UNIVERSITY of EDINBURGH

DEPARTMENT of AGRICULTURE

HONOURS THESIS

Title Factors affecting the Germination and Establishment of Monogerm Sugar Beet.

Author's Name Robert A R Meikle

Degree B.Sc.

Date May 1981

FACTORS AFFECTING THE GERMINATION AND

ESTABLISHMENT OF MONOGERM SUGAR BEET

by

ROBERT ANDREW ROSS MEIKLE

in partial fulfilment of the requirements

for the Degree of BSc in Agriculture

with Honours in Crop Production Science.

at the

School of Agriculture

University of Edinburgh

May 1981

ACKNOWLEDGEMENTS

I should like to express my sincere thanks to Dr Fred Harper for his guidance and much appreciated criticism throughout the preparation of this thesis. I would also like to thank the staff at the crop production glass-houses for their help with the experiments. Thanks are also due to Mrs E Grosse and Typerite for their excellent typing of the script.

CONTENTS

Page

	LIST OF FIGURES			(i)
1.	INTRODUCTION			1
2.	REVIEW OF LITERATURE			3
	2.1	Introduction to Review		3
	2.2	Seed production and seed treatments		3
		2.2.1	Introduction	3
		2.2.2	A brief history of monogerm seed	4
		2.2.3	Sugar beet breeding	4
		2.2.4	Varieties	5
		2.2.5	Commercial seed production	5
		2.2.6	Post harvest treatments	6
		2.2.7	Pelleting	7
		2.2.8	Alternative seed treatments	8
	2.3	Factors affecting germination		12
		2.3.1	Introduction	12
		2.3.2	Seed storage	12
		2.3.3	Germination tests	13
		2.3.4	Effects of inhibitors on germination	14
		2.3.5	The germination process and the involvement of air and water	15
	2.4	Factors affecting the establishment of beet		19
		2.4.1	Introduction	19
		2.4.2	Plant population	20
		2.4.3	Cultivation and seed bed preparations	21
		2.4.4	Drilling	21
		2.4.5	Field emergence	23
		2.4.6	Predicting sugar beet emergence	24
		2.4.7	The effect of seed size on emergence	26
		2.4.8	Other emergence factors	27

CONTENTS (Contd)

Page

	2.5	Alternative methods of establishing beet		29
		2.5.1	Transplanting	29
		2.5.2	Autumn sowing	30
		2.5.3	Conclusion	30
3.	EXPERIMENTS			31
	3.1	Materials and Methods		31
		3.1.1	Description of seed lots	31
		3.1.2	Seed characteristics	31
		3.1.3	Germination and emergence tests	32
		3.1.4	The effect of water-soluble inhibitors on cress seed germination	36
	3.2	Results		36
		3.2.1	Seed characteristics	36
		3.2.2	The germination experiments	38
		3.2.3	The emergence test	57
		3.2.4	The effect of water-soluble seed extracts on cress seed germination	57
	3.3	Discussion		60
		3.3.1	Germination tests at standard temperatures and 7.5°C with untreated seed	60
		3.3.2	Seed characteristics	63
		3.3.3	Germination tests with treated seed	64
		3.3.4	The emergence test	69
		3.3.5	Conclusion	70
4.	GENERAL DISCUSSION: THE POTENTIAL OF SEED TREATMENTS TO IMPROVE FIELD EMERGENCE			71
5.	SUMMARY			76
6.	CONCLUSIONS			77
7.	REFERENCES			78

LIST OF FIGURES

Page

1. The relationship between seed weight and

true seed weight for Seed Lot 1 39

2. The relationship between seed weight and

true seed weight for Seed Lot 2 40

3. The relationship between seed weight and

true seed weight for Seed Lot 3 41

4. The relationship between seed weight and

true seed weight for Seed Lot 4 42

5. The relationship between seed weight and

true seed weight for Seed Lot 5 43

6. Germination pattern of five seed lots under

standard germination test conditions 47

7. Germination pattern of five seed lots under

low temperature conditions (constant 7.5°C) 49

8. Germination pattern of three seed lots (advanced in water) under standard germination test conditions 52

9. Germination pattern of five seed lots (advanced

in water) under low temperature conditions 53

10. Germination pattern of five seed lots (with GA₃, solution in place of distilled H₂0 in the petri-dishes) otherwise under standard

germination test conditions 55

11. Germination pattern of three seed lots (advanced in GA₃ solution) under standard germination

conditions 56

1. INTRODUCTION

The final yield of sugar obtained from the sugar beet crop (Beta vulgaris L.) is determined by many factors, including: the length of the growing season, the incidence of diseases and pests, and the prevailing weather conditions (Biscoe, Draycott & Jaggard 1980). These factors have a direct or indirect effect on photosynthesis and consequently the amount of assimilate translocated to the roots. It is therefore the duration of the leaves which can produce assimilate (i.e. excluding excessively shaded and diseased leaves) that determines the maximum potential yield.

The actual yield obtained is determined by the amount of solar radiation intercepted and utilised by the leaf canopy over the - whole growing season. Scott, English, Wood & Unsworth (1973) have shown that a good linear relationship between sugar yield and both total incident and intercepted radiation exists.

The grower cannot of course control the amount of incident radiation but his objective must be to ensure as much of it is intercepted as possible. However under the ambient Spring soil temperatures of the U.K., sugar beet is very slow in emerging and establishing, and complete leaf coverage does not occur until July.

It was estimated by Scott & Jaggard (1978) that just over 50% of the total radiation is intercepted by crops in the ground between April and October. The rest strikes bare ground and is wasted.

There is clearly considerable scope for improving the efficiency of interception by achieving full leaf coverage sooner It is in fact the slow rate of growth over the early part of the season that results in late leaf coverage and poor

-1-

-2-

radiation interception. Growers would welcome any feasible method of improving the early growth of the crop.

The objective of this thesis is to discuss the early stages of sugar beet growth and the various problems affecting establishment that exist with sugar beet seed, both in its production and performance in the field.

-3-

2. REVIEW OF LITERATURE

2.1 Introduction to review

The literature reviewed in this thesis considers three main areas, namely seed production, seed germination and field establishment. Standard methods and various alternative procedures for treating and testing beet seeds are firstly described and discussed and then related to field performance. It will be explained why establishment is still a problem in modern crops and why seed quality is of major importance.

