corniculatus L. ’Viking’), common sweetclover (common), hairy vetch (Vicia villosa Roth, common) and corn (Zea mays L.’Pioneer 3737’) were planted as first year rotation crops. Two treatments each of alfalfa, red clover, and birdsfoot trefoil were planted; one managed as a green manure left unharvested during the seeding year, and the other managed as a hay crop harvested two (birdsfoot trefoil) or three (alfalfa and red clover) times during the seeding year. Corn plots were divided into four subplots fertilized with 0, 75, 150, or 225 kg N ha“, as ammonium nitrate (NH4N03). Establishment and harvest procedures for legumes and corn are de- scribed elsewhere (Griffin & Hesterman, 1990; manuscript in preparationl). ‘ See Chapter One, pp. 6-12. 66 Following plowing on May 5, 1988, all plots (8 rows by 15.2 m) were planted to potatoes (’Shepody’), with a row spacing of 0.92 m, and seed pieces 0.25 m apart. Aldicarb (Temik 15G) was applied at 3.4 kg ai ha‘1 at planting. Split-plot N rates (same as 1987) were applied as 75 kg N ha'1 at planting and the remainder, if any, sidedressed prior to hilling on May 26, 1933. Metolachlor (Dual) and metribuzin (Lexone) were applied preemergence at 2.2 and 0.56 kg ai ha", respectively. Foliar insecticides and fungicides were applied to potatoes as needed during the growing season. During September 30-October 6, 1988, the center two rows of each subplot were harvested, and tubers were separated into; ’B’ grade (<114 g), ’A’ grade (115-280 g), oversize (>281 g), and culls (second growth, soft rot, etc.). Soil agg plant sampling (:9: pathogens, Soil samples were taken 4 times during the first year of the rotations to determine soil population densities of P. penetrans. Legume plots and corn plots that received no N fertilizer were sampled on May 5, July 23, November 5, 1987, and May 12, 1988. These subplots were sampled to evaluate the effect of rotation crop species independently of N fertilizer manage- ment. Fifteen to twenty 2.5 cm dia by 20 cm deep soil cores were taken randomly in each plot. Cores were composited, mixed thoroughly, and subsampled (approximately 500 g) for analysis. P. penetrans was extracted from soil using the centrifugation-flotation technique (Jenkins, 1964). Root samples were taken on July 23 and November 5, 1987, to determine root infestation of different rotation crops by P. penetrans by the shaker-incubation method. Soil samples taken in May, 1988, were also used to determine pre-plant inoculum levels of V. dahliae, using the dilution plate method described by Nicot and Rouse (1987), with an alcohol agar medium (Nadakavukaren and Homer, 1959). 67 WW Samples of approximately 5 kg of ’A’ size tubers were collected from each N rate subplot in 1988 for evaluation. Ten tubers were randomly selected from each sample, thoroughly washed, and evaluated for severity of common potato scab. Each of the ten tubers was placed in one of five scab severity classes based on the percent of tuber surface area exhibiting scab lesions; 0- 1, 1-5, 5-10, 10-25, and more than 25%. A weighted average scab severity was then calculated for each rotation crop/N rate combination. The same ten tubers used for scab evaluation were also evaluated for black scurf and growth cracking. If black scurf was observed on any of the ten tubers, a value of 100 was recorded; absence of black scurf was recorded as zero. The number of tubers (out of ten) exhibiting growth cracks was also recorded. Specific gravity was determined by the weight in air to weight in water ratio on 2 kg of tubers from the original sample. Experimental design was a randomized complete block with 4 replications. Treatment design was a split plot, with 1987 rotation cr0p as whole plots and 1988 N fertilizer rate as split plots. Rotation crop means were compared using Least