Temperature and the Integrated Control of Plant Height

* Unfortunately, figures provided for this presentation contained a combination of graphic images and text which couldn't be translated into html. Consequently, figures are missing some labels.

Growing plants in greenhouses is first and foremost a protection against the elements. But once the basic requirements of life are assured (temperature, light and water) we can start to manipulate the environment in ways which will enhance the quality of our crops and provide the kind of growth which will produce the desired end product. In this way the environment in a greenhouse (whether we are growing forest seedlings, vegetable transplants or flowering plants) is as much a tool in the nursery business as the tractor, sprayer or pruning shears.

It was not always this way. For many years the greenhouse was considered merely a shelter; somewhere to grow crops at a time of year when outdoor growth was slow or impossible. Heat was controlled by simple thermostats which perhaps allowed us to change temperatures on a day/night basis, but little else. So what changed? If I had to choose a single word it would be "electronics" The introduction of microprocessors opened up a new world of greenhouse climate control, much as computers have revolutionized all other aspects of our society. As computers started to make their way into production greenhouses in the early 1980's, horticulturists suddenly realized that their ability to control the environment around their crops was quickly outstripping their knowledge of just what environmental conditions to apply. Traditional views of crop temperature requirements, for example, called for a day temperature somewhat higher than night temperature. Nobody, it seemed was quite sure why, but it simply seemed logical that plants should be warmer during the active daily growth period than at night.

In 1986 a group of researchers at Michigan State University began an in depth study of temperature effects on a range of floricultural crops. In a series of experiments which had local growers scratching their heads at the time, they grew Easter lilies, chrysanthemums and poinsettias under a wide range of temperatures conditions. What was so hard for the growers to accept at the time was not so much that some of the temperatures were out of the range of those commonly used to grow their flower crops, but that these researchers had the strange idea that it might be acceptable to have nighttime temperatures warmer than those during the day. What the Michigan group discovered has had a major impact on the greenhouse floriculture industry and may have far-reaching implications for the forest nursery industry.

We have learnt over the past several years that temperature has a complex effect on plant growth and development and further that the way in which temperature is controlled (over successive diurnal cycles) has a profound effect on processes such as flowering, leaf development and stem elongation. In some cases, the relationship between temperature and process is quite straightforward. Take for example a measure of vegetative development in Easter Lily - leaf unfolding (Figure 1). No matter how temperature conditions are set, day or night, the rate of leaf unfolding is strictly dependent on the average daily temperature (Karlsson et al, 1988). Other species have shown similar responses so quite simply if we want to speed up a crop of Easter lilies we can turn up our thermostats or set the computer to control the greenhouse at a higher set point.

Doing so, however, may not quite give the results that we are looking for. While working with chrysanthemum in a series of experiments looking at growth and development in relation to a range of day/night temperatures, Meriam Karlsson at Michigan State discovered that plant height was much more strongly dependent on the difference between day and night temperature than on the absolute value of either (Karlsson et al, 1989). To explain the effect she coined the term DIF.

The horticultural industry quickly seized on the idea that plant height might be controlled simply by changing the way in which temperature conditions were cycled each day. More research showed that many species apparently responded in the same way and it was not long before the concept was being applied in greenhouses across North America and Europe.

Simply put, the rate of stem elongation depends on the relationship between day and night temperature. As day temperature increases relative to night temperature, SER increases and vice versa. DIF is commonly expressed in terms of day temperature minus night temperature and influences cell expansion in the internodal regions of the stem. In this data for chrysanthemum cultivar BGA, internode length at full flower increases from 1.09 at -12 DIF to 2.99 at +12 DIF under exactly the same daily average temperature (Table 1). Numerous experiments in our growth facilities and others have confirmed this response.

Figure 1: Leaf unfolding of Easter Lily in relation to average temperature (From Karlsson et al, 1988).

Table 1: The Effect of DIF on internode length of in "Bright Golden Anne" chrysanthemum grown at 18C average daily temperature. (From: Karlsson et al, 1989)
DIF (Day temp - Night temp)	Internode Length (mm)
-12	10.9
-8	12.8
-4	15.3
0	18.2
4	21.6
8	25.5
12	29.9

What is so attractive about the concept of DIF? Before the nature of this temperature response was known horticulturists had very little ability to control the height of their crop. We now know that the temperature regimes which were used for so many years exacerbated problems of height control. The tendency of plants to stretch (particularly under the lower light conditions of early Spring) was counteracted with growth regulators such as B-Nine, cycocel and more recently paclobutrazol and the new generation of triazole growth regulators such as Uniconazole (Sumagic). The trend towards reduced chemical use and increased reliance on integrated methods to achieve cultural goals favours strategies based on environmental control. Another reason that DIF is of very great potential usefulness is that it is virtually impossible to "over do it" when using temperature to control plant height. This contrasts with the potential damage which can ensue from over use or overdosing of plant growth regulators (in many cases the effects of potent growth regulators such as uniconazole take weeks if not months to wear off and plants can be stunted even after sale and transplant. This is particularly evident in some bedding plant species treated with uniconazole.