In the review and thereafter whole fruits will be referred to as seeds, embryos as true seeds, and the rest of the fruit as the seed coat and the ovary cap (which is considered to be part of the seed coat), as the seed cap unless otherwise indicated.

2.2 Seed production and seed treatment

2.2.1 Introduction

This section is mainly concerned with monogerm sugar beet seed as multigerm seeds are no longer used for sugar production.

The considerable difficulties encountered in breeding and producing good quality monogerm seed are outlined and modern varieties described.

As freshly harvested seed is of poor quality and cannot be used successfully in practice because it contains both empty seeds and multigerms the various processing procedures and alternatives are described in detail. The limitations of certain treatments are considered, as perfect seed lots cannot be produced on a commercial scale.

-4-

2.2.2 A brief history of monogerm seed

The advantages of seeds containing only one true seed were first realised in the early 1900's, but early Breeding Projects in the U.S.A. failed to produce plants with more than 75% single flowers. The idea was abandoned until the 1930's when it was re-investigated in Russia. Plants with complete monogermity were produced but other growth features were poor and commercial varieties were not available until the 1960's due to severe breeding difficulties (Orlovskii 1957) & (Savitsky 1952). Today however, monogerm varieties have almost completely replaced multigerms.

2.2.3 Sugar beet breeding

Breeding sugar beet is complicated by its biennial habit which makes the period taken to produce new varieties much longer than with most other species. It is also an open or out pollinating crop which makes uniformity difficult to achieve. However, several methods of breeding can be used, e.g. mass selection where the largest and best shaped roots are selected to produce seeds which are planted and the best progeny selected for further seed production. The method can be used to reliably improve existing varieties when crossed with other plants with desirable characters. Other methods used singly or in various combinations include hybridization polyploidy and male sterile lines. Some of the main objectives in sugar beet breeding are:-

1 The monogerm character for commercial crops but multigerm pollinators are still necessary.

2 High sugar yield per ha as the best compromise between root yield and sugar %.

3 Low bolting tendency.

4 Good germination and early growth.

-5-

5 High purity.

6 Good root shape.

7 Disease resistance to virus yellows & downy mildew.

(Ellerton 1978).

2.2.4 Varieties

All nine varieties on the recommended list are genetic monogerm and produce over 90% single plants after processing (NIAB 1980). All modern varieties produce similar yields in trials, and have sugar contents over 17%. There are varietal differences in downy mildew resistance, bolting susceptibility, size of tops, and field emergence. Nomo and the newly recommended Monoire show low bolting tendencies when sown early, while Sharp's Klein Monobeet is poor in this respect. Vytomo and Salohill are much less resistant to downy mildew than Monoire or Bush Mono "G".

2.2.5 Commercial seed production

Sugar beet seed is usually grown away from root crops because of downy mildew and virus yellows carry over and possible pollen contamination from bolters (Scott & Longden 1973). Most of the seed used for root crops is triploid hybrid monogerm which is produced by mixing and then sowing together an inbred diploid male sterile monogerm, and a tetraploid multigerm to act as a pollinator. The mixed progeny is mechanically separable into triploid Monogerm and multigerm seed (Johnson 1980, personal communication).

When seed crops are in flower problems such as pollen contamination from weed beet, wild relatives or low fertilisation overall due to unsuitable weather can occur, but the amount of viable seed produced can be low for a number of reasons even if

-6-

good fertilisation is achieved (Scott & Longden 1973), (Battle & Whittington 1969a). The maturity of the fruit at harvest can affect seed performance. More mature fruits have better developed embryos and a lower concentration of germination inhibitors (Scott & Longden 1973) and should therefore give better establishment of the root crop.

The actual numbers maturing at harvest depends on the season and the date harvested. Wet seasons delay rate of maturing and often seed is harvested early to reduce losses by windshake. Tekrony & Hardin (1969) looked for the location of under-developed seed on seed plants and found them on all parts but their occurrence was more frequent on tertiary lateral branches. Field removal is therefore not possible and it is also difficult to do so in processing unless there is a relationship between "seed" size and under-developed seeds.

2.2.6 Post harvest treatments

The standard post harvest treatment to raw seed is a complicated and exacting operation but in outline the following is carried out: -

1) Pre-cleaned (to remove stick & leaf) and dried to below 15% moisture at a temperature not greater than 120°F (49°C).

2) Cleaned with aspiration using round hole screens to remove small and dead seeds.

3) Monogerm seed is separated from multigerm seed using a combination of round hole and slotted screens.

4) The monogerm seed extracted is graded to size specifications - usually 2.75 - 4.00 mm.

5) The graded monogerm seed is gravity separated to extract light "weak" seeds to improve germination.

-7-

6) The product is soaked in a fungicide solution of organic Ethyl Mercury Phosphate (E.M.P.) for twenty minutes - by using a concentration of 40 ppm. E.M.P. and a ratio weight of three parts water to one part seed.

7) The final seed product is then regraded to size specifications (Lindsay 1980, personal communication).

8) Pelleted.

2.2.7 Pelleting

Nearly all the seed sown in the U.K. is pelleted as there are several advantages over raw-rubbed seed. The major advantage is more accurate drilling and others are safe dressing with fungicides, protection from mechanical damage and sowing depths and spacings are easily checked (Charlesworth 1978).

There are many materials and methods which can be used for Clays such as Cellite, Montmorillonite, Vermiculite or Bentonite, or other materials such as cork, peat, chalk, sand or even beet cortex can be used. Non-clay materials may need a sticker such as ethyl cellulose to hold the pellet together. Clay structures generally adhere with water only, but they need sufficient physical strength to resist damage in the drill but breakdown easily on contact with soil water.

The three main methods of pelleting are stamping, coating or rolling. Stamping is a dry process and is therefore rapid but any additive must be evenly distributed throughout the pelleting material, while coating and rolling permit layering of additives anywhere within the pellet.

Drying after pelleting is necessary and slows the procedure down (Longden 1975).

-8-

The most popular process in the U.K. is the Germain's "Filcoat". About 1600 tonnes of seed are pelleted annually with a secret clay-based medium (Charlesworth 1978), plus methiocarb insecticide @ 4 Kg/1000 Kg seed (Johnson 1980, personal communication). Manganese Oxide can also be incorporated for use in deficient soils (Farley & Draycott 1978).

