Published July, 2001

SOYBEAN
Effects of Photoperiod on Growth and Development of Soybean
Floral Bud in Different Maturity
Lingxiao Zhang,* Ruifang Wang, and John D. Hesketh

ABSTRACT croscopic dissection (Evans, 1969). However, the effect

Growth and development of most soybean [Glycine max (L.) of photoperiod on soybean floral bud growth, as a sole
Merr.] plants is sensitive to photoperiod effects. Therefore, it is impor- event, is less discussed. Some efforts have been made
tant to understand and quantify the processes involved. Studies were to understand and quantify the thermal–photoperiod
conducted to determine growing degree days required for growth of requirement for the rates of floral bud growth of differ-
the floral bud from initiation to open flower of different maturity ent soybean MGs (Ephrath and Hesketh, 1991; Zhang
groups (MGs) in both field and controlled environments. Eighteen et al., 1993, 1995). The objective of this experiment was
soybean strains, including ‘Clark’ back-cross near-isolines differing in
to further understand the thermal–photoperiod require-
maturity, were sown in the field at five different dates in 1992 and
1993. In growth chamber studies, plants of two strains differing in
ments of soybean cultivars with different MGs so that
maturity were used and moved to 12-, 14-, and 16-h daylengths after more information can be used for predicting phenology
floral buds were initiated in 12 h. Results from both field and growth and culture management to properly apply optimal cul-
chamber studies indicated that planting dates had a significant effect tural practices.
on the thermal requirement for bud growth in the late-maturing strains
used. Shorter photoperiods during the bud growth period accelerated
growth rates to open flower. Furthermore, in the growth chamber MATERIALS AND METHODS
study, flowering was inhibited under 16-h daylength in a late-maturity
(MG V) strain when plants were transferred immediately after floral Field Experiments
bud initiation (FBI) under 12 h. Plants remaining 8 d after FBI before In 1992 and 1993, a total of 18 soybean varieties, maturity
they were transferred to 16 h were not significantly delayed in flow- group (MG) ranging from 00 to VIII, were planted in a Flana-
ering. This study indicated that photoperiod length and treatment gan silt loam (fine montmorillionitic, mesic Aquic Arguidoll)
duration affects soybean FBI and floral bud development in a quanti- soil at the University of Illinois Crop Science South Farm at
tative way, which resulted in a profound photoperiod response in late Urbana, IL. Seven of them will be discussed in this paper.
maturity-group soybean under field conditions. Their MG ranking, associated maturity genes (if known), and
stem termination genes are listed in Table 1. There were five
planting dates, ranging from early May to late July, in both
he understanding of photoperiod ⫻ temperature
T interaction in controlling flowering processes is one
of the key problems in soybean growth and production.
years (Table 2). Plots were 8 m long with 0.76 m between
rows and 0.05 m between plants in the row. Three replications
were used in a randomized complete block design.
Understanding and quantifying this interaction often Plants were dissected using a light microscope to determine
directly affects soybean breeders and producers when when a knob-like structure of a floral bud primordium first
appeared on the plant (FBI). Flowering dates were recorded
selecting varieties, determining plant dates, and pre- when 50% of plants in a plot had open flowers. Fresh and
dicting flowering dates, maturity dates, and final yields. dry weights per plant, plant heights, leaf number (or node
Previous studies from laboratory and fields have shown number on the main stem), and leaf area of 10 plants were
the effects of photoperiod on flowering dates (and sub- recorded on the flowering date for each plot.
sequent phenological events) and final yields (Board The accumulative growing degree days (GDD) were used
and Hall, 1984; Gregan and Hartwig, 1984; Egli et al., to estimate the thermal requirement of each growth stage.
1989; Jones and Liang, 1978; McBlain et al., 1987; Sin- Accumulative GDD are calculated by the sum of averages of
clair et al., 1991; Thomas and Raper, Jr., 1976; Wang et daily high (ⱕ30⬚C) and low (ⱖ10⬚C) temperatures and then
al., 1987). subtracting a base temperature, which is 10⬚C in this case.
The period from seedling emergence to flowering Emergence dates were used instead of planting dates for the
consists of two phases: (i) emergence to floral bud initia- beginning of the first growth stage because of seedling emer-
tion (FBI) and (ii) FBI to open flower. Most research gence variations for different planting dates due to soil condi-
tions (especially water availability).
in the past has focused on mechanisms for photoperiod
Because the emergence dates of individual correlated plant-
and temperature effects on FBI, as determined by mi- ings of the 2 yr were close (Table 2), the average dates of 2 yr
were used for calculations and statistical analysis. Average
L. Zhang, Delta Res. and Ext. Cent., Mississippi State Univ., P.O. photoperiods (sums of daily photoperiods divided by days)
Box 197, Stoneville, MS 38776; R. Wang, Dep. of Agron., Beijing were also calculated for emergence to FBI and FBI to flow-
Agric. Univ., Beijing, China; and J.D. Hesketh, Dep. of Crop Sci., ering (Table 3).
Univ. of Illinois, Urbana, IL 61801 and USDA-ARS, 1201 W. Good-
win Ave., Urbana, IL 61801. Received 8 Aug. 2000. *Corresponding
author ([email protected]).
Abbreviations: FBD, floral bud development; FBI, floral bud initia-
Published in Agron. J. 93:944–948 (2001). tion; GDD, growing degree days; MG, maturity group.

