Crop Science 42:242-247 (2002)
© 2002 Crop Science Society of America
FORAGE & GRAZING LANDS
Flowering in Crimson Clover as Affected by Planting Date
Twain J. Butler*,a,
Gerald W. Eversb,
Mark A. Husseyc and
Larry J. Ringerd
a Texas A&M Univ. Agricultural Research and Extension Center, 1229 N HWY 281, Stephenville, TX 76401
b Texas A&M Univ. Agricultural Research and Extension Center, P.O. Box 200, Overton, TX 75684-0200
c Soil and Crop Science Dep., Texas A&M University, College Station, TX 77843
d Dep. of Statistics, Texas A&M University, College Station, TX 77843
* Corresponding author (t-butler{at}tamu.edu)
 |
ABSTRACT
|
|---|
Understanding factors that affect flowering of crimson clover (Trifolium incarnatum L.) could improve management decisions to optimize utilization by improving season of use. The experiment was a split-plot randomized complete block design with three replications at College Station, TX, in the 1997-1998 and 1999-2000 growing seasons, and Overton, TX, in the 1998-1999 growing season. Main plot treatments of two crimson clover cultivars and subplot treatments of six planting dates (PDs) were used to evaluate the effect of date to reach 50% budding and 50% flowering based on day of year (DOY), days after planting (DAP), photothermal index (PTI), and growing degree days (GDD) under field conditions. Correlations with 50% bud and 50% flower were almost identical. ‘Columbus’ planted in the autumn flowered an average of 49 d later than ‘Tibbee’. Date to reach 50% flowering was best correlated with DOY (r = 0.93 and 0.97) and DAP (r = 0.92 and 0.98) for Columbus and Tibbee. Date to reach flowering was not as highly correlated with PTI (r = 0.66 and 0.82) or GDD (r = 0.71 and 0.85) for Columbus and Tibbee, thus temperature could not be used to predict flowering. Planting after 21 December delayed flowering in Tibbee 2 to 9 wks, whereas, Columbus planted after 21 December did not flower. It is important to plant early in the growing season or to use later-maturing cultivars to maximize the length of the growing season and possible total production in grazed environments.
Abbreviations: DAP, days after planting • DL, daylength • DOGS, day of growing season • DOY, day of year • GDD, growing degree days • PD, planting date • PLS, pure live seed • PTI, photothermal index
|
INTRODUCTION
|
|---|
CRIMSON CLOVER is one of the most important annual clovers for overseeding into perennial warm-season grass sods becuase of its excellent seedling vigor, early forage production, and early maturity. Studies reporting prediction equations for growth and development of cool-season annual clovers have not been reported. It would be beneficial to accurately predict crop growth in order to optimize management, production, and utilization. Predicting date of flowering could be a valuable management tool to optimize production by estimating length of growing season and improving season of use.
Some plants are not capable of flowering upon germination, but must pass through a juvenile period lasting from a few days to many years (Dennis, 1984). Some species must reach a certain growth stage, photoperiod, temperture, accumulate sufficient photosynthate, or a combination of these factors before flowering. Juvenility is most often seen in the field as delayed flowering in later-sown crops. Therefore, early planting is essential for growth in annual species that must pass through juvenility (Fisher, 1999). There is little information regarding the effects of juvenility on flowering of cool-season annual clovers.
Most plants flower in response to certain vernalization and photoperiod requirements. Vernalization is a cold requirement, which must be met in order to induce flowering for some species. It is a cumulative process and is achieved at a faster rate at lower temperatures. Vernalization is also reversible by induction of high temperatures when a plant is at an early stage of development. Crimson clover can develop as a summer annual without an obligatory cold requirement for flowering, thus it has no vernalization requirement (Knight and Hollowell, 1958).
Photoperiodism (response to light duration, light quality, and radiant energy) is a mechanism that enables plants to respond to daylength (DL) so that they flower at a specific time of the year as determined by the length of the day. Flowering of crimson clover begins when the DL exceeds 12 h (Knight, 1985). As photoperiod increased (14 to 18 hrs) under greenhouse conditions, length of time to flowering decreased (Knight and Hollowell, 1958).