2.2.8 Alternative seed treatments

Sugar beet seed has to be processed between harvesting and sowing to eliminate inferior seed, and give good seed a better chance of good and fast emergence. The standard procedure listed removes empty and shrivelled seeds and multigerms, and gives protection against some pests. and pathogens but often field performance is far from perfect, therefore better treatments should be sought. The problem in finding suitable alternatives is not that treatments do not improve seed performance when used singly but more often because there are practical difficulties or incompatibility with existing treatments or simply the cost and time involved (Tonkin 1979).

(i) Washing

Longden (1973) showed how 21 rinses in 3.5 hours gave the best response in glasshouse and field trials with natural and partially processed seed; but when agitated, washed seeds showed a greater tendency to lose seed caps and therefore this was an unsuitable treatment. Washing had similar effects on the performance of both mature and immature seed.

(ii) Advancing

The object of advancing is to allow seeds to imbibe water and develop physiologically up to the point of

-9-

radicle elongation and then be dried back, holding the seeds at this point before sowing. This reduces the period between sowing and emergence in the field.

The optimum advancing technique for sugar beet was determined by Longden (1971), (equal weights of seed and water in airtight containers for 24h, repeated 3 more times with a 48h air drying interval after each soaking). Advancing increases the number of cells in the true seed but the treatment does not increase the size of individual cells so that the germination process does not therefore result in large changes in embryo volume before the radicle elongates, but like washing, advancing increases the tendency for premature loss of the seed cap. An extremely gentle drill mechanism would therefore be required. Fluid drills may be developed for practical use.

While washing was calculated to be equivalent to a very much higher than normal rainfall over the seed plant, rainfall effects are more similar to advancing than washing, i.e. a small amount of water in contact with the seed for a longer time.

(iii) Osmotic priming

This treatment is also designed to get the seed in a better physiological state for quick emergence. Trials with various concentrations and priming periods in polyethylene glycol or salt (KNO₃ + K₃ P0₄ H₂0), (Longden, Johnson, Darby & Salter 1979) gave poor and inconsistent results and no reduction in the

-10-

time spread of germination. Priming is therefore too unreliable to use in practice.

(iv) Water steeping

Longden et al (1979) also looked at the effects of a 24h soak in 10 x the seed volume of water @ 1°C followed by re-drying. Steeping gave uniform germination but poor field performance. Scott, Wood & Harper (1972) however found that a 24h soak in water @ 20°C hastened emergence and improved final germination. The steeping process probably has effects on physiological developments and also removes germination inhibitors.

(v) Plant growth regulators

Scott et al (1972) compared steeping seed in solutions of kinetin (K), 6-benzyl-aminopurine (BA) and gibberellic acid (GA₃) with water only. All solutions gave better field emergence than water but also showed a response to concentration. The optimum concentrations were BA 1 - 10 ppm, K 50 - 100 ppm, GA₃ - 100 ppm which nearly doubled the seedling weights.

(vi) Size grading of seed

Longden, Scott & Wood (1974) devised a size grading method which could produce a seed lot containing 90% monogerms and a germination of 80%, providing the unprocessed raw seed had at least 50% germination. The method included grading by thickness to remove multigerms, and grading by diameter and aspiration to remove small and non-viable seeds. The variation in size range was also reduced in the process. However, an ideal sequence for size grading cannot be

-11-

formulated as there is much variation between original seed lots due to effects of season, seed production agronomy and variety. The optimal adjustment and sequence for one seed lot may be totally unsuitable for another.

(viii) Treatment combinations

Any number of treatments can be used on a seed lot but to be of value they should be

a) fully compatible,

b) additive in effect or, preferably,

c) synergistic (Longden 1976).

Longden (1976) tried washing, E.M.P. steeping, Thiram soaking and advancing in sequence on a seed sample of Amono graded 3.18 - 3.57 mm. One or more of the treatments were omitted on other Amono samples so that a total of 16 different combinations of treatments were obtained for comparison. When analysed, different aspects of seed performance were affected and complex interactions occurred - some combinations gave faster emergence, others larger shoots. In general, as more treatments were applied detrimental effects were observed, e.g. E.M.P. had positive effects on seed performance when used singly, but negative in combination, (when additional soaking let Hg penetrate the embryo). It was probable that there was considerable overlap in the effects of single treatments and there was no evidence of cumulative or synergistic benefits from combinations. Therefore only one treatment involving water or a solution should be used on a seed lot and in practice this is the E.M.P. steep to control seed borne fungi.

-12-

2.3 Factors affecting germination

2.3.1 Introduction

This section is mainly concerned with laboratory seed testing as affected by storage, natural inhibitors and various treatments. The most important factors in a petri-dish test are the access of oxygen to the true seed and of water which is discussed in some detail. The importance and limitations of germination tests are considered.

2.3.2 Seed storage

Sugar beet storage is not a problem in the U.K. The cool ambient conditions allow commercial seed to remain usable for several years. Longden & Johnson (1974b) studied the effects of storage temperature and water content with time on seed performance with pelleted and unpelleted Amono and Monobeet. Both higher water content and temperature lead to quicker loss of viability over the range 5 - 18% H₂0 and 2 - 22°C; in the extreme case of storage under conditions of 18% and 22°C, seeds were completely dead after 13 months storage. Seed stored in open containers @ 10°C produced about 9% fewer seedlings per annum relative to the initial value (= 100) over five years, while both pelleted and unpelleted stored in sealed thick polythene bags showed almost no decline with a water content of 5 - 10%.

Sometimes an increase in germination over the first year of storage was observed. This was most probably due to a combination of saprophytic fungi on the seed surface dying and loss of post harvest dormancy. Storing at lower temperature (2°C) gave seedlings of lower dry weight when tested. Storage of beet seed at about 10°C is considered appropriate and this temperature is relatively cheap and easy to maintain.

-13-

In warm humid climates seed viability does decline and methods of minimising losses are required. Basu & Dhar (1979) found that soaking in 5 x the volume of water followed by re-drying before storage reduced the loss of viability over three months of storage. No chemical solution had a better effect than water but the exact mode of action of the hydration/dehydration process is not known. There may be an effect on inhibitors in the seed or on free radicals.

Another problem is that seed vigour deterioration occurs before loss of viability can be determined by standard germination tests. Seed vigour is a measurement of the ability of seeds to germinate or emerge under non-optimal conditions. Low vigour is associated with slow emergence which is extremely undesirable for high sugar yield.