944
 ZHANG ET AL.: EFFECTS OF PHOTOPERIOD ON SOYBEAN FLORAL BUD DEVELOPMENT 945

Table 1. Summary of soybean strains planted in 1992 and 1993, Table 3. Approximate average photoperiod experienced from
with maturity group (MG) ranking and associated termination floral bud initiation (FBI) to 50% of plants flowering for five
genes. planting dates and six Clark near-isolines differing in maturity
Flowering Stem group (MG) genes. The sum of daily photoperiods was divided
Strain MG gene termination by days to get the average.
L71-920 I e1e2e3 Dt2 Planting
L63-3117 II e1e2e3 Dt2
Strain MG 1 2 3 4 5
L63-2404 III e 1 E2 e 2 Dt2
Clark-63 IV e 1 E2 E3 Dt2 h
L74-441 VI⫹ E1e2e3 Dt2
L71-920 I 15.5 15.5 15.4 14.8 14.2
L65-3366 V E 1 E2 E3 Dt2
L63-3117 II 15.6 15.5 15.2 14.7 14.2
Cook VIII
L63-2404 III 15.5 15.5 15.1 14.7 14.0
Clark-63 IV 15.5 15.3 14.9 14.6 13.8
L74-441 VI⫹ 15.2 14.9 14.6 14.2 13.4
Growth Chamber Experiment L65-3366 V 15.1 14.7 14.4 14.0 13.4
Seeds of Clark near-isolines L63-3117 (MG II) and L65-
3366 (MG V) were sown in 30 and 50 pots (0.3 m high and GDD for both years for the early MG genotypes. Colder
0.2 m diam.), respectively. The potting mix was a 1:1:1 ratio night temperatures and rapidly changing daylengths
of Flanagan silt loam, peat moss, and perlite (vol./vol.). After may have caused this response. We have calculated the
seedlings emerged, all pots were transferred to one growth
chamber with a 3- by 1.5-m2 floor area. Seeds were overly
GDD requirements for FBI for each MG by using 8⬚C
sown, and seedlings were thinned to two plants per pot 7 d as the base temperature instead of 10⬚C. The conclusion
after emergence. One plant from each pot was used to deter- was similar.
mine FBI. The light source was a combination of fluorescent Figure 1 shows the overall relationship between day-
and incandescent lamps providing 450 mol m⫺2 s⫺1 photosyn- length, planting date, and duration of the two growth
thetically active radiation measured with a LI-6200 sensor for stages measured for MGs I, IV, and V. The figure shows
12 h. Pots of the two isolines were transferred to growth that the period from emergence to flowering (including
chambers under 14- and 16-h photoperiods, with 10 pots per floral bud growth) reduced dramatically when photope-
treatment, right after FBI was detected. Another 20 pots were riod reduced during the late growing season. This indi-
transferred to the longer photoperiods 8 d after FBI. The cated that photoperiod influenced or regulated the total
same number of pots was left in the 12-h treatment each time.
Only incandescent lamps of 25 to 30 mol m⫺2 s⫺1 photosynthet-
dates of growth of early vegetative growth and floral
ically active radiation were used to extend daylengths, min- bud growth. It had showed in all early and later MGs
