Species
Determinate
Indeterminate
Bean
9 . 7 3
1 0 . 6 2
Soy bean
3 . 9 7
3 . 9 6
Mean
6 . 8 5
7 . 2 9
Type
Species
F
-test
ns
* * CV( % )
=6 . 8
Mean
1 0 . 1 7
3 . 9 7
Type
*Species
ns
-.,J wGrowth Type
Species
Determinate
Bean
3
•5 1
Soy bean
1
•3 4
Mean
2 . 4 3
Type
F-test
* * C V( % ) =
1 8 . 8
Indeterminate
5 . 3 6
2 . 0 0
3 . 6 8
Species
* *Mean
4 . 4 3
1
•6 7
Type
*Species
ns
-.1 ..,.biological yield to seed yield was very similar as indicated by t he similar harvest indices ( Table 3 . 1 3 ) . Nit rogen utilization expr e s sed as nitrogen harvest indic e s was a l s o simi lar between the two typ e s (Table 3 . 14 ) .
3 . 4 D I SCUSSI ON
3 . 4 . 1 P lant growt h : Growth curves have been used as an approach t o plant growth analysis in this experiment because i t o ffered a number of advantage s (Hunt , 1 982 ) . Some o f the advantages which a r e direc t ly relevant to this experiment are i ) t he difficulties involved in t h e pairing o f plants when applying t he c lassical formulae c an be avoided , ii ) informat ion from a l l sampling occasions have been used i n
determining the value o f the derived quantities , whereas in the
c lassical method only data from two consecu t ive harves t s c a n be u s ed . The dif ference between the two growth types of both spec i e s in their dry matter accumulation wa s quite similar ( Figure 3 . 1 a and b ; Table 3 . 1 )
i . e .
tLe indetermina t e cultivars accumulated signif ic antly more dry-mat ter than the det erminate cultivars at the late grainfilling stage . However , there was not a c omplet e c e ssation o f
accumu lation o f dry-matter i n the vegetat ive por tion of the plant a t the beginning of f lowering for the determinate cultivars . The r e s u l t is similar t o that reported by Egli and Leggett ( 1 9 7 3 ) in the ir experiment using field-grown soybeans over two seasons .
Since total dry-ma t ter yie ld depends on the deve lopment o f leaf area and the rate o f net phot osynthesis ( Turner and B e gg , 1 98 1 ) , the higher leaf area o f the indet e rminate types ( Figure 3 . 3a and b ) reflec t s direc t ly the advantage o f the indeterminate typ e s in
Growth type
Species
Determinate
Indeterminate
Bean
0 . 6 6
0 . 5 2
Soy bean
0 . 6 1
0 . 5 5
Mean
0 . 6 4
0 . 5 4
Type
Species
F-test
n s
* * C V( % )
=5 . 2
Mean
0 . 5 9
0 . 5 8
Type
*Species
n s
-..) 0'\Growth type
Species
Determinate
Indeterminate
Mean
Bean
0 . 8 8
0 . 7 7
0 . 8 2
Soy bean
0 . 7 4
0 . 7 5
0 . 7 4
Mean
0 . 8 1
0 . 7 6
Type
Species
Type
*Species
F-test
ns
ns
ns
CV
{ % )
=9 . 9
-J -J
producing dry-ma t ter . C onsequent ly , the vegetative dry weight ( le a f +
stem) of the indeterminate cultivars of both spec ies was about twi c e that of the determina te cultivars at physiological maturi t y ( Figure 3 . 1a and b ) . A c omparison of total dry weight be tween t he two g rowth types at each growth stage ( Table 3 . 1 ) showed significant d i f ferenc e s only in the later growth stages o f soybean . Similarly there was no difference in leaf area between the two types during the e arly growth stages ( Table 3 . 3 ) . The s e differenc e s between the two growt h t yp e s we.,-e similar to those reported by Chaturvedi
et a Z .
( 1 980 ) ; i n cowpe a the indeterminate cultivars produced more dry-ma t t e r i n leaves , s t em and t otal plant than those of determinate cultivars .The net a ssimi l a t ion rate (NAR) c a lculated from f i t ted curves showed an interest ing pattern ( Figure 3 . 4a and b ) . The NAR valu e s o f indet erminate cultivars of both bean and soybean increased at the later stage of reproduc t ive growt h . Koller
et a Z .