2.3.3 Germination tests

The requirements for a standard germination test for many species including Beta vulgaris are defined by the International Seed Testing Association (Anon 1966). The conditions for sugar beet are 16 hours @ 20°C, and 8 hours at 20 - 30°C per day with germination counts at 3 and 14 days. Light is not essential and the test can be carried out on top of or between filter paper, or in sand. The seed should be pre-washed for 1 hour in water at 25°C.

The object of the standard test is to gain information about the field planting value of the seed under test and to compare it with other seed lots. The defined conditions are designed to allow the seed lot to express regular, rapid, and complete germination, and also be repeatable within the limits of the random sampling. However, sugar beet field emergence is not accurately predicted by the standard test (Brown 1980).

-14-

Hibbert and Woodwark (1969) have tried other laboratory tests using pleated paper or flat paper in sealed containers at different temperatures and periods of counting. Results were similar but not interchangeable and the inherent variability of seedlots makes emergence prediction a procedure with low precision. Sand is considered to be an unsuitable medium for germination tests but Snyder and Filban (1970) in the U.S.A. praised a test for emergence potential of seed from a standardised sand tray.

Hibbert, Thomson and Woodwark (1975) and Reiff (1976) cited by Johnson (1979) found that pleated paper gave good laboratory germination, as the contact between seed and paper was better than with flat filter paper. Brown (1980) found low temperature results more accurate for field emergence prediction, but this procedure is lengthy. Bonscheur (1975) cited by Johnston (1979) found that the speed of germination but not the final value was affected by varying temperatures and water contents, but Heydecker, Orphanous and Chetram (1969) recommended that care should be taken not to penalise seed by either excess or a lack of water when under test (using garden varieties of red beet).

2.3.4 Effects of inhibitors on germination

Many substances have been isolated from the seed coats of sugar beet which are potentially inhibitory to germination, most of them are organic acids. Snyder, Sebeson & Fairley (1965) considered Oxalate to be the major inhibitor but did note some complex interactions with others. It was concluded that effects were specifically inhibitory rather than osmotic. Sebeson, Mitchell & Snyder (1969) studied the effects of inhibitors on Alpha-amylase activity on starch solutions (hydrolysis of starch is an essential process before germination commences). The

-15-

inhibitory effects of Caffeic, Ferulic, Gallic, P-hydroxy, Benzoic and Vanillic acids, which are known to exist in beet seeds, were examined using Alpha-amylase and excised embryos. Gallic acid was found to be most inhibitory and, in general, the degree of inhibition increased with concentration in both experiments, but the effects were less pronounced on the embryos, indicating that some form of de-toxification must occur in the seed. Battle & Whittington (1971) showed that the maternal genotype influenced the early field behaviour of the progeny through control of the level of inhibitors in the seed. In earlier experiments, Battle & Whittington (1969b) showed that the inhibitors were situated in the perianth and pericarp, and also that the various other acids, including Abscisic, were involved. It was postulated that the free phenolic acids were in equilibrium with acetone insoluble esters involved in lignin bio-synthesis. Immature clusters would have a relatively higher proportion of P-coumaric acid associated with an earlier stage in lignin biosynthetic pathway.

Inhibitors may also act as 0₂ acceptors and affect germination in this way (Heydecker, Chetram & Heydecker 1971).

2.3.5 The germination process and the involvement of air and water

Water and oxygen must be taken up before germination can occur. The figures on Pages 16 and 17 show that the embryo is concealed inside the seed coat which acts as a barrier to both, but the basal pore does permit entry. The pore is not usually open, but contains the remains of the vascular connections between the mother plant and the embryo.

Perry and Harrison (1974) almost completely prevented germination by blocking the pore with Vaseline. Entry between seed cap and the seed coat does not occur as Heydecker et al (1969) had earlier described.

-16-

(i) Monogerm seed of variety Monohill (Aura 1975).

(ii) Opening of the seed cap and radicle emergence after two days (Aura 1975).

-17-

(iii) Diagrams showing external view and out-section of a true seed (Lakon & Bulat 1958 cited by Aura 1975).

(iv) Diagram of a transverse section of a monogerm seed (Perry & Harrison 1974).

-18-

Severe inhibition has also been described in many papers including Chetram & Heydecker (1967) and Heydecker & Chetram (1971) by excess water in the test substrate. Excess water tends to be taken up and held in the basal pore by capillary action. Consequently, oxygen can only enter by diffusion, through the water at a very slow rate.

Perry and Harrison (1974) applied Fick's Law to the dimensions of the basal pore and estimated that, when air filled, the oxygen diffusion flux was 4.5 ml/h and when water filled only 8.5 x 10^-3 μl/h. An embryo requires about 0.8 μl/h to germinate and therefore the process is inhibited when the pore is full.

The observed uptake of oxygen in a water filled pore was 0.14 μl/h. The difference from the water filled pore estimate was due to microbial respiration.

As germination continues cell number increases (Longden 1971) then the radical elongates forcing the seed cap open thus making more oxygen available to the embryo facilitating faster elongation. Coumans, Côme & Gaspar (1976) showed that in a wet medium removing the cap before germination resulted in more seed germinating if positioned "face up" but not if "face down". Therefore the external film of water could also inhibit germination. Peto (1964) cited by Heydecker and Chetram (1971) chipped the seed cap beforehand and improved germination by both allowing more oxygen in and reducing the mechanical effort required by the radical to remove the cap. Heydecker and Chetram (1971) viewed germination as more than a physiological process, i.e. complex ecological and microbiological components are involved too. When 8 ml of water was used in a laboratory test dish it was excessive, but adding aureomycin to inhibit bacteria leaves more oxygen for embryos to use. Washing seeds upsets the ecological balance between bacteria, fungi and the inhibitors in the seed coat, and

-19-

the depressed laboratory germination of seeds treated with fungicides may be explained by a similar change to the micro-environment.

As germination is improved by cap removal or inhibitor removal, whether water is excessive or not, 0₂ uptake and respiration must therefore precede germination and not be a consequence of it! However, Heydecker et al (1971) noticed that in red beet seeds when the concentration of inhibitors was high, the oxygen uptake was high also. Therefore, some process different from normal respiration was occurring. There may be competition for oxygen between different metabolic pathways with imbibed seeds - inhibiting respiration at low concentrations and uncoupling, i.e. preventing access to the true seed, if at higher concentrations but an actual mechanism has not been found.