imizing photosynthate contributing to biomass production though the degrees of the reduction were varied.
among photoperiod treatments. Temperatures were 25 and Effects of photoperiod on floral bud growth rates
20⬚C during the light and dark period, respectively, giving were shown best when plotted as a function of average
means of 22.5, 22.9, and 23.3⬚C for the 12-, 14-, and 16-h daylength during FBI to open flower (Fig. 2). Data in
treatments. Fig. 2 were the means of eight data points. From the
figure, we can clearly see that the slope increases when
Data Analysis the MG increases. There was a similar trend in both
All data were analyzed by one-way analysis of variance FBI and FBD. Because early MGs (IV and less) had
(ANOVA), and means were separated by Fisher’s Protected relatively short periods of FBI, the influence of photope-
LSD (SAS Inst., 1989). Only LSD is listed in the results tables. riod could not been detected clearly, which was indi-
Regression analysis was also performed between average pho- cated by the flatness of the slope of those regression
toperiod hours experienced and accumulative GDD required lines (one was even in a negative number). Under field
for FBI and floral bud development (FBD) in different MGs. conditions, the duration of bud initiation and develop-
ment was most affected by photoperiod in the strains
RESULTS AND DISCUSSION from MG IV to VIII, which were late MGs. Both deter-
Field Experiments minate and indeterminate strains showed the same pat-
tern of response, indicating that the stem termination
The average daylengths during floral bud initiation gene did not respond to photoperiod.
were not much different for the first three plantings in
early MGs (Table 3). The accumulative GDD required Table 4. Average growing degree days (GDD) required for floral
to produce a flower for early MGs (MG IV or less) bud growth and development (base temperature was 10ⴗC) for
were less than that required for the MG V near-isoline soybean Clark near-isolines differing in maturity for planting
(Table 4). However, the first planting required fewer dates used over 2 yr (1992 and 1993).
Planting
Table 2. Soybean planting and emergence dates for 1992 and Strain MG† 1 2 3 4 5
1993.
GDD
Planting L71-920 I 156 185 180 162 175
L63-3117 II 163 195 194 173 187
Year 1 2 3 4 5
L63-2404 III 185 212 193 181 184
1992 Planting date 13 May 1 June 15 June 1 July 21 July Clark-63 IV 189 211 192 198 177
Emergence date 22 May 8 June 21 June 9 July 28 July L74-441 IV⫹ 289 236 235 236 220
1993 Planting date 10 May 1 June 20 June 10 July 25 July L65-3366 V 294 224 228 232 212
Emergence date 28 May 10 June 24 June 15 July 30 July
† MG, maturity group.
946 AGRONOMY JOURNAL, VOL. 93, JULY–AUGUST 2001