Donnelly and Cope (1961) reported that there was an 8-d spread in dates (+8) in northern Alabama to reach full bloom among six crimson clover cultivars [‘Auburn’ (0), ‘Autauga’ (0), ‘Dixie’ (0), common (+4), ‘Talladega’ (+6), and ‘Chief’ (+8)]. Evers et al. (1995) reported a 35-d spread in dates (+35) in northeast Texas to reach 50% flowering among eight cultivars of crimson clover [‘AU Robin’ (0), Auburn (+5), ‘Flame’ (+5), Tibbee (+7), Dixie (+9), Chief (+12), ‘American’ (+23), and Columbus (+35)]. Only three cultivars were common to both studies, but differences among the three were observed.
Temperature interacts with photoperiod to induce flowering. Exposure to low temperatures during the winter often causes earlier flower induction, but actual flower development is retarded by low temperatures (Spedding and Diekmahns, 1972). Aitken (1974) reported that lower temperature retards flowering in subterranean clover, but increases leaf yield per plant.
In crimson clover, photoperiodism is also conditioned by temperature. Von Gliemeroth (1943) reported that under greenhouse conditions, crimson clover germinating at low temperatures accelerated plant development, which led to earlier flowering and maturity. Knight and Hollowell (1958) reported that crimson clover flowered earlier when plants were shifted from lower temperature to higher temperatures at both natural (11–14 h) and artificial (16 h) photoperiods. However, at 16-h photoperiods both high (18–21 °C) and low (0 °C) night temperatures delayed flowering, compared with 10 to 13 °C temperature treatment when plants were grown from germination to maturity. Under natural photoperiod (11–14 h), high night temperature (18–21 °C) inhibited flowering.
There is a need to determine the relationship between crimson clover growth and environmental factors in order to better plan management decisions. If flowering could be predicted, management decisions such as PD, harvest dates, and stocking rates could be modified to maximize utilization. Nor is it known if intermediate- and late-maturing cultivars of crimson clover would respond the same way to environmental factors. Our objective was to determine if DOY, DAP, PTI, and GDD could be used to predict flowering of intermediate- and late-maturing crimson clover cultivars by using several PDs across 3 yr.
|
MATERIALS AND METHODS
|
|---|
A randomized complete block design experiment with three replications was initiated in the autumn of 1997 and 1999 at College Station, TX (30°38' N, 96°26' W), and with four replications in the autumn of 1998 at Overton, TX (32°17' N, 94°58' W). The College Station study was in a Boonville fine sandy loam soil (fine, smectic, thermic Vertic Albaqualfs), and the Overton study was planted in a Bowie fine sandy loam (fine-loamy, siliceous, semiactive, thermic Plinthic Paleudults).
At College Station, both Tibbee and Columbus crimson clover cultivars were planted in 3 rows spaced 30.5 cm apart and 3.0 m in length. Clovers were planted at 16.8 kg pure live seed (PLS) ha-1 with a hand seeder (Planet Jr., Powell Manufacturing Co., Bennetsville, SC) at approximately monthly intervals from October through March. At Overton, the same cultivars were planted into a prepared seedbed at 18 kg ha-1 PLS with a small plot drill. Plots consisted of seven rows 4.6 m long spaced 18 cm apart. Both sites were fertilized according to soil test recommendations prior to planting. At Overton, 67 kg ha-1 P2O5, 112 kg ha-1 K2O, and 2800 kg ha-1 lime were applied, and no fertilizer was needed at College Station. At both College Station and Overton, the DL at 1 September was 
12.5 h, which decreased to 10.2 h on 21 December and then increased to 14.1 h by 31 May. Clover planted in the autumn was exposed to a decreasing photoperiod followed by an increasing photoperiod in the spring. Clover planted after 21 December was only exposed to increasing photoperiods.
Maximum and minimum daily temperatures were collected from the Easterwood airport in College Station (2 km from plot area) and with a model 101 temperature probe (Cambell Scientific, Logan, UT) in Overton (adjacent to plot area). Growing degree days were calculated by averaging daily minimum and maximum temperature minus a base temperature of 0 °C. Fukai and Silsbury (1976) reported that at 30 °C, leaf death proceeded more rapidly than leaf appearance of subterranean clover (T. subterraneum L.). In these studies, when maximum temperature exceeded 30 °C, a threshold model that used an adjusted maximum temperature similar to the one used by Dufault (1997) was used to account for the reduced growth associated with high temperatures:
where Tmax = maximum daily temperature, Tmin = minimum daily temperature, Tbase = base temperature, Tu = upper threshold limit, and Tadj-max = adjusted maximum daily temperature. A PTI was calculated by summing the proportion of DL per 24-h period multiplied by the total heat units (Masle et al., 1989).
where DLi = number of daylight hours in each day.