Coumans et al (1976) viewed the seed coat as a physiochemical barrier to oxygen. It (a) restricted diffusion and (b) actually absorbed oxygen so that very little if any reached the embryo through the seed coat.

Chetram & Heydecker (1967) and Heydecker et al (1969) found that hydrogen peroxide in solution was an excellent way of supplying oxygen to the embryo to improve germination.

2.4 Factors affecting the establishment of beet

2.4.1 Introduction

Regular sugar beet stands used to be achieved by sowing seed at a high rate and subsequently hand hoeing unwanted plants after emergence. However, now that monogerm seed, precision drills and suitable herbicides have been introduced, regular stands can be achieved without handwork. Nevertheless, modern methods of establishment are not always completely successful as Bray (1980)

-20-

has shown. Nearly half the area "drilled to a stand" receives at least a small amount of handwork.

The main reason for this is that sugar beet field emergence is difficult to predict even when laboratory germination is known. This section discusses the factors affecting emergence but firstly optimum or target populations are considered.

2.4.2 Plant population

Hull and Jaggard (1971) reviewed attempts to determine the population for maximum yield of sugar and found many factors, viz. soil type, irrigation, sowing technique and fertilizers, affected this optimum population. They generally concluded that this was 65,000 plants/ha on a fertile soil rising to 85,000 on poorer soils, but a few thousand above or below did not seriously depress yield. Goodman (1966) recommended 74,100/ha with a leaf area index (L.A.I.) of 2.8 (which does not intercept all available radiation but ensures there are no non-productive leaves). The above recommendations were based on hand-hoed situations only where a dense crop was sown and subsequently thinned. Draycott & Durrant (1974) looked at populations in relation to other cultural practices, and showed 50,000 or above was adequate without hand-hoeing. It was also shown that between a rectangularity of 1:2 and 1:1 yield was not affected at higher populations (86,000/ha) but Hull & Jaggard (1971) showed that 45 cm rows were more suitable for high populations and 60 cm for lower populations to reduce interplant competition in both cases. Interplant competition was shown to reduce individual plant yields considerably by Draycott & Durrant (1974) who compared sugar yields from plants grown in a competition free plot (22,000/ha with minimal nutrient and light competition and adequate soil water) with denser stands.

-21-

At 22,000/ha individual plants yielded 315 g of sugar/plant but at 81,000/ha individuals yielded only 113 g, but a much higher total yield/ha.

2.4.3 Cultivation and seed bed preparations

Cultivation for sugar beet should make the best use of the available environmental conditions over the preceding Autumn and Winter. The following practices should ensure a reasonable quality seed bed in the Spring.

1) Plough early with a reversible plough to gain the benefits of an even surface for an even depth of weathering, and as long a weathering period as possible.

2) Use as few passes as necessary with wide wheel extensions, to minimise excessive consolidation. Use wide implements and tandem arrangements.

3) After weathering use shallow cultivations only to avoid bringing clodding unweathered material up into the seed bed.

4) Form a coarse tilth below the surface for drainage but a finer tilth on the surface for water conservation.

5) A level seed-bed should be achieved by use of straight and rolling tined harrows or power harrows, so that precision drills can be used and drilling depth controlled (Spoor 1978, Clare 1976 and Rose 1972).

2.4.4 (i) Drilling

Precision drills are now almost universal in use and are essential for the "drilling to a stand" technique. A precision

-22-

drill is defined as one which selects and deposits seed at predetermined distances. Common features of precision drills are: land wheel drive, minimum seed drop, boat shaped coulters and flat rollers (Rose 1972). In the U.K. the "Stanhay" pinched rubber belt drill is used, but disc types are popular elsewhere (Hull & Jaggard 1971). Ten or twelve row machines are necessary to compensate for slow forward speed of precision drills, but the N.I.A.E. have a test drill accurate at 11 km/hr (Hayward 1978). Munday (1977) has shown that no commercial drill sows perfectly. Doubles, singles or multiple misses and inconsistent spacings are always observed, but the seed can confound the drill performance when doubles are due to pellets with extra embryos and misses due to dead or empty ones. However, better drills that reduce the amount of seed roll and that are more accurately space-calibrated are required for "drilling to a stand".

Beet seeds should not be sown below 3.8 cm due to the small perispermic reserves. Early sowing should be shallow (< 2 cm) for good emergence. If sown later then 3 cm is better as the surface dries out (Hull & Jaggard 1971, Hibbert et al 1975).

Alternative drilling techniques such as fluid drilling of pre-chitted seed (Longden et al 1979, Currah 1978) have as yet unsolved technical problems and cannot be used for fast and even emergence in sugar beet.

(ii) "Drilling to a stand"

Drilling to a stand" is only successful with precision drills, good emergence and relatively weed free fields. In the U.K. 12 - 15 cm spacing in 50 cm rows is commonly used to achieve 74,000/ha but Fletcher (1974) has shown that no universally recommended spacing) is possible as localised factors are involved. The technique works best for April sowing when compensatory growth

-23-

is adequate to make up for irregular spacing. Neeb and Winner (1970) cited by Hull and Jaggard (1971) deliberately mixed good and dead seed to encourage irregular spacing and reduce population and still found a linear relation between population and yield up to 80,000/ha. However, Thomson (1956) cited by Hull and Jaggard (1971) also deliberately obtained an irregular stand by random hand-singling and found 0.5 t/ha less from an irregular stand than from a similar regular, hand-singled stand. Knott, Parker & Mundy (1976) found that with "drilling to a stand" irregularity effects were worse with low populations, made with wide rows and spacing, and at the same time found 70,000/ha was optimal for a fen soil, but only 50-56,000 for a silt.

2.4.5 Field emergence

Aura (1975) categorised four factors involved in emergence:

1) The germination energy of the seed.

2) The appearance of pathogens.

3) Mechanical soil resistance.

4) Soil, air & water content.

(i) Germination Energy. Perry (1973) showed monogerm final emergence to be reduced by high soil water levels and by compaction, but not by low soil temperature. However, seed lots responded differently showing that seedling vigour is important.

(ii) Effect of fungicide. Heydecker & Chetram (1971) showed that seed fungicide treatment improved field performance even if laboratory tests showed the reverse.

-24-

(iii) Hegarty and Royle (1978) measured the impedance of soils covering sugar beet seeds and showed a negative linear correlation with final emergences. Perry (1973) also showed how soil capping caused by irrigation water reduced emergence by 30%.