Significant Difference (LSD) or Duncan’s Multiple Ranges test (DMRT). LSD or DMRT were not calculated unless main effect was significant at 0.10 level. Statisti- cal analyses of P. penetrans populations were performed on square root transformed data to equalize the variance among treatments. N rate means were compared using linear and quadratic orthogonal contrasts. RESULTS AND DISCUSSION Waging, The pre-plant population density (Pi) of P. penetrans in the soil of individual plots at the beginning of the two-year rotations (May, 1987) was similar across the experimental area, ranging from 38 to 60 P. penetrans 100 cm" soil. By mid-season (July, 1987), significant differences were observed in popula- tion densities of P. penetrans associated with the different rotation crOps (Table 3.1). Among legume crops, sweetclover supported the lowest P. penetrans population, while hairy vetch supported the highest (60 and 638 P. penetrans 100 cm" soil + g'1 root, respectively). This agreed well with the observations of MacDonald and Mai (1963) who observed population increases of 112 and 358% for sweetclover and hairy vetch, respectively. Populations associated with the remaining legume crops were similar, exhibiting 3 to 8-f old increases from Pi in May, 1987. In contrast to the increases in P. penetrans in legume treatments, populations under the growing corn crop were lower than Pi. By November, 1987, there were differences in P. penetrans populations associated with both rotation crop species and green manure or hay crop manage- ment of the same species. Relative population levels between rotation crops in November were similar to those found in July, 1987. Sweetclover and'hairy vetch still supported the lowest and highest P. penetrans populations, respectively, among legume crops. Likewise, P. penetrans in soil and roots of corn plots was lower than all legume species. The low population level associated with corn was contrary to the results of Dickerson ct a1. (1964), who observed more rapid root infection and 68 69 Table 3.1. Effect of rotation crops on population densities of £5agxigggfigg_ m and Win in 1987-1988 at the Michigan State University Montcalm Research Farm in Entrican, M1. V. 'a colony forming g&_pgggggag§ 100 cm”8 soil 0 g‘1 root units g'1 soil Rotation Crop 5/871' 7/87 11/87 51881’ 5/88 Alfalfa GM: 43 143bcd§1 112ef 10 4.0 Alfalfa May 50 424ab lbldef 18 4.0 Red Clover GM 48 335ab 202cde 32 2.5 Red Clover May 38 278bc 454ab 21 3.0 8. Trefoil GM 57 269bc 435abc 6 5.5 8. Trefoil May 44 375ab 308bcd 18 3.5 Sweetclover GM 47 60cd 107ef 21 5.0 Hairy Vetch GM 44 638a 679a 43 3.5 Corn 48 27d 47f 7 5 . 0 Prob.>F 115 0.012 0.0001 us ' MS ‘T Population density for soil only. 3 GM 8 Green Manure, no harvest during seeding year 5 Means with same letter in each column are not statistically different at 0.10 level using Duncan's Multiple Range test. 1 Duncan's Multiple Range test performed on square root transformed data. 70 population growth by P. penetrans on corn than on potato. The effect of legume harvest management on the population of P. penetrans was minimal. At the end of the seeding year, red clover hay supported a higher population than red clover managed as a green manure (Table 3.1), but no significant differences due to legume harvest were seen for alfalfa or birdsfoot trefoil. Research comparing legumes used as green manure and hay is limited, but MacDonald and Mai (1963) reported that infestation of hairy vetch and many other crops by P. penetrans increased as defoliation became more severe. Although we found that all legumes supported high p0pu1ations of P. penetrans, other research has indicated that yield reductions are unlikely because losses caused by P. penetrans were usually delayed until the second or third production year (Willis and