( 1 9 7 0 ) int erpre t ed the increase in NAR during the lat t er part o f the growing season a s a response o f the pho t o synthetic aparatus t o an increased demand for assimilates , which was due to the rapid growth o f the seed . Sharmaet a Z .
( 1 982 ) reported that NAR ' s from pod development to maturity o f sixteen soybean cultivars were strongly correlated wi th yield .One of the maj or difference between the two growth type s was root growt h . The indeterminate cult ivars produced more root dry weight than those of the determinate cultivar s at every growth s tage compared ( Figure 3 . 5 ; Table 3 . 4 ) . Cultivars of soybeans were found to differ considerably in root growt h , e specially in pro li f erat ion during late development ( Raper and Barber , 1 970 ; Mitche ll and Rus s e l l , 1 9 7 1 ) . However these workers did not find any relationship between yielding
ability and nutrient a s s imilation due to dif ferenc e s in root ing behavior . More recently Silberbush and Barber ( 1 984 ) repo r t ed t ha t cultivars with more ext ensive root syst ems absorbed more P and K and this was found t o be a s sociated with higher grain yield in s oybean . 3 . 4 . 2 f ixation and nitrogen distribution :
Variat ion in the ability of grain legumes to fix nitrogen has been reported both between and within spec ies ( Graham , 1 98 2 ) . I n this experiment , beans had a lower activity than soybeans ( Figu r e 3 . 5 a and
HowW..tY) (IJ tiAl .eV\J totctl
Vl'trroyV\ Cl� �o1� t'fp-tJ
wa.� $ti'Vlilo.Y CFi�
L<.Y-L--:;, .ce:;) ,
b ) . Similar observations were quoted by Graham and Halliday ( 1 9 7 7 )t\
both in t erms o f total a c t ivity and specific activity. A d i f ference in total nitrogen fixat ion activity after f lowering betwee n the two growth types was found only in soybean ( Table 3 . 5 ) . There have been only a f ew studies which compared nitrogen fixation direc t ly b e tween the two growth type s ; Hardy
et a Z .
( 1 9 7 1 ) reported that d et erminate and indeterminat e soybeans o f similar maturity dates had s imilar nitrogen f ixat ion activity . In contrast Graham ( 1 98 1 ) quot ed f i e ld studies which indicated that the indeterminate cultivars o f bean( Phaseo Zu s vu lgaris )
were consistant ly superior to that of mo s t determinate bush types i n total nitrogen f ixat ion .When t o t a l plant nitrogen ( dependent variable ) i s r egre ssed with nodule dry we ight ( independent variable ) , the inter c e p t r e pr e sents
the amount of nitrogen actually absorbed from the original seed and the soil and the regr e s s ion coef ficient ( B 1 ) is an 'index o f ni t rogen fixing e f ficiency� Pioneering thi s method , Dobereiner ( 1 9 6 6 ) f ound that this 'index o f nitrogen f ixing effic iency ' seemed to be c o n s t ant for each plant spec ies and appeared to be independent of mo s t
experiment the determina t e cultivars of both species had s igni f i c a nt ly higher regre ssion coe f f icients than tho se of the indeterminate
cult ivars ( Figure 3 . 9 ) . Thi s would sugge s t that the nodul e s o f the determinate cultivar of both bean and soybean in thi s experiment are more e fficient than tho s e o f the indeterminate cultivar in f ixing nitrogen .
Since in thi s experiment no mineral nitrogen was added a ft e r the first week from seed ling emergenc e , and the p o t mixture had very low organic nitroge n , the change in t o t a l plant nitrogen wit h t ime should be a relatively accurate means of determining the t o t a l amount of nitrogen f ixed as we l l . In this case the indetermina t e cultivars of both species had signi ficantly higher total plant nitrogen than the determinate cult ivars ( Figure 3 . 1 0 ) .