(iv) Aura (1975) considered oxygen uptake was only seriously hindered by water in the seed or the surrounding film, when soil water potentials were close to zero, i.e. in very wet soils. Poor contact between soil and seed could reduce germination by restricting water uptake in dry soils, if the soil water potential and was less than - 10 atm no emergence occurred. Aura also noted water diffusion to be slower through the seed than through the soil. Pelleted seed emerges better in wet soil conditions (Perry 1973), and Hibbert et al (1975) postulated that pelleted seed may have a higher water requirement and should be sown slightly deeper in dry conditions. Aura (1975), however found pellets restricted oxygen uptake in very wet conditions.

2.4.6 Predicting sugar beet emergence

The standard germination test is fair to the seed in that individual seeds have minimal stress and therefore if germination is possible it should occur. However, sugar beet growers require information on field emergence potential and therefore tests with inbuilt stresses may be more appropriate. Perry (1973) has shown how field stresses affect beet seed lots differentially and prior knowledge of this would be very useful for seed selection for a particular situation.

-25-

Brown (1980) suggested low temperature tests (5-7°C) as viable seeds which germinate at 20°C may not do so at lower temperatures which would be experienced in the field. Longden, Johnson & Love (1970) developed a radiography test for laboratory emergence and Longden and Johnson (1974a) compared a radiography test with other methods, i.e. leachate conductivity, a pleated paper test (Hibbert & Woodward 1969), and growing in compost, for prediction of field emergence.

The radiograph prediction is based on x-ray photographs of seed, assessed visually into "good", "dead" and "uncertain" categories. Filled cavities, shrivelled seed, empties and double embryos can be identified with this method. However, not all seed classified as good will germinate, and therefore radiography tends to over-estimate laboratory germinations. Germination in compost was predicted equally well with radiography and standard germination tests. For field performance the leachate conductivity method was hopelessly inaccurate with pelleted seed and poor with unpelleted seed and was therefore discarded as a prediction method. The compost test was more accurate but took three weeks and was roughly equivalent to a radiography prediction. However, radiography does not work for pellets. Overall the pleated paper and the standard test gave the best prediction. It was noted that a low laboratory germination always resulted in poor field performance, but a high laboratory value could result in a high or low field result.

The method of prediction of field performance was a linear regression, e.g. Y = 0.64 X + 6.2 where Y is field emergence (%) and X is laboratory germination (%) but using a field factor is simpler. A field factor of 72% would mean a grower could expect 72% of the laboratory germination to emerge in the field.

-26-

The field factor can be applied to this formula (Bleasdale 1963 cited by Longden & Johnson 1974a).

P x 100

N = ----------

G x F

Where:-

N = no. seeds required/ha

P = desired population, plants/ha

G = laboratory germination %

F = field factor.

However, as sugar beet establishment is variable, estimating the field factor may not always be accurate enough for practical use of the formula, but it shows that improving the field factor by preparing better seed beds will reduce the number of seeds required to produce a regular stand.

2.4.7 The effect of seed size on emergence

Snyder & Filban (1970) compared fruits graded 2.58 - 2.98 mm and 3.77 - 4.12 mm in diameter sown at different depths, 3.2 cm and 5.1 cm in sand. The deeper and smaller seed reduced establishment. The hypocotyls of the deeper sown seeds were heavier as increased impedance forced them to widen.

In the field, emergence was as shown:-

	Depth of	sowing
Seed size (mm)	3.2 cm	5.1 cm
2.58 - 2.98	37%	20.5%
3.77 - 4.12	51%	38%

Snyder and Filban (1970) recommended seed rates should be increased by 33 per cent when using small seed, and that small seed should always be sown at less than 2.5 cm. When soil

-27-

moisture is limiting, deep sowing of large seed should be used. Lindsay (1980 personal communication) showed that seed less than 2.75 mm is eliminated in processing and so only large seed is used in practice. There is nevertheless still a size range as investigated by Scott, Harper, Wood & Jaggard (1974). The effects of seed size on development and yield of sugar beet were studied. small (2.8 - 3.6 mm), Medium (4.0 - 4.4 mm) and Large (5.2- 5.6 mm) seed were used.

Overall the embryo weight was about 20% of the "seed" and larger seeds had fewer empty cavities. They confirmed that the large seed within a seed lot gave a higher percentage emergence and was more reliable.

After processing, larger seed still performed better.

Plants produced from larger seed had higher root: shoot ratios but at harvest plants from medium and large seed produced similar yields. Possibly selecting for seed size also selected different genotypes.

2.4.8 Other emergence factors

(i) Fertilizers

Providing the recommended rates are used (100 Kg/ha N, 100 Kg/ha P₂0₅ & 200 Kg/ha K₂0 MAFF 1979), fertilizers do not have a large effect on establishment except for nitrogen which can be used to encourage early leaf growth. Nitrogen should be applied in the Spring at least two weeks before sowing to allow high localised concentrations to be diluted by rainfall before scorch damage of seedlings occurs (Last & Draycott 1979). The other nutrients, and, if required, Na and Mg, can be applied in the Autumn (Draycott 1977).

-28-

(ii) Fungicides

Byford (1977) studied the effects of mercury fungicides on emergence. Maneb, Captafol and 2-(Thiocyanome-thylthio) benzothiazole (T.C.M.T.B.) seed treatments were compared with the standard E.M.P. treatment. No treatment was as good as E.M.P. and only maneb was similar. It has been estimated that national emergence would be 10% lower if E.M.P. was replaced by another fungicide. Other environmental factors had much larger effects on emergence overall, but E.M.P. treatment is definitely justified as the untreated control had a 20% lower emergence.

(iii) Pests

Pest control measures in addition to seed dressing are sometimes required for pests which effect establishment. Millipedes and beetles can be troublesome but Gamma HCH worked into the seed bed before sowing gives control. Wood mice which dig up seeds and seedlings can be controlled by traps or poisoned food if numbers are high after a mild winter (Farmers' Weekly 1979).

(iv) Weeds

Weeds are no longer a major problem to establishment as good herbicide control is now possible with soil acting herbicides such as chloridazon or ethofumesate (Bray 1980).

(v) Cold injury

Cary (1975) showed that seedlings on the point of emergence were more sensitive to frost damage than at any stage before or after, and that no chemical used to promote winter hardiness in other species

-29-

affected sugar beet similarly. However, germination at low temperature and osmotic treatments did reduce sensitivity to frost damage.