Fig. 1. Effect of planting date on floral bud development in soybean strains differing in maturity. P1–P5 indicate planting groups. Vertical dash
lines indicate emergence dates. Horizontal dash lines indicate the time of floral bud growing period from emergence to open flowers on 50%
of plants in a field plot. Roman numerals represent maturity groups (MGs). The curved dot line shows photoperiod changes during the
growing season. Numbers in parentheses represent average photoperiod experienced during floral bud growth.

Fig. 2. Growing degree days (GDD) required from emergence (VE) to floral bud initiation (FBI) and floral bud development (FBD) for soybean
strains differing in maturity group (MG) and growing under different photoperiods encountered during the growing season at Urbana, IL.
 ZHANG ET AL.: EFFECTS OF PHOTOPERIOD ON SOYBEAN FLORAL BUD DEVELOPMENT 947

Table 5. Vegetative and phenological parameters at flowering for plants that were sown at five different dates. Data represent pooled
values from 1992 and 1993 growing seasons.
Planting
Parameters Strain MG† 1 2 3 4 5 LSD
Leaf no. L71-920 I 5.2 5.3 5.3 4.7 4.0 0.44
L63-3117 II 5.1 7.0 5.6 5.2 4.0 0.61
L63-2404 III 6.5 7.1 6.3 6.0 5.5 0.54
Clark-63 IV 6.7 7.9 7.4 6.6 5.1 0.77
L74-441 VI⫹ 13.4 11.8 9.3 9.2 7.6 0.79
L65-3366 V 14.3 13.7 10.6 9.1 7.9 0.68
Leaf area, cm2 L71-920 I 383 461 374 290 215 37
L63-3117 II 426 681 422 369 181 55
L63-2404 III 484 814 485 432 334 71
Clark-63 IV 670 834 687 534 285 79
L74-441 VI⫹ 2056 1329 822 768 527 201
L65-3366 V 2698 2838 1016 763 554 230
Plant height, cm L71-920 I 17.0 25.6 25.4 22.0 21.1 3.3
L63-3117 II 19.7 33.6 27.0 26.5 21.0 2.7
L63-2404 III 22.6 36.6 31.1 29.7 27.6 4.1
Clark-63 IV 27.4 39.3 40.3 35.5 27.7 4.4
L74-441 VI⫹ 72.0 63.7 54.0 46.1 33.5 11.2
L65-3366 V 75.2 70.7 55.9 46.4 34.9 10.8
Dry weight/plant, g L71-920 I 2.6 2.8 2.1 1.8 1.5 0.22
L63-3117 II 2.8 4.4 2.4 2.2 1.2 0.31
L63-2404 III 3.1 5.2 3.1 2.9 2.4 0.34
Clark-63 IV 4.1 5.5 4.6 3.9 2.0 0.42
L74-441 VI⫹ 15.9 10.6 6.1 5.6 3.6 0.89
L65-3366 V 21.9 21.9 8.2 5.5 4.3 1.77
† MG, maturity group.

Effects of planting dates on the nodal position of the soybean (L63-3117, Treatment 1), the prolonged photo-
first flower were also significant (Table 5). Late MGs period (16 h) had less effect on floral bud growth,
produced floral buds at the much lower position when perhaps due to its relatively short growth period. For
planted after July due to much shorter daylength-initi- late-maturity soybean (L65-3366, Treatment 2), 14 h
ated floral bud at an earlier stage. It also indicated that significantly delayed flowering, whereas 16 h inhibited
the floral bud could be initiated at any plant node if flowering. However, floral buds one-half to two-thirds
daylength was sufficiently short. Vegetative growth be- full size were observed in the 16-h treatment at the end
fore flowering was influenced by planting dates, FBI,
and flowering dates, especially in the late-maturity strains
as expected (Table 6). The greater the thermal require-
ment for flowering, the more the plant produced vegeta-
tive biomass. This sink for photosynthate and N may
have contributed some to slower bud growth rates.

Growth Chamber Experiment

This growth chamber study demonstrated the rela-
tionship of photoperiod and floral bud growth in a more
significant way by using artificially increased daylength
(Fig. 3). The long-day photoperiod delayed and inhib-
ited floral bud growth, and it largely depended on the
starting time of this treatment and how long the treat-
ment was applied and the MGs tested. In early MG

Table 6. The effects of planting dates on main-stem nodal position

of the first flower for Clark near-isolines differing in maturity.
Values are means from 1992 and 1993.
Planting
Strain MG† 1 2 3 4 5 LSD
No.
L71-920 I 2.0 2.0 2.7 2.9 2.0 0.35
L63-3117 II 2.0 2.0 2.3 2.7 1.9 0.32
L63-2404 III 3.5 3.5 3.2 4.0 2.4 0.44 Fig. 3. Summary of the effect of timing of the photoperiod treatment
Clark-63 IV 4.0 3.7 3.4 4.3 2.2 0.49 on floral bud growth in Clark near-isolines L63-3117 (MG-II) and
L74-441 VI⫹ 9.2 10.2 7.5 7.0 3.9 0.81
L65-3366 V 10 11.0 8.9 7.7 4.4 0.70
Clark L65-3366 (MG-V). All plants were under 12 h before treat-
ments started on the day floral buds were initiated, or for L65-
† MG, maturity group. 3366, 8 d after floral bud initiation (FBI).
948 AGRONOMY JOURNAL, VOL. 93, JULY–AUGUST 2001