The similarlity between acetylene reduction rate , but difference in total plant nitrogen , between the two bean c u lt ivar s could par t ly be due t o t he actual acetylene reduc tion data whic h
showed high plant to plant variability . Nodule weight ( fresh o r dry ) has t raditionally been used to indicate dinitrogen f ixat io n in l e gumes
( Bell and Nutman , 1 9 7 1 ; Brockwe l l , 1 9 7 1 ) . In soybean the higher nit rogen fixation of the indeterminate cult ivar was due to a higher nodule dry weight ( F igure 3 . 7 ) , s ince the 'nitrogen f ixing
efficiency ' , as indicated by the regres sion t echnique , was lower in the indet erminate cultivar ( Figure 3 . 9 ) . On the other hand , the superior nodule growth in the ind e terminat e bean cultivar did not result in higher nitrogen f ixation rate . Thi s could , in part , b e due to the nodul e activity in the bean being more sensit ive t o
environmenta l variability e . g . temperature than nodule dry weight ( Rennie and Kemp , 1 98 1 ) .
The t o t a l plant nitrogen in each cultivar was highly dependent on the total plant biomass ( correlation coefficient s > 0 . 94 ) . A similar relationship in soybean was previously reported by Hanway and Weber ( 1 97 1a ) ; S paeth and Sinc lair ( 1 983 ) ; and Anderson a nd Vas i l a s
( 1 985 ) . Many studies examining the distribution in nitrogen by soybean have shown that the loss o f nitrogen from vegetative t i s sue coinc ides with the accumulation o f nitrogen in seed ( i . e . Hanway and Weber 1 9 7 1c ; Derman
et a l . ,
1 9 78 ) . The rat io o f nitrogen in the grain to t o t a l above ground plant nitrogen or 'nitrogen harves t index ' wa s not s igni ficantly different between the two growth types ( Table 3 . 1 4 ) . Thi s indicate s that the distribut ion o f nitrogen to grain was related to the size o f the nitrogen poo l , which was , in turn , c lo se ly related to the dry weight rather than to the dif ference in nitrogentranslocation .
3 . 4 . 3 Yield and yield component s : C omparisons o f yie ld d i ff erence s between determinat e and indeterminate legume types have recorded
similar values in soybean ( Egli and Legge t t , 1 97 3 ) , higher valu e s in indetermina t e type of cowpea ( Chaturvedi
et a l . ,
1 980 ) or higher values in the det erminate type of soybean ( Hickset a l . ,
1 96 9 ) . The result s from this present experiment showed that the indetermina t e type o f both species out-yielded the det erminate counterpart (Tables 3 . 8 and 3 . 9 ) . This advantage mainly came from a higher number of pods per plant (Table 3 . 10 ) . The production of yield in a grain cropdepend s upon the accumulation o f dry-ma t t e r in seeds . The amount of seed yie ld produced is the product of the number o f seeds and their size , determined by the rate and dur a t ion of dry-matter acc umulat ion (Wein and Ackah , 1 97 8 ) . In thi s experiment the reproduct ive period o f
the two growth types o f each species was the same , so this indic a t e s that the rate of dry-ma t ter accumulation into the seed was likely t o b e the key fac tor i n c ausing the yield dif ference . The indetermi na t e types could achieve higher yie ld because they produced a larger source capac ity , as well a s bene fit ing during the early stages of pod f illing f rom the canopy which s t i l l had a large proportion o f r e l a t ive ly young leaves . In c ontra s t , determinate types have to derive muc h o f their photosynthat e from ageing leaves ( Shibles , 1 980 ) . Assessment o f the 'harvest index ' could ident i fy a pattern i n the relationships
between the total bioma s s of the mature plant and the fract ions cont ributed from the seeds . There was no difference in the harve s t index between the two growt h types o f both species i n the present experiment ( Table 3 . 1 3 ) . Previous work had indicated that the harve s t index o f a cultivar was a stable charact er i s t ic irrespe c t ive o f
substantial differenc e s in environmental conditions during plant growth ( Spaeth
et a l . ,
1 984 ) .Since plant s in thi s experiment were large ly dependent on symbiosis to satisfy t heir nitrogen demand , then the dif ference in total nitrogen between two growth type s ( Figure 3 . 8 ) c ould only be due to the difference in nitrogen fixation . It had been r e ported tha t where cowpea is dependent o n symbiotically-f ixed nitrogen , superior
nitrogen fixing geno type s were also higher yielding ( Graham and S c ot t , 1 983 ) . The resul t s in thi s experiment could also draw similar
CHAPTER 4
A c omparative study o f nitrogen f ixat ion , nitrogen
distribution and yield of two soybean cultivars with determi nate and indet erminat e growth type s .