(vi) Soil temperatures

The effects of soil temperature are more important than the actual sowing date as Scott et al (1973) have shown. The date that the amount of accumulated day degrees over 5°C begins to increase rapidly is a good indicator of the onset of growing weather. However, this can occur at any time between lst March and 20th May, but most commonly between the lst and 10th of April. It is usually best to sow just before the onset of growing weather.

2.5 Alternative methods of establishing beet

2.5.1 Transplanting

Scott & Bremner (1966) investigated the potential of transplanting with multigerm varieties as an alternative to contemporary practises and found that an extra ten tonnes/ha of roots and more tops could be obtained. The technique resulted in more fangy and globular roots developing, but the desired plant populations were easily obtained. The larger leaf area duration of the transplants made them more drought sensitive, but in the experiment drought did not seriously check yield. High populations were not required as it was realised that it was the period of ground cover and not the L.A.I. obtained that had the main effect on yield.

In 1966 it was postulated that commercial success would be possible if suitable mechanisation at an acceptable cost could be developed but it was not forthcoming in the U.K. However, in Japan and Bavaria transplanting is used.

-30-

More recently ADAS have started to re-investigate the technique and preliminary trials with monogerm varieties in paper mini-pots show promise of a 40% yield improvement (Farmers' Weekly 1980).

Some other advantages have been noted, namely a reduced dirt tare and better root shape which contrasts with Scott & Bremmer, but handling costs are still prohibitive.

2.5.2 Autumn sowing

Wood & Scott (1975) showed that October and September sowing gave full leaf cover by mid-June, thus making more efficient use of radiation, up to 40% of the available between April and June. Thereafter bolting occurred and final sugar yields were never better than Spring sown crops, even if hand rogued. Ethephon applied at 10 g/1 partly controlled bolting but killed 55% of the plants. At a lower dose (2 g/1) it was ineffective.

It was concluded that Autumn sowing was unsuitable for the U.K. unless extremely bolting resistant varieties could be developed. This is unlikely as flowers are needed for breeding.

Autumn sowing is also inadvisable as virus yellows and downy mildew would be carried over more easily. Autumn sowing is practised in warmer climates, viz. Italy and Japan where low temperature induced bolting is not a problem.

2.5.3 Conclusion

In this review, sugar beet establishment has been shown to be the "weak link" in the production of high sugar yields. The various alternatives for all stages from seed production to establishment all have too many drawbacks to use in practice. The following experimental section is principally concerned with methods of testing and improving the laboratory performance of beet seed, which may result in better field establishment and higher yields.

-31-

3. EXPERIMENTS

3.1 Materials and Methods

3.1.1 Description of Seed Lots

The general objective of the experimental work was to compare the germination performance of five different lots of sugar beet seed under several different test conditions with various treatments, and relate the performance to features of the seed lots. The five seed lots used were as follows:-

Lot I, variety Monotri, harvested in 1977;

Lot 2, variety Monotri, harvested in 1975;

Lot 3, Bush Mono '"G", harvested in 1979;

Lot 4, Amono, harvested in 1979;

and Lot 5, Nomo, harvested in 1975.

Lots 1 2 and 5 had been lightly rubbed after harvesting but not further treated, while Lots 3 and 4 had received the complete commercial processing and pelleting procedure as described in Sections 2.2.6 and 2.2.7.

3.1.2 Seed Characteristics

One hundred air-dried seeds from each lot were weighed, (before and after de-pelleting by washing in running tap water where necessary). The true seeds were dissected out from all seeds by lifting the seed cap with a dissecting needle, after soaking in water for 12 hours, and then weighed individually. The relationships between total and true seed weights were determined. Moisture contents of each seed lot were also determined using an infra-red moisture meter and milled seed samples.

-32-

3.1.3 Germination and Emergence Tests

The purpose of Experiment 1 was to test all five seedlots under conditions similar to those defined by the International Seed Testing Association (ANON 1966), as described in Section 2.3.3.

Germination counts were taken at intervals to determine the final germination percentages, and mean germination time (M.G.T.).

The procedure used for this test was as follows:-

1. Four groups of fifty seeds were counted out from each of the rubbed seed lots. Small, damaged and multigerm seeds were excluded as it was assumed further processing would also have removed them.

2. Four groups of fifty pelleted seeds were counted out from the commercial seed lots (without selection). The pelleting material was removed by washing.

3. All seeds were soaked in tap water for approximately l½ hours.

4. After a short period of air drying, each lot was placed in a 9 cm petri-dish containing three Whatmans' grade 181 filter papers and 5 ml of distilled water. The dishes were covered and placed in a temperature controlled incubator without illumination. The temperature was maintained at 20°C for 16 hours and 25°C for 8 hours per day (standard temperatures).

5. In this test germination counts were made initially at 2 day intervals, but the counting interval increased as germination approached completion.

6. Seeds were counted as having germinated when the radicle had forced the seed cap open and could be seen emerging from it.

-33-

7. As some of the selected seeds contained more than one true seed, despite the attempt to exclude them, they were considered to have germinated if one or more radicles appeared.

This procedure was chosen after considering the observations of Chetram and Heydecker (1967), Perry and Harrison (1974), and Hibbert and Woodwark (1969). Five ml of distilled water on three filter papers was known not to be excessive for germination. Experiment 1 was repeated at the end of the experimental period, (October 1980 - February 1981) to test for changes in germination performance in any of the seed lots.

The aim of Experiment 2 was to test seed germination at a lower temperature than the standard recommendation. It was suggested by Brown (1980) that spring-sown seed experiences seed bed temperatures well below those recommended for the standard test (ANON 1966). The procedure used was the same as in Experiment 1 except that the incubator was maintained at a constant 7.5 C, and also seed lots 2 & 5 were started two days before the others. This was because it was anticipated that these lots would take longer to reach the period when most seeds germinate. The temperature was selected to be low enough to allow germination, but not so low as to considerably prolong the duration of the experiment (Brown 1980).

The petri-dishes used in Experiment 2 which still contained ungerminated seeds after 30 or 32 days for Lots 1, 3 & 4, and 2 & 5 respectively were transferred to a cabinet at standard temperatures as germination at 7.5°C was considered to have been complete.

Germination counts were taken to assess the proportion of seeds which would germinate under standard conditions but not at 7.5°C. After a further 19 days germination at standard temperatures was considered complete.