Thomp- son, 1969; Thompson and Willis, 1975). No statistically significant differences in P. penetrans populations (soil only) due to 1987 rotation crOp were detected immediately prior to planting the 1988 potato crop. Although no significant differences existed, the lowest and highest populations (7 and 43 P. penetrans 100 cm‘3 soil) were observed following corn and hairy vetch, respectively. The similarity in P. penetrans soil populations immediately prior to planting potatoes means that, although high populations were associated with most legume crops in 1987, this would have little impact on the subsequent potato crop. VI gamma ggii ingnlgm lava]; [gnawing giflfing grng, There were no differences in soil inoculum levels of V. dahliae midway through the rotation cycle (May, 1988, immediately prior to the susceptible potato crop) related to the 1987 rotation crop (Table 3.1). As reported previously (cg. Davis and McDole, 1978; Rowe et a1., 1987), the absence of a susceptible potato crop for periods of 1 to 3 years is not sufficient to reduce the soil p0pu1ation of this pathogen. The experimental area 71 utilized had grown potatoes every second or third year for many years, and levels of V. dahliae in this area are sufficient to cause damage to susceptible crops. A mean soil inoculum level of 4 colony forming units (ef u) V. dahliae g"1 air-dried soil was observed. W No significant rotation crop by N rate interaction was observed for any tuber quality parameter examined. Thus, main effects of rotation crop and N rate will be discussed separately. Although potatoes had not been grown in this experimental area for three years prior to initiating rotations in 1987, symptoms of common scab were prevalent on the 1988 potato crop (consistently in excess of 5%, below which is considered acceptable for marketable tubers). This may be partially due to the use of a scab- susceptible potato variety (’Shepody’) in 1988. Because of the high incidence of potato scab, data are reported as percent of ’A’ grade tubers exhibiting scab lesions on more than 25% of surface area. Averaged across N rates, scab severity was influenced slightly (P-0.09) by the previous rotation crop (Table 3.2). Twenty-eight percent of tubers following red clover, as a hay crop, were severely seabbed. Rouatt and Anderson (1950) also noted that scab severity following red clover was similar to that in continuous potato, and much higher than in soybean (Glycine max Merr.)-potato rotations. Scab following red clover was significantly higher than following alfalfa green manure, birdsfoot trefoil hay and sweetclover green manure. Several previous reports have shown lower scab severity following alfalfa (cg. Wheeler, 1946). Except for potatoes following red clover hay, severe scab ranged from 10 to 19%. 72 Table 3.2. Influence of 1987 rotation crop on incidence of severe potato scab and black scurf, and on specific gravity of subsequent (1988) potato crop at the Michigan State University Montcalm Research Farm in Entrican, M1. Tubers with Plots with Tubers with Specific 1987 Crop Severe ScabT Black Scurf Growth Cracks Gravity ................... y .....--.-........-. Alfalfa GM: 11.9b 6.3b§ 5.0 1.080 Alfalfa May 13.8ab 8.3b 6.3 1.083 Red Clover GM 14.4ab 18.8b 6.9 1.083 Red Clover May 28.1a 18.8b 3.1 1.083 Birdsfoot Trefoil GM 16.3ab 8.3b 6.3 1.081 Birdsfoot Trefoil May 10.06 25.0b 1.3 1.082 Sweetelover 10.0b 12.5b 9.4 1.082 Hairy Vetch 16.8ab 62.5a 11.9 1.081 Corn 18.8ab 18.8b 4.4 1.082 Prob.