Problems with DIF

For all the positive, practical implications of the DIF concept, there are some drawbacks and perhaps the greatest single one relates, ironically, to the practicality of using the technique. To all intents and purposes in most commercial greenhouses whether the crop is forest tree seedlings or poinsettias we are working under positive, and sometimes extremely positive DIF conditions. Late spring and summer temperatures may exceed 30C unless we have very efficient cooling control, yet night temperatures drop naturally to the range of 12 to 15C (and it is simply not practical to turn the heat on just to try and maintain negative DIF conditions). Further, we have shown in many plants that maintaining night temperatures warmer than day temperatures leads to foliar chlorosis, a phenomenon which may be related to an altered carbohydrate metabolism. From a plant science stand-point there is also the question of why DIF works at all! I think we can all relate to the idea that temperature affects plant growth in a very direct way, but why should what is after all a rather abstract concept (a difference between temperature applied during the night or during the day) affect stem elongation rate?

Added to all this it became clear after a few years that DIF did not always work in all situations and the reasons for this apparent breakdown in the pattern were not clear. These questions and concerns started us in the late 1980's on a series of experiments designed to investigate the exact nature of the DIF response and how we might improve on the basic concept.

Until that time all the information on DIF had been gathered from experiments involving daily or weekly measurements of stem length with rulers and calipers. We knew that stem elongation responded in a definable way to day and night temperature, yet we had no idea of how that growth was distributed over the day and night periods. It seemed that to proceed beyond our basic understanding of the stem growth phenomenon we would need to take a much closer view of elongation over relatively short time periods. To achieve this we needed to resolve stem growth to micrometer increments recorded over time intervals of no more than 5 minutes. By connecting the uppermost node of the plant to a device known as a linear voltage displacement transducer (LVDT) and linking the LVDT to a computer we were able to monitor stem position over very short time intervals and integrate the data stream to produce an elongation rate. In the initial experiments we chose chrysanthemum, but have found that the system will work well with most plants which have a vertical growth habit.

The first thing that we noticed was that growth rate was by no means constant throughout the diurnal period (Figure 2). Under constant temperature conditions or when night temperature exceeds day temperature, most growth occurs at night. However, when day temperature exceeds night temperature stem elongation during the day and night periods are approximately equal.

In all cases there is a distinct rhythmicity to elongation and within those patterns several very reproducible features. We saw, for example that under constant temperature conditions and whenever day temperature exceeded night temperature there was a distinct peak in growth coincident with the transition from night to day. Thereafter growth generally declined through the day. When night temperature exceeded day temperature there was no immediate increase in growth rate at the night/day transition. Instead growth rate dropped and then began a slow recovery followed by another drop. Lets come back to these details in a minute and deal with some of the practical implications of these observations.

The very distinct rhythmicity of growth response prompted us to ask some questions about whether these growth patterns would proceed independently of all environmental cues. Did, for example the light/dark transitions and temperature treatments serve simply to modify a rhythm of stem elongation which was inherent in the plants underlying physiology and how flexible would that physiology be. We found that under certain circumstances rhythms did indeed proceed under constant conditions (Figure 3). The SER rhythm was clearly cued very strongly by the light dark transition, but indeed the rhythm of SER actually continued in a weak, but distinguishable way for several cycles when plants were subject to continuous lighting. Not so in darkness,(Fig. 4) however, suggesting an absolute requirement for some light signal to maintain the rhythm or perhaps the fact that stem elongation activity is lost quickly when photosynthesis and a constant supply of carbohydrate is no longer available.

How could this new-found knowledge of both the endogenous nature of the SER rhythm and the influence of various environmental conditions be used to give us better control over stem development? We started to examine more closely the patterns which by now had become very familiar to us. Quite clearly stem elongation was influenced through the application of DIF - however negative DIF of the kind used in these experiments is clearly not an option for commercial growers. We have already dealt with the practical obstacles to the imposition of any negative DIF so we needed to look for other ways to achieve similar results.

Stem elongation of chrysanthemum "Envy" subjected to continuous lighting and constant 18.3C following 2 consecutive day/night cycles.