of the experiment. When exposure to longer photoperi- provide valuable information for predicting growth stage
ods was delayed 8 d after FBI (Treatment 3), duration and maturity of soybean that can be used by crop manag-
of floral bud growth was hardly affected by 14- and 16-h d, ers. More studies are obviously needed before one can
but pod set was, as reported earlier (Acock and Acock, predict soybean phenology with confidence, especially
1995; Zhang et al., 1993). Zhang et al. (1995) reported for cultivars grown at the lower latitudes.
a temperature ⫻ photoperiod interaction on the GDD
requirement of floral bud growth in these same strains. REFERENCES
Duration of floral bud growth in the MG II near-isoline Acock, M.C., and B. Acock. 1995. Photoperiod sensitivity during
of Clark (L63-3117) was less affected by photoperiod soybean flower development. Biotronics 24:25–34.
than in the MG V near-isoline (L65-3366). McBlain et Board, J.E., and W. Hall. 1984. Premature flowering in soybean yield
al. (1987) showed a similar result in that daylengths of reductions at non-optimal planting dates as influenced by tempera-
ture and photoperiod. Agron. J. 76:700–704.
18 h or longer had a marked effect on time to flower Cregan, P.B., and E.E. Hartwig. 1984. Characterization of flowering
in MG II and earlier MGs, indicating that 16 h is not response to photoperiod in diverse soybean genotypes. Crop Sci.
long enough to induce a photoperiod response in MG II. 24:659–662.
Photoperiod effect on floral bud growth is a quantita- Egli, D.B., R.A. Wiralaga, T. Bustaman, Z.W. Yu, and D.M. Tekrony.
1989. Time of flower open and seed mass in soybean. Agron.
tive or accumulative effect. Floral bud growth and devel- J. 79:697–700.
opment stage is in early growth stage. To an early matur- Ephrath, J.E., and J.D. Hesketh. 1991. The thermal–photoperiod re-
ing soybean, this period was relatively short, and the quirement for floral bud growth. Biotronics 20:1–8.
accumulative effect of photoperiod influence was lim- Evans, L.T. (ed.) 1969. The induction of flowering: Some case histories.
ited and hard to be detected. Therefore, this effect may Melbourne Univ. Press, Melbourne, Australia.
Jones, P.G., and D.R. Liang. 1978. Simulation of the phenology of
be invisible or insignificant. Meanwhile, for a later ma- soybeans. Agric. Syst. 3:295–311.
turing soybean, even the FBD process can be long McBlain, B.A., J.D. Hesketh, and B.L. Bernard. 1987. Photoperiod
enough to be affected by photoperiod influence and and temperature effects on reproductive phenology in soybean
become significant. For the same reason, pod-setting isolines differing in maturity genes. Can. J. Plant Sci. 67:105–116.
SAS Institute. 1989 SAS/STAT software: Usage and reference. Ver-
period was at the late stage of growth. Even an early sion 6. 1st ed. SAS Inst., Cary, NC.
maturing soybean can accumulate significant photope- Sinclair, T.R., S. Kitani, K. Hinson, J. Bruniard, and T. Horie. 1991.
riod effect on pod-setting stage. It was indicated that Soybean flowering date: Linear and logistic models based on tem-
photoperiod effect on soybean from first flowering to perature and photoperiod. Crop Sci. 31:786–790.
maturity is a quantitative process that is controlled by Summerfield, R.J., H. Asumadu, R.H. Ellis, and A. Qi. 1998. Charac-
terization of the photoperiodic response of post-flowering develop-
three pairs of genes: E1/e1, E2/e2, and E3/e3 (Summerfield ment in maturity isolines of soybean [Glycine max (L.) Merrill]
et al., 1998). However, this is not the focus of this paper. ‘Clark’. Ann. Bot. (London) 82:765–771.
Quantifying the process of floral bud growth and de- Thomas, F.J., and C.D. Raper, Jr. 1976. Photoperiodic control of seed
velopment is a difficult but important step towards accu- filling for soybeans. Crop Sci. 16:667–672.
Wang, J.B., A. McBlain, J.D. Hesketh, J.T. Wooley, and R.L. Bernard.
rately predicting flowering and maturity of a soybean 1987. A data base for predicting soybean phenology. Biotronics
variety and MG. Predicting flowering and crop maturity 16:25–38.
using a computer model is very difficult because of com- Zhang, L.X., R.F. Wang, and J.D. Hesketh. 1993. Photoperiod effects
plex effects of daylength and temperature on processes on soybean floral bud growth. p. 149. In Agronomy abstracts. ASA,
Madison, WI.
involved. It requires systematic experiments to accu- Zhang, L.X., R.F. Wang, and J.D. Hesketh. 1995. Separating photope-
rately determine the relationship between daylength riod and temperature effects on the degree day requirement for
hours and MGs. The results reported in this paper can floral bud events in soybean. Biotronics 24:59–64.