>F 0.09 0.0005 MS MS ‘T Scab lesions on more than 25% of tuber surface area. 3 GM 8 Green Manure, not harvested during the seeding year. 5 Means with the same letter in each colm are not statistically different at 0.10 level using Duncan's Multiple Range Test. 73 Severe scab decreased linearly (Pa-0.012), from 21.6 to 8.8% of ’A’ grade tubers, as N fertilizer rate increased from 0 to 225 kg ha'l, suggesting that increasing rates of NHiNO8 reduced soil pH to levels inhibitory to S. scabies (Figure 3.1A). This inhibitory effect of N fertilizer was observed for all rotation crops. Blodgett (1940) also found that scab severity was reduced if N was applied to potatoes. Lapwood & Dyson (1966) used non-ammonium based N fertilizers at rates similar to those used in this study, and observed that scab increased with N rate, suggesting that the increase in scab was a result of delayed tuber initiation at higher N rates. It has also been suspected that high NH,‘+ levels in the soil, from N fertilizer, may directly inhibit S. scabies. Because ammonium nitrate was used as the N fertilizer source in this research, this may be a factor in decreased scab. However, several observation suggest that the N status of the growing potato plant and possibly reduced soil pH, not the N form in the soil, influenced scab severity. First, potatoes without N fertilizer following most rotation crops were clearly N def ieient by mid-July, 1988. And second, soil samples taken from selected rotations at this time contained extremely low or no detectable NH‘l' to a depth of 1.2 m, regardless of N rate (data not shown). Previous reports also support the use of ammonium- bascd N fertilizers in soil acidification to control S. scabies. The incidence of black scurf was also influenced by the crop preceding potato (Table 3.2). The percentage of plots with black scurf was generally between 6 and 25%. However, almost 63% of the potato plots following hairy vetch exhibited black scurf symptoms. Since R. solani, which causes black scurf, is known to be strongly affected by residue composition, the low C/N ratio of hairy vetch (approxi- mately 12) may have allowed survival until infection could occur. Black scurf increased as N rate increased to 150 kg ha" (Pa-0.001; Figure 3.1B), when averaged across all rotations. As shown by many researchers (Wright, 1941; '8 8 1 ... ta 1 10d Severe Scab (X) (A) 8 1:5 as lack Seu of P16 N o 8. a 10 u q d (B) Specific Gravity . (D) 7‘11 110 21:. M Applied in 1988 (kg/ho) Figure 3.1. Effect of fertilizer N rate on (A) severe scab, (B) black scurf, (C) growth cracking, and (D) specific gravity of ’A’ grade tubers at Montcalm. Regression equations for A, B, C, and D, respectively: Y-21.94-0.057N, 113-0.99; .Y-8.76+0.099N, R2=0.67; Y=1.68+0.039N, 123-097; and Y-1.0805+0.0000133N, R3=0.87, where Y=response and N-fertilizer N applied (kg ha"). 75 Davey and Papavizas, 1959, 1963), R. solani infection of many plant species increases when exogenous N is applied, primarily due to the increased saprophytic ability of R. solani when adequate soil N was available. Tuber quality parameters not directly influenced by soil-borne pathogens, like growth cracking and specific gravity, were not affected by previous rotation crop (Table 3.2). Growth cracking of tubers, generally caused by moisture or nutrient conditions that promote rapid tuber bulking, increased linearly (P<0.001) as N fertilizer rate increased to 225 kg ha’1 (Figure 3.1C). When no N fertilizer was applied, 2.2% of tubers exhibited noticeable cracks, while 10.8% of tubers were cracked when 225 kg N ha'1 was applied. Specific gravity also increased linearly (P-0.045), from 1.080 to 1.083, as N rate increased from 0 to 150 kg ha‘1 (Figure 3.1 D). CONCLUSION Although considerable research has examined the effect of individual pathogens on potato yield and tuber quality, production of potato is usually affected by multiple pathogens simultaneously. For this reason, it is valuable to examine the relative efficiency of different rotation crops in increasing or decreasing pathogen populations or disease severity. Table 3.3 summarizes the relative