Stem elongation of chrysanthemum "Envy" subjected to continuous darkness and constant 18.3C following 2 consecutive day/night cycles.

Despite the fact that in general there was more growth occurring at night than during the day, our attention was drawn to one particular characteristic of the daytime SER pattern. The peak which occurred at the start of the day period under constant, and positive DIF conditions accounted for up to 20% of the total daytime growth. We reasoned that elimination of that burst of growth at the night/day transition (or dawn in the natural world) could have a significant effect on the total daily amount of stem elongation. Since the growth peak seemed to be enhanced at higher daytime temperatures (and reduced when day temperature dropped) we decided to look at the effects of dropping temperature for various periods at the start of the day.

Choosing our standard 18.3C regime we dropped the air temperature around the plants to 8.3C for either 2,4 or 6 hours beginning immediately at the lights on event (Figure 5). The results were startling. SER dropped rather than increased in the early part of the day but the decline in growth rate did not last until the end of the pulse. It seemed rather that the low temperature shock induced a timing mechanism which temporarily suppressed growth. If cool conditions persisted, the timer was reset so that the growth suppressing effect continued.

Overall the effect of the low temperature pulse was to significantly reduce daily growth. The effects of the low temperature pulses were, however, transitory and had no effect on the overall periodicity of the rhythm indicating once again the very strong relationship between environmental cues and the underlying endogenous rhythm. The cues of light and darkness are of paramount importance. 4 and 6 hour pulses, as we might expect, tended to be more effective than those given for shorter time periods but 2 hours at lower temperature at the start of the day was quite effective in reducing daily stem elongation.

Low temperature pulses given later in the day also reduced stem elongation, but because they did not coincide with the early growth peak they were less effective in limiting stem elongation. This raises the question of when and how temperature drops should be timed to effectively control stem growth.

While temperature pulses or the imposition of day-long low temperature (i.e. negative DIF conditions) are less effective if the low temperature does not start at the beginning of the day period, some useful work in Scandinavia has established that temperature can be dropped before dawn (or a lights on event) and still retain the effectiveness of the negative DIF treatment (Table 2)

The effect of temperature pulses of 2, 4 or 6h duration at 8.3C imposed on chrysanthemum plants grown at 18.3C over successive 11h day/ 13h night cycles.

Table 2: Average internode length (mm) of "Garland" and "Surf" chrysanthemum subjected to positive DIF (21/19C) or negative DIF (14/22C) with temperature shifted at daybreak or before daybreak. (From Jacobsen and Amsen, 1992).

So far we have dealt mainly with the responses of chrysanthemum to DIF and variations on the DIF technique. How general is the DIF response and how can it be extended to other species? Work which we have undertaken over the past 18 months sheds some more light on the nature of the DIF and related responses and may provide some insights which will allow your own assessment of the usefulness of the technique in your nurseries. We have recently undertaken a survey of various flowering plants to determine how general the diurnal SER patterns are and to assess whether DIF operates in a similar way in different plants. DIF did indeed appear to be quite effective in influencing stem elongation in several species, we have been surprised at the diversity of response and the fact that DIF will affect growth quite differently in different plants.

To illustrate we can consider two bedding plant species which tend to cause greenhouse growers some problems with height control under late winter/spring conditions - zinnia and snapdragon. In both cases DIF has a measurable effect, but the effects are not the classical ones which we found with chrysanthemum (Figures 6 and 7).

Growth in both species follows a sigmoid pattern and - the three lines represent growth in the 3 DIF regimes. Here, in zinnia there is no significant difference between growth at 0 and -5 DIF.

In snapdragon the growth curves follow a similar format, but in this case there is no difference between 0 and +5 DIF. The high resolution analysis bears this out in terms of total daily stem elongation. The fact that growth is affected differently by DIF at different stages of development is yet another complication which needs to be taken into consideration in assessing plant performance in relation to DIF conditions.

For producers there are some very practical implications of these findings. Lets take the zinnia and snapdragon examples. In zinnia there is really no advantage in attempting to drop nighttime temperatures to attain negative DIF conditions - stem elongation will be similar to equal day/night temperature conditions. On the other hand considerable advantage would accrue from a negative DIF situation in snaps. To make matters even harder for us the "limited DIF" low temperature pulse treatments will have to be adjusted in these two species since the early day peak in growth is only present in zinnias. In snapdragons an equivalent pulse is evident at the start of the dark period!