Plots were monitored every 2 d to visually estimate when plots reached 50% bud and 50% flower stages each year. The relationship between PD, based upon day of the growing season (DOGS) starting from 1 September, DOY starting from 1 January, DAP, PTI, and GDD to reach the 50% bud and 50% flower stages were determined by multiple regression and analysis of variance procedures (Steel and Torrie, 1980; SAS Inst., 1991). All observations were used to develop regression equations, but only the means were plotted in the figures. Means were separated using a Fisher's protected LSD, and the slopes between cultivars and PDs were compared within each growing season using a partial F-test for pairwise comparisons at the 0.05 significance level (Neter et al., 1996).
|
RESULTS AND DISCUSSION
|
|---|
Average daily temperatures in autumn and spring were similar to the 30-yr average in College Station, TX (Fig. 1)
. However, temperatures in January and February were 7 °C warmer than normal with the temperature never falling below 0 °C. These higher winter temperatures caused the accumulation of GDD to occur at a more rapid rate than the 30 yr average. A similar pattern occurred at Overton during 1998-1999 growing season. Average daily temperatures were about 5 °C above the 30 yr average for January and February, and similar for the remaining months. However, there was little difference in accumulation of GDD between the 1998-1999 growing season and the 30-yr average at Overton.
View larger version (46K):
[in this window]
[in a new window]
|
Fig. 1. Monthly average daily temperature and accumulated growing degree days (GDD) by year and long-term average at College Station and Overton, TX.
|
|
Planting date influenced DOY, DAP, GDD, and PTI to reach the 50% bud stage (Table 1). The transition from vegetative to reproductive stage did not occur for Columbus when planted in the spring. It appears the required juvenility stage for flowering reported by Fisher (1999) was not adequate when the appropriated DL for Columbus was reached. When planted in the fall, Columbus reached the 50% bud stage 6 to 8 wk later than Tibbee at College Station and 5 wk later at Overton. As PD was delayed, the DOY to reach the 50% bud stage was later for both intermediate-maturing Tibbee and late-maturing Columbus crimson clover (Table 2). Day of year to reach the 50% bud stage was correlated to DOGS for Columbus and Tibbee (r = 0.91 and 0.96), and it increased curvilinearity with time (Fig. 2)
. Day of year for both cultivars to reach the 50% flower followed the same trend, and was generally 2 wk later than the 50% bud stage. Exceptions were the September PDs for Tibbee and the October PD in 1997-1998 for Columbus, which were 3 wk later (Table 3). Day of year was correlated with DOGS to reach the 50% flower stage for Columbus and Tibbee (r = 0.93 and 0.97) and increased curvilinearity with time. Graphs of regression equations for 50% flower are not presented since they were almost identical to the 50% bud stage.
View this table:
[in this window]
[in a new window]
|
Table 1. Influence of planting date for Tibbee and Columbus crimson clovers to reach 50% bud stage based on day of year (DOY), days after planting (DAP), growing degree days (GDD), and photothermal index (PTI).
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Regression equations for ‘Columbus’ and ‘Tibbee’ crimson clovers to reach the 50% bud stage based on day of year (DOY), days after planting (DAP), growing degree days (GDD), and photothermal index (PTI) versus day of the growing season beginning on 1 September.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2. Relationship between day of the growing season (DOGS) starting 1 September and day of year (DOY), days after planting (DAP), growing degree days (GDD), and photothermal index (PTI) for ‘Columbus’ and ‘Tibbee’ crimson clover to reach the 50% bud stage at Overton (O) and College Station (CS).