population of P. penetrans (November, 1987 sampling), and severity of scab and black scurf, each of which was significantly affected by rotation crop. Ranking is based on differences shown by Duncan’s Multiple Range Test (Tables 3.1 and 3.2), with the lowest ranking in each category having the lowest pathogen population or disease severity. The sum of individual rankings provides an indication of the general suitability of different rotation crops for disease prevention in potato production. None of the rotations examined is expected to completely control disease problems if 76 Table 3.3. Ranking of rotation crops based on population of root-lesion nematode (Movedaer, 1987), and severity of cm potato scab and black scurf at the Michigan State University Montcalm Research Farm in Entrican, MI. Root-Lesion Potato Black Ranking 1987 Crop Memetode Scab Scurf Sum -------------------- Ranking ---------------------- Alfalfa CM 1.5T 1.0 1.0 3.5 Alfalfa Hay 1.5 1.5 1.0 4.0 Red Clover CM 2.0 1.5 1.0 4.5 Red Clover May 3.0 2.0 1.0 6.0 Birdsfoot Trefoil CM 2.5 1.5 1.0 5.0 Birdsfoot Trefoil May 2.5 1.0 1.0 4.5 Sweetelover 1.5 1.0 1.0 3.5 Hairy Vetch 4.0 1.5 2.0 7.5 . Corn 1.0 1.5 1.0 3.5 T’ Ranking based on Duncan's Multiple Range Test; within each column, rankings more than 1.0 units apart are significantly different at 0.10 level of probability. 77 used as the sole management strategy. Alfalfa (as green manure or hay), sweetclover, and corn were equally beneficial (ie. lower sum ranking) in rotations, with similar sum rankings of 3.5 to 4.0. Red clover and hairy vetch, both as green manure, had the highest sum rankings and tended to significantly increase all disease parameters examined, and thus would not be recommended if disease problems were evident. The other rotations, which had similar sum rankings, did not profoundly affect any parameter. With these rotations, neither pathogen control nor increased disease severity would be expected. REFERENCES Bird, G.W. 1981. Management of plant parasitic nematodes in potato production. pp. 223-243 in J.H. Lashomb and R. Casagande (ed.) Advances in potato pest management. Hutchison Ross Publ. Co., Stroudsburg, PA. Blodgett, RM. 1940. A second report on the ef feet of agronomic practices on the incidence of Rhizoctonia and scab of potatoes. Amer. Potato J. 17:290-295. Brodie, B.B., J.M Good, and W.E. Adams. 1969. Population dynamics of plant nematodes in cultivated soil: effect of sod-based rotations in Cecil sandy loam. J. Nema. 1:309-312. Davey, CB, and GB. Papavizas. 1959. Effect of organic soil amendments on the Rhizoctonia disease of snap beans. Agron. J. 51:493-496. Davey, CB, and GB. Papavizas. 1963. Saprophytic activity of Rhizoctonia as affected by carbon-nitrogen balance of certain organic soil amendments. Soil Sci. Soc. Amer. Proc. 27:164-167. Davis, J.R., and RE. McDole. 1978. Influence of cropping sequences on soil-borne populations of Verticillium dahliae and Rhizoctonia solani. pp.399- 405 in B. Schippers and W. Gams (ed.) Soil-borne plant pathogens. Academic Press, London. Dickerson, O.J., H.M Darling, and GD. Griffin. 1964. Pathogenicity and population trends of Pratylenchus penetrans on potato and corn. Phytopathology 54:317-322. Ellison, J.H., W.C. Jacob, and HS. Cunningham. 1951. Effect of soil reaction on the performance of certain scab-resistant and susceptible potato varieties. Amer. Potato J. 28:721-727. Emmond, G.S., and R.J. Ledingham. 1972. Effects of crap rotation on some soil-borne pathogens of potato. Can. J. Plant Sci. 52:605-611. Francl, L.J., R.C. Rowe, R.M Riedel, and L.V. Madden. 1988. Effects of three soil types on potato early dying disease and associated yield reduction. PhytOpatholo- gy 78:159-166. Goss, R.W., and MM Afanasiev. 1938. Influence of crop rotations under irrigation on potato scab, Rhizoctonia, and Fusarium wilt. Nebraska Agr. Exp. Stn. Bull. 317. 