With so much variation in response is DIF still a valuable technique for plant producers? There is no question that it is. Variable response is something which must be expected, however, and as we have shown certainly not all species will respond in the same way. It is encouraging for the general application of DIF techniques in a wide range of greenhouse crops that temperature is a factor of very great importance. While stem elongation patterns appear to be strongly entrained by light/dark transitions, it is temperature which can modify the basic rhythms. While light conditions may affect overall growth and vigour, under most circumstances, the DIF response remains intact.

Figure 6:
Life cycle growth curves for zinnia subjected to O, +5 and -5C DIF regimes.

Figure 7:
Life cycle growth curves for snapdragon subjected to 0, +5 and -5C DIF regimes

In recent months we have conducted several experiments under controlled conditions in greenhouses under natural lighting. The early day growth peaks which may play such an important part in the overall daily stem elongation response in several species seem to be initiated by early morning (pre-dawn) lighting when the spectral balance is shifted to the far-red end of the spectrum (Figure 8). We are now investigating the possibility that temperature interacts with this spectral shift in influencing the elongation response.

The control of temperature and more specifically the DIF technique is clearly a powerful production tool but is it sufficient to produce the carefully graded plants which are the mainstay of the horticulture and forestry industries? In some cases environmental manipulation may be all that is necessary, but increasingly we are looking at DIF as a component in an integrated approach to plant growth control. Chemical growth regulators are used extensively in horticulture and have for many years been the method of choice for controlling excessive stem elongation. When these chemicals are used in combination with DIF, we have found that although the growth regulator effect predominates, DIF still acts and in a manner which appears to be quite independent of the effect of the growth regulator. Chemical growth regulators primarily affect the synthesis of gibberellic acid in the plant. When that synthesis is blocked stem elongation growth is suppressed. It appears from our most recent work that DIF does not affect gibberellin synthesis, although it may very well affect plant sensitivity to gibberellin. In practical terms this means that DIF can be

used in combination with growth regulators in an integrated way to limit stem elongation. This is especially relevant when we consider the mode of action of the new generation of chemical growth regulators (the triazoles) now under trial and consideration for registration. This is because plants are dose-sensitive to these chemicals and the quantity of active ingredient applied to the crop can be varied to provide a graded control of elongation. If DIF is used as the fundamental elongation control strategy, then growth regulators can be used to "top-up" the response as required.

Above all stem elongation control and the ultimate size and stature of plants is not something which, as producers, we can assure with a single treatment (whether that is an application of a chemical growth regulator or an adjustment of temperature). Rather it is an ongoing process of assessment and action which requires careful record keeping, but record keeping of a very specialized kind. In the greenhouse industry the techniques has become known as graphical tracking. To apply it we simply need to monitor crop growth at intervals and determine the definable, low resolution pattern of stem elongation during the production cycle (Figure 9). With this information we can construct a graph to represent the long-term pattern of elongation, and modify it to represent the desired response. During crop production representative plants can be monitored and DIF and/or growth regulators used to give the desired growth trace.

Figure 8:
Daily stem elongation rate for zinnia grown at 18C constant daily temperature in a greenhouse under natural, March lighting conditions. A: Stem elongation rate. B: Incident photosynthetically active radiation

Forestry seedlings which undergo almost linear growth will clearly exhibit very different growth patterns to those shown by Easter lilies depicted in Figure 9. However, the ultimate benefit of this analysis is putting control firmly in the hands of the grower and it is the first and most important step in the application of DIF and other growth control techniques. It is through this type of analysis that the effectiveness and applicability of DIF to individual crops can best be determined.

Jacobsen, L.H. and M.G. Amsen. 1992. The effect of temperature and light quality on stem elongation of chrysanthemum. Acta Hort. 305:45-50

Karlsson, M.G., R.D. Heins and J.E. Erwin. 1988. Quantifying temperature-controlled leaf unfolding rates in `Nellie White' Easter Lily. J. Amer. Soc. Hort. Sci. 113:70-74.

Karlsson, M.G. R.D. Heins, J.E. Erwin, R.D. Berghage, W.H. Carlson and J.A. Biernbaum. 1989. Temperature and photosynthetic photon flux influence chrysanthemum shoot development and flower initiation under short-day conditions. J. Amer. Soc. Hort. Sci. 114:158-163.

Blacquiere, T and J.A. Bakker. 1992. First European workshop in thermo- and photo-morphogenesis on cultivation of ornamentals. Acta Hort. 305.

Erwin, J.E., M. Ascerno, F. Pfleger and R.D. Heins. 1992. New Guinea impartiens production. Minnesota Com. Flower Growers Bull. 41:1-15.

Erwin, J.E. R.D. Berghage and R.D. Heins. 1991. Circadian temperature effects on plant chlorophyll content. HortScience 26:105

Erwin, J.E. and R. D. Heins. 1990. Temperature effects on lily development rate and morphology from the visible bud stage until anthesis. J. Amer. Soc. Hort. Sci. 115:644-646.