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Influence of planting date for ‘Tibbee’ and ‘Columbus’ cultivars of crimson clovers to reach the 50% flower stage based on day of year (DOY), days after planting (DAP), growing degree days (GDD), and photothermal index (PTI).
|
|
As PD was delayed, DAP to the 50% bud stage decreased, which resulted in progressively shorter growing seasons (Table 1). Days after planting was highly correlated to DOGS, at which 50% bud stage occurred for Columbus and Tibbee (r = 0.92 and 0.98). However, DAP decreased in a curvilinear manner more rapidly for Tibbee than Columbus (Fig. 2). This data is in agreement with Knight and Hollowell (1958) that as photoperiod increased, time to flowering decreased.
Days after planting to reach the 50% flower stage followed the same trend as the 50% bud stage. Days after planting were highly correlated with DOY to reach the 50% flower stage for both Columbus and Tibbee (r = 0.92 and 0.98). This suggests that management practices typically applied to earlier-maturing cultivars of crimson clover may not apply to Columbus. Intermediate-maturing Tibbee required a shorter DL to reach the 50% bud and flower stages, and therefore was better able to compensate for later PDs than Columbus.
The use of GDD demonstrates the influence of temperature, or heat units, on flowering. Columbus required about 800 to 1000 more GDDs to reach the 50% bud stage than Tibbee when planted at College Station (Table 1). At the more northern location in Overton, Columbus only required an additional 500 to 600 GDDs than Tibbee. There was a tendency for the number of GDDs to reach the 50% bud stage to decrease as PD was delayed, especially at Overton. Pooling years, there was a significant quadratic response between DOGS and GDD for both cultivars (Table 2). However, the relationship was not as accurate as DOY or DAP, which supports the report by Knight (1985) that flowering of crimson clover is primarily controlled by DL. Growing degree days only accounted for 50 and 77% of the variability in Columbus and Tibbee (Table 2). However, Columbus planted in February and March did not reach the 50% bud stage due to temperatures exceeding 30 °C, which illustrates that temperature could also influence development.
Quantity of GDD to reach 50% flowering were generally from 200 to 300 higher than 50% bud for Tibbee and from 300 to 400 higher for Columbus (Table 3). The range in GDD among PD within years was about 400 GDD. The number of GDD required to reach the 50% flower stage was not highly correlated for Columbus or Tibbee (r = 0.71 and 0.85). Typically, the number of GDD required to reach the 50% flower stage for Tibbee decreased for the autumn PDs (October to December) and then increased for the spring PDs (February to March). When planted in February and March, Columbus did not flower, so comparable data is not available. The range in GDD among PD within years was similar for both 50% bud and flower stages.
The PTI is a climatic parameter that combines the influence of heat units and DL. Tibbee required 400 to 500 less PTI units than Columbus to reach the 50% bud stage (Table 1). Although the regression equations were significant and followed the same trend as GDD, the coefficients of determination for PTI (R2 = 0.43 and 0.69) were lower than DOY, DAP, and GDD (Table 2). Because average daily temperature is part of the PTI, these values followed the same trends as GDD to reach the 50% flower stage (Table 3). The PTI did not accurately predict flower stage for Columbus or Tibbee (R2 = 0.43 and 0.67) (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 4. Regression equations for ‘Columbus’ and ‘Tibbee’ crimson clovers to reach the 50% flower stage based on day of year (DOY), days after planting (DAP), growing degree days (GDD), and photothermal index (PTI) vs. the day of the growing season beginning on 1 September.
|
|
In this study, flowering appeared to be controlled primarily by DL; however, temperature played a limited role. Tibbee crimson clover planted at the recommended dates for Texas (mid-September to early November) flowered from 23 March to 3 April, whereas the later PDs flowered progressively later (Table 3). Tibbee planted in February and March did not flower until 90 to 111 DAP (Table 3). The late-maturing Columbus crimson clover flowered from 9 May to 14 May when seeded at the recommended PD, but did not flower when planted in February and March. Knight and Hollowell (1958) reported that Dixie, Chief, and common crimson clover would not flower when night temperatures reached 18 to 21 °C. This suggests that crimson clover must either pass through a juvenility phase, or it requires low night temperatures (vernalization) or a critical DL before it can flower. It is probably a combination of factors, because Tibbee was able to flower under these high night temperatures in the spring.
|
CONCLUSION
|
|---|
Different PDs were used to quantify the effects of temperature and DL on reproductive development of an intermediate- and late-maturing cultivar of crimson clover. Generally, as PD was delayed, the rate of development was accelerated. This corresponds to a shorter growing season, which could equate to reduced production. From a practical viewpoint, it is important to plant early in the growing season (as soon as temperature and moisture allows) to maximize the length of the growing season and possibly total production in grazed environments. Another way to extend the growing season is to use later-maturing cultivars, which will remain vegetative for a longer period.