78 79 Heald, CM, and G.W. Burton. 1968. Effect of organic and inorganic nitrogen on nematode populations in turf. Plant Dis. Rep. 52:46-48. Heale, J.B., and 1. Isaac. 1963. Wilt of lucerne caused by species of Verticillium. IV. Pathogenicity of V. albo-atrum and V. dahliae to lucerne and other crops; spread and survival of V. albo-atrum in soil and in weeds; effect upon lucerne production. Ann. Appl. Biol. 56:439-451. Hooker, WJ. 1959. Survival of Streptomyces scabies in peat soil with various cover crops. Phytopathology 46:677-681. Jenkins, W.R. 1964. A rapid centrifugal-flotation technique for separating nematodes from soil. Pl. Dis. Rep. 48:692. Johnson, L.F., A.Y. Chambers, and HE. Reed. 1967. Reduction of root-knot of tomatoes with crop residue amendments in field experiments. Pl. Dis.Rep. 51:219-222. Kimpinski, J. 1979. Root lesion nematodes in potatoes. Amer. Potato J. 56:79-86. Lapwood, D.H., and P.W. Dyson. 1966. The effect of nitrogen on the formation of potato tubers and the incidence of common scab (Streptomyces scabies). Plant Path. 15:9-14. Linford, MB., F. Yap, and J.M Oliviera. 1938. Reduction in soil populations of the root-knot nematode during decomposition of organic matter. Soil Sci. 45:127- 141. MacDonald, D.H., and W.F. Mai. 1963. Suitability of various cover crops as hosts for the lesion nematode, Pratylenchus penetrans. Phytopathology 53:730-731. Martin, MJ., R.M Riedel, and R.C. Rowe. 1982. Verticillium dahliae and Pratylenchus penetrans: interactions in the early dying complex of potato in Ohio. Phytopathology 72:640-644. ‘ Menzies, JD. 1963. Survival of microbial pathogens in soil. Bot. Rev. 29:79-122. Nadakavukaren, MJ., and CE. Horner. 1959. An alcohol agar medium selective for determining Verticillium microsclerotia in soil. Phytopathology 49:527-528. Nicot, P.C., and DJ. Rouse. 1987. Precision and bias of three quantitative soil assays for Verticillium dahliae. Phytopathology 77:875-881. Papavizas, G.C. 1970. Colonization and growth of Rhizoctonia solani in soil. pp. 108-122 in LR. Parmeter, Jr. (ed.) Rhizoctonia solani, biology and pathology. Univ. of California Press, Berkeley, CA. 80 Papavizas, GB, and CB. Davey. 1960. Rhizoctonia disease of bean as affected by decomposing green plant materials and associated microfloras. Phytopathology 50:516-522. Riedel, R.M, and R.C. Rowe. 1985. Lesion nematode involvement in potato early dying disease. Amer. Potato J. 62:163-171. Rouatt, J.W., and R.G. Atkinson. 1950. The effect of the incorporation of certain cover crops on the microbiological balance of potato scab infested soil. Canad. J. Res. Sect. C. 28:140-152. Rowe, R.C., J.R. Davis, ML. Powelson, and DJ. Rouse. 1987. Potato early . dying: causal agents and management strategies. Plant Disease 71:482-489. Thompson, L.S., and CB. Willis. 1975. Influence of f ensulf othion and f enami- phos on root lesion nematode numbers and yield of forage legumes. Can. J. Plant Sci. 55:727-735. Walker, J.T. 1971. Pratylenchus penetrans (Cobb) populations as influenced by microorganisms and soil amendments. J. Nema. 1:260-264. Werner, H.O., T.A. Keisselbach, and R.W. Goss. 1944. Dry-land crop rotation experiments with potatoes in northwestern Nebraska. Nebraska Agr. Exp. Stn. Bull. 363. Wheeler, EJ. 1946. The residual effect of crop rotations on potato yield and the presence of potato scab. Mich. Agric. Exp. Stn. Quart. Bull. 38:326-332. Willis, C.B., and LB. Thompson. 1969. Effect of root-lesion nematode on yield of four forage legumes under greenhouse conditions. Can. J. Plant Sci. 49:505-509. Wright, E. 1941. Control of damping-of f of broadleaf seedlings. Phyt0patholo- gy 31:857-85. HICHIGRN STQTE UNIV. LIBRQRIES \IHIWHIIH \ll‘ ”Ill” I11 I? Ill W1 |\ IN!” “NW |\ 1 1| 31293007908480