Erwin, J.E., R.D. Heins, R.D. Berghage, B.J. Kovanda, W.H. Carlson and J.A. Biernbaum. 1989. Cool mornings can control plant height. GrowerTalks 52:73-74

Erwin, J.E., R.D. Heins, and M.G. Karlsson. 1989. Thermomorphogenesis in Lilium longiflorum Thunb. Amer. J. Bot.

Erwin, J.E., M. Strefeler and R.D. Heins. 1994. Diurnal temperature fluctuations affect vegetable development. Minnesota Com. Flower Growers Bul. 42:6-10.

Erwin, J.E., P. Velguth and R.D. Heins. 1994. Diurnal temperature fluctuations affect Lilium cell elongation but not division. J. Expt. Bot. 45:1019-1025.

Grindal, G. And R. Moe. 1994. Effects of temperature drop and a short dark interruption on stem elongation and flowering in Begonia x hiemalis Fotsch. Scientia Hort. 57:123-132.

Heins, R.D. and J.E. Erwin. 1990. Understanding and applying DIF. Greenhouse Grower 8:73-78.

Hicklenton, P.R. J. Tutty and D.N. Kristie. 1992. DIF, limited DIF and stem elongation patterns in pot chrysanthemums. Greenhouse Canada December 28- 29.

Jacobsen, L.H., M.G. Amsen and O.F. Nielsen. 1991. Negative DIF: Mean room temperature control and its effect on shor-day plants. Tidsskr. Planteavl 95:441- 447.

Jensen, H.E.K. 1993. Influence of duration and placement of a hight night temperature on morphogenesis of Dendranthema grandiflora Tzvelev. Scientia Hort. 54:327-335

Jensen, H.E.K. 1994. Effects of duration and degree of pulse-DIF temperature on plant height and floewering of Kalanchoe blossfeldiana v. Poelln. Scientia Hort. 59:45-54

Langton, F.A., K.E. Cockshull, C.R.J. Cave and E.J. Hemming. 1992. Temperature regimes to control plant structure. Current U.K. R & D. Acta Hort. 327:49- 59.

Lecharny, A., H.Scwall and E. Wagner. 1985. Stem extension rate in light-grown plants. Plant Physiol. 79:625-629.

Miller, W.B., P.A. Hammer and T.I. Kirk. 1993. Reversed greenhouse temperature affect carbohydrate status in Lilium longiflorumThunb. `Nellie White'. J. Amer. Soc. Hort. Sci. 118:736-740.

Moe, R., N. Glomsrud, I. Bratberg and S. Valso. 1992. Control of height in poinsettia by temperature drop and graphical tracking. Acta Hort. 327:41-48.

Moe, R. and R.D. Heins. 1990. Control of plant morphogenesis and flowering by light quality and temperature. Acta Hort. 272:81-89.

Moe, R., R.D. Heins and J.E. Erwin. 1991. Effect of day and night temperature alterations, and plant growth regulators on stem longation elongation and flowering of the long-day plant Campanula isophylla Moretti. Scientia Hort. 48:141-151.

Moe, R. and L.M. Mortenson. 1992. Second European symposium on thermo-and photo-morphogenesis in plants. Acta. Hort. 327.

Tutty, J.R., P.R. Hicklenton and D.N. Kristie. 1992. The dynamics of chrysanthemum stem elongation in relation to day and night temperatures. Acta Hort. 305:61.

Tutty, J.R., P.R. Hicklenton, D.N. Kristie and K.B. McRae. 1994. The influence of photoperiod and temperature on the kinetics of stem elongation in Dendranthema grandiflorum. J. Amer. Soc. Hort. Sci. 119:138-143.

Ueber, E. and L. Hencricks. 1992. Effects of intensity, duration and the time of a temperature drop on growth and floewring or Euphorbia pulcherrima Willd. Ex Klotzsch. Acta Hort. 327:33-37

Vogelzang, J., L. Cuijpers and M. De Graaf- v.d. Zande. 1992. Growth regulation of bedding plants by reversed day/night temperature only? Acta Hort. 305:37-43

Treatments Day/Night Temperature (C)	"Garland"	"Surf"
21/19 Shift at daybreak	15.4a	8.0a
14/22 Shift at daybreak	9.8b	5.8b
14/22 Shift 2h beforedaybreak	10.2b	5.6b

TEMPERATURE AND THE INTEGRATED CONTROL OF PLANT HEIGHT

Problems with DIF