-34-

The remaining ungerminated seeds were dissected to determine qualitatively if the true seeds were shrivelled or absent, or apparently normal.

Experiment 3 was a repeat of both Experiments 1 and 2 with advanced seed. The aim was to test the effect of advancing as a seed treatment, on germination performance at both standard and low temperatures. The advancing procedure used was as follows (Longden 1971):-

1. Samples from the pelleted seed lots were washed to remove the pelleting material.

2. Approximately 1000 seeds from each of the rubbed seed lots were weighed and placed in a 9 cm petri-dish.

3. Stage 2 was repeated with the de-pelleted seeds.

4. Tap water was added to each dish in an amount approximately equal to the weight of each seed sample in each dish.

5. The dishes were covered for 24 hours at room temperature, then uncovered with the contents spread out to dry for a further 48 hours.

6. Stage 5 was repeated twice, so that each seed sample received a total of three advancing cycles.

The seeds were then counted and set up as for Experiments 1 & 2, except that the 1½ hour pre-soak was omitted. The test at standard temperature was named Experiment 3 (i), and the low temperature test Experiment 3 (ii). It was noticed that during the advancing procedure some seeds germinated, and seed caps became detached from others, particularly in the de-pelleted samples. This was also observed by Longden (1971, 1973). However, only seeds entirely intact after the advancing treatment were selected for testing, and in Experiment 3 (i) seed Lots 3 & 4

-35-

were not used as there were too few intact seed left in the advanced seed stocks.

The aim of Experiment 4 was to test germination performance with a solution of Gibberellic acid (GA₃) in place of distilled water. The procedure used was therefore the same as in Experiment 1 except that 5 ml of a 100 ppm solution of GA₃ was placed in the petri-dishes, and germination counts were initially made at daily intervals. The concentration used had earlier been found to be optimal for improving germination performance if used in a 24 hour steep before testing (Scott et al 1972).

Experiment 5 was a repeat of Experiment 3 (i) (advancing and testing at standard temperature) except that, the tap water used for advancing was replaced by a 100 ppm GA₃ solution. The aim of this experiment was to test the effects of advancing with GA₃ on germination performance. As Experiment 5 was also effectively a treatment combination test, either additive effects or interactions may be observed.

The final test (Experiment 6) was an emergence test. Four replicates of 100 seeds or pellets from each seed lot were sown 2 cm deep in trays containing John Innes number 3 compost. The trays were placed in an illuminated (16 hour photo-period) glass house at approximately 16°C. The aim of this experiment was to examine the relationship between germination and emergence from compost, of the seed lots.

After 16 days seedlings were counted and cut off at soil level. The dry weights of the cut seedlings were assessed after oven drying for 24 hours at 90°C. A final count was made 27 days after sowing for slower emerging seedlings but no dry weights were recorded.

-36-

3.1.4 The effect of water-soluble seed extracts on cress seed germination

1. 30 g of seeds from Lots 1, 2 & 5 were pulverised in a hand mill.

2. The milled sample obtained was mixed with 80 ml of distilled water, periodically shaken and left in sealed bottles for 2 days.

3. Five ml of liquid extract was pipetted into petri-dishes containing 3 Whatmans' grade 181 filter papers and 50 cress (Lepidium sativum) seeds. The dishes were incubated as in Experiment 1 with daily counts.

4. Stage 3 was repeated with the extract diluted to 0.5, 0.25, 0.1 and 0.01 of the original concentration. Cress seeds were counted as having germinated when radicles >1 mm were observed. The final germination percentages and M.G.T. of the cress seeds were determined. Lots 3 & 4 were not used for this determination as chemical treatments in the processing procedure may have interfered with cress germination.

3.2 Results

3.2.1 Seed Characteristics

The mean values for various seed characteristics are presented in Table 1. A large proportion of the pelleted seed lots was in the form of the clay pelleting material, but after de-pelleting the mean seed weights of the pelleted seed lots (Lots 3 & 4) were similar to the mean weights of Lots 1 & 5. However, the mean seed weight of Lot 2 was lower than the others.

-37-

TABLE 1: SEED CHARACTERISTICS

SEED			SEED LOT
CHARACTERISTICS	1	2	3	4	5
Mean Pelleted weight (mg)	-	-	69.4	69.1	-
Mean seed weight (mg)	10.39	8.67	11.24	11.57	11.29
Standard Deviation	±1.62	±2.10	±2.11	±2.41	±1.37
Co-efficient of variation	±15.6	±24.2	±18.8	±20.8	±13.2
Proportion of pellet as seed (%)	-	-	16	17	-
Moisture content of seed (%)	10.4	10.2	9.4	9.2	9.8
Mean true seed weight (mg)	3.73	2.92	4.66	3.8	3.46
Standard Deviation	±1.3	±1.13	±1.10	±1.14	±0.87
Co-efficient of variation	±34.9	±38.7	±23.3	±30.1	±24.9
Mean proportion of seed as true seed	36.5	33.6	41.9	32.3	30.1
Standard Deviation	±13.45	±11.65	±8.13	±9.08	±7.49

-38-

The mean true seed weights of Lots 1, 4 & 5 were similar, while Lots 2 and 3 were lower and higher than the other lots. The mean proportion of the total seed weight as true seed weight was similar for all the seed lots except Lot 3 which had a greater proportion than the others. The moisture contents of the five seed lots were also similar.

The relationships between seeds and true seed weights are presented in Figures 1-5. Integers plotted on the figures indicate the number of seeds with the same seed and true seed weights.

The relationships revealed highly significant correlations (P = 0.001) with Lots 2, 3 & 4, a good correlation with Lot 5 (P = 0.01) but a non-significant correlation with Lot 1.

The slopes of the regressions are slightly greater than 1 for Lots 2, 3 & 4 but much less than 1 for Lots 1 & 5. There would appear to be an association between the correlation coefficient and the slopes of the regressions. However when the actual true seed weights and mean proportions of the seed weight as true seed weight are compared Lots 1 and 5 have the largest and smallest standard deviations in both cases (Table 1).

The coefficients of variation for the true seed weights of all 5 lots are all larger than the respective coefficients for the total seed weights indicating that there is relatively more variation in the true seed weights.

3.2.2 The Germination Experiments

The germination patterns for the six germination experiments are presented in Figures 6-11. The final germination percentages and mean

-39-