Day of year and DAP were closely correlated for Columbus and Tibbee to reach the 50% bud and flower stages, which suggest that flowering is primarily controlled by photoperiod. Conversely, the number of GDD and PTI required to reach flowering was not as highly correlated, further suggesting that DL is primarily responsible for flowering in crimson clover. High temperatures inhibited flowering, which illustrates that temperature plays an important role in flower development. However, temperature as GDD or PTI did not improve the accuracy of predicting the occurrence of flowering over DOY or DAP alone.
The growth stage is important to determine when clover should be harvested for hay or when grazing should be terminated to allow for reseeding. Sometimes weather does not permit planting at the optimal time. Predicting when flowering occurs based upon PD is useful in making management decisions concerning utilization and reseeding.
Received for publication March 14, 2001.
|
REFERENCES
|
|---|
- Aitken, Y. 1974. Plant development and shoot structure. p. 20–40. In Flowering time, climate and genotype. Melbourne University Press, Carlton South, VIC, Australia.
- Dennis, F.G. 1984. Flowering. p. 237–264. In M.B. Tesar (ed.) Physiological basis of crop growth and development. ASA and CSSA, Madison, WI.
- Donnelly, E.D., and J.T. Cope. 1961. Crimson clover in Alabama. Bul. 335. Auburn Univ. Agric. Exp. Stn., Auburn, AL.
- Dufault, R.J. 1997. Determining heat unit requirements for broccoli in coastal South Carolina. J. Am. Soc. Hortic. Sci. 122:169–174.
- Evers, G.W., P.T. Smith, S.D. Blackerby, and D.S. Doctorian. 1995. Comparison of crimson clover varieties in northeast Texas. p. 1–5. Forage Research in Texas. No. CPR-5252. Texas Agric. Exp. Stn., College Station, TX.
- Fisher, M.J. 1999. Crop growth and development: flowering physiology. p. 81–90. In D.S. Loch and J.E. Ferguson (ed.) Forage seed production. Vol. 2:81–92. CAB International, Cali, Columbia.
- Fukai, S., and J.H. Silsbury. 1976. Responses of subterranean clover communities to temperature. I. Dry matter production and plant morphogenesis. Aust. J. Plant Physiol. 3:527–543.
- Knight, W.E. 1985. Crimson clover. p. 491–502. In N.L. Taylor (ed.) Clover science and technology. Agron. Monogr. 25. ASA, CSSA, and SSSA, Madison, WI.
- Knight, W.E., and E.A. Hollowell. 1958. The influence of temperature and photoperiod on growth of crimson clover (Trifolium incarnatum L.) Agron. J. 50:295–298.
- Masle, J., G. Doussinault, G.D. Farquhar, and B. Sun. 1989. Foliar stage in wheat correlates better to photothermal time than thermal time. Plant Cell Environ. 12:235–247.
- Neter, J., M.H. Kutner, C.J. Nachtsheim, and W. Wasserman. 1996. Multiple regression – II. p. 260–326. In Applied linear statistical models. 4th ed. McGraw-Hill, Chicago, IL.
- SAS Institute. 1991 SAS users guide. Release 6.12. SAS Inst., Cary, NC.
- Spedding, C.R.W., and E.C. Diekmahns. 1972. Reproduction. p. 300–305. Grasses and legumes in British agriculture. Commonwealth Agriculture Bureaux, Farnham Royal, Bucks, England.
- Steel. R.G.D., and J.H. Torrie. 1980. Principles and procedures of statistics: A biometrical approach. 2nd ed. McGraw-Hill, New York.
- Von Gliemeroth, G. 1943. The influence of germination-temperature and day length on development of crimson clover. J. Landwirtsch. 39:123–150.
This article has been cited by other articles:
|
|
|
|
 
G. W. Evers and G. R. Smith
Crimson Clover Seed Production and Volunteer Reseeding at Various Grazing Termination Dates
Agron. J.,
October 3, 2006;
98(6):
1410 - 1415.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