AB STRACT
Two s oybean cultivars Matara ' ( de terminate ) and Amsoy '
( inde t e rmina t e ) were studied in c ontrolled environment conditions t o identi f y the differenc e s in nitrogen fixa t i o n , nitrogen d i s tribu t ion , nitrogen re-distribution and yie ld . The emphasis in thi s experiment was to d e t ermine the key factors tha t may contribut e towards the difference between the two growth type s .
The indet ermina t e cultivar produced twice a s much s eed yield t he determinate cultivar . Vegetative dry-ma t t e r component s o f the ind e terminat e cultivar were significant ly higher than those o f the determina t e cult ivar . This was largely because the indet e rmina t e cultivar had more leaf area per plant , a longer period o f root growth and more nodules than that of the determinate cultivar .
The superiority o f the indeterminat e cultivar in a c e tylene reduct ion ac tivity came primarily from a rapid increase in the nodule mass , since the specific acetylene reduct ion act ivity of the
determinat e cultivar was generally higher than that of the
indet ermina t e cultivar . There was a linear relationship between 4integrated nitrogen fixation activity ' and the nitrogen accumulat ion
About 30-40% o f seed nitrogen o f both cultivars c ame f r om re distribution . The s t em o f the indetermina te cultivar re-dis t r ibut ed a higher proportion of nitrogen t o the seed than that o f the det ermina t e cultivar .
Among several plant characters t e st ed ( viz . dry-wei ght o f root s , nodules , s t ems l eaves and pods , leaf area per plant , a c e t y lene reduc tion a c t ivity and t o t a l plant nitroge n ) leaf area was identi fied as one of the key fac t ors in determining the difference between t he two growth type s .
4 . 1 INTRODUCT ION
From t he previous experiment ( reported in Chapter 3 ) , i t wa s evident that the indet e rminate cultivars o f both bean and s oybea n produced approxima t e ly t hree times more t o t a l plant nitrogen than the determinate cultivar s . However , only the indeterminat e soybean s howed signif icant ly mor e acetylene reduction a c t ivity than tha t o f the
determinat e cultivar . The indeterminate cultivars o f both bean a nd soybean also had higher nodule dry weight s than tho s e of the
determinat e c u l t ivars . With these advant ages , the indetermi na t e bean and soybean cult ivars produced two and three t ime s more seed yie ld respectively than in the determinate cult ivars .
However , there are disagreement s in the literature about which growth types had an advantage in yield . For example , Egli a nd Leggett ( 1 9 7 3 ) and Beaver
et a l .
( 1 985 ) recorded s imi lar values of grain yield between the two growth types in soybea n , whereas Hickset a l .
( 1 9 6 9 ) report ed higher values in the determinat e type . Cha turvediet a l .
Furthermore , Beaver and Johnson ( 1 98 1 ) f ound that the determina t e cultivars performed le s s predictably than the indeterminat e cult i va r s when they compared the yie ld stability o f the two growth types adapted to the northern United S t a te s . So the yield difference between t he two growt h types may not only be due to genetic differenc e s but a ls o t o other factors such a s environmental c onditions .
Several worker s have estimated that between 50 and 6 0 % o f the nitrogen in seed comes f rom the re-distribution of nitrogen from vege t a t ive t i s sue ( Hanway and Weber , 1 9 7 1b ; Egli
et a Z . ,
1 97 8a ) .Increasing the quantity o f nitrogen ava ilable for grain development in soybean could be accomp l ished , therefore , by increasing whol e plant nitrogen accumulation and /or increasing the proport ion o f vegetat ive nitro gen translocated to t he developing grain ( Jeppson
et a Z . ,
1 9 7 8 ) .In a previous experiment , the indetermina t e cultivars had a great e r mass o f vegetative plant part s
( i . e .
leaf and stem ) and hen c e they also had bigger nitrogen pools . I t would also be o f int ere s t to know if the re-distribut ion o f nitrogen in the two growth typ e s wasdifferent : variability for nitrogen translocat ion from vegetative tissues to the developing grain may exi s t between the two c u l tivar s or conversely , differenceS in seed nitrogen cont ent may be mai nly due t o difference s i n sizes o f the nitrogen poo l s .
The obj e c t ives o f this experiment we re
1 ) To identify the differenc e s in nitrogen f ixation , nit rogen distribution and yie ld betwe en the two growt h types ( with the emphas i s