Chapter 1 General introduction
1.1 Flowering and fruitset – significance to Australian viticulture
Flowering and fruitset are critical stages in grapevine development. The processes and sequence of events that occur during flower development and flowering are extremely sensitive, and disruption by endogenous or exogenous factors can be detrimental to their success. The success or failure of flowering and fruitset affects the quantity of grape production, which from an agronomic perspective, is detrimental to profitability.
Recent Australian research has estimated that the number of bunches per vine provides the greatest source of annual yield variation at around 60%. This is followed by the number of berries per bunch (30%), which is determined by fruitset, and berry weight (approximately 10%) (Martin et al. 2003). In the 1950s and 1960s considerable progress was gained in understanding bud fruitfulness of Sultana which was, at the time, the predominant variety grown in Australia. However, the study of fruitset was largely neglected, perhaps because Sultana is grown in hot regions where poor fruitset is not considered to significantly affect yield (May 2004).
The earliest review of flowering and fruitset in Australia provided recommendations to reduce the incidence of poor fruitset (De Castella 1933). However, that information was based on international experiences and therefore, its applicability to Australian viticulture is uncertain. Later, the results of studies conducted in Australia led to the conclusion that there were few available treatments to improve set and those that were available often had undesirable side effects (Coombe1959). In his review of flowering and fruitset, May (2004) found that the earlier findings of Coombe (1959) were still valid. May (2004) also concluded that low and inconsistent-yielding varieties have a significant economic impact on the Australian viticultural industry, and that as an industry we are limited in the amount of control that we have because of the gaps in our knowledge.
The expansion of Australian viticulture into cool climate regions of Australia, the production of a wider range of varieties, and the publicised deficit of relevant and recent research has produced a demand for research focussed on grapevine flowering and fruitset.
1.2 The relationship between molybdenum and Merlot
In Australia, own-rooted Merlot vines have a reputation as being difficult to establish in the first few years. Merlot is also commonly affected by a number of vegetative symptoms which collectively have become known as the ‘Merlot problem’ (Robinson & Burne 2000). In a case study in Renmark, South Australia, Robinson and Burne (2000) associated the symptoms of the Merlot problem on one-year-old vines with excessively high nitrogen values in the petioles, a phenomenon that can be caused by molybdenum (Mo) deficiency. Based on previous work on brassicas, sodium molybdate was applied to the affected vines at a rate of 1 kg per 1000L and, over a period of four weeks, vine growth improved and nitrogen levels in the petioles declined.
Once established, Merlot typically produces low and inconsistent yields. In 2002, a preliminary report was released which cited Mo-deficiency as the possible cause of poor yield of Merlot (Williams & Bartlett 2002). In those experiments, vines sprayed with sodium molybdate in springtime had improved yield by up to 750%, a function of an improvement in the weight of coloured berries per bunch.
In 2003, Gridley conducted further experiments with molybdenum and Merlot grown on various rootstocks to evaluate those interactions and to validate the earlier findings by Williams and Bartlett (2002) and their ongoing research. Gridley (2003) attempted to estimate fruitset; however, perhaps due to methodological problems, found no significant differences between treatments. Gridley agreed with Williams and Bartlett (2002) and concluded that Mo-treatment affected yield via its effect on berry weight.
Chapter 2 Literature review
Grapevines are perennial plants that when grown under temperate conditions produce a crop that matures in the autumn of each year. The yield of grapevines is determined over two seasons and thus the potential for influencing yield spans a period of approximately 18 months. In the first season inflorescence primordia are initiated in the bud, however it is not until the following spring that flower initials are formed and the potential for fruit is realised.
2.1 Components of yield
2.1.1 Season 1
Fruitfulness is defined as the number of bunches per node, and is determined in the season prior to fruit emergence (Dry 2000). Fruitfulness can be measured after bud initiation by observing the number of inflorescence primordia present inside the compound buds. However, it is not until budburst in the following season that the number of shoots per node and hence the number of bunches per shoot is known (Tassie & Freeman 1992).
2.1.2 Season 2
At around budburst in the season of fruit emergence, individual flowers are formed on the inflorescences. However not all of these flowers will becomes berries. The number of berries per bunch is determined after flowering and fruitset, and in the final stages of the season the berries enlarge to realise their final weight.
2.2 Bud initiation
There is no definable endogenous or exogenous ‘trigger’ for floral induction in grapevines, however there are numerous factors that contribute to its initiation (May 2000). Uncommitted primordia (also known as anlagen) can potentially develop into inflorescence primordia, tendril primordia or shoot primordia depending on exogenous factors such as light interception and temperature, and the availability of endogenous growth substances, organic nutrients and water (May 1964).
2.3 Differentiation of inflorescences
The first stage of differentiation is the formation of the bract primordium. This is followed by differentiation of the inner and outer arm. Formation of branch initials on the inner and sometimes the outer arm is the final stage of differentiation and is complete, in the middle part of the shoot, approximately three weeks after flower opening (May 1964; Mullins et al. 1992). From this point there is no further development of the inflorescence primordium until the following spring, however, anecdotal evidence suggests that the extent of branching of anlagen that is reached at this stage correlates well with the number of flowers produced in the following spring (May 1964). While the buds are dormant, morphological development ceases until either exogenous conditions promote growth and budburst occurs, or when endogenous sources of growth inhibition are removed (Swanpoel & Archer 1988; May 2000).
2.4 Flower structure and development
As the buds begin to swell in the spring following bud initiation, growth of the inflorescence primordium recommences. There is further branching, the branches elongate and several flower primordia form on each branch primordium (May 1964; Swanpoel & Archer 1988). Flower development begins with the sepals. A continuous ring of symmetrical tissue forms at the flower primordia to become the calyx. Soon after the appearance of sepals, the petals appear. The flowers of Vitis vinifera are perfect, that is they are hermaphrodite. At around the same time as the corolla is forming stamen primordia begin to develop (Agaoglu 1971; Srinivasan & Mullins 1981). By the time budburst occurs the filament and anther have differentiated and connecting cells have formed at the edges of the petals, fusing them together to form the calyptra or ‘cap’. Following this the pistil, made up of two carpels each containing two ovules, the egg cells and the pollen are formed (Agaoglu 1971; Srinivasan & Mullins 1981). Within ten to fifteen days after the appearance of inflorescences, differentiation of all the essential flower organs is complete, and within twenty days the floral parts are fully developed and visible (Winkler & Shemsettin 1937; Swanpoel & Archer 1988).
Flowers on an inflorescence do not develop simultaneously (Swanpoel & Archer 1988). Flower differentiation tends to decrease towards the terminal point of the inflorescence such that the state of their development depends on their position on the inflorescence (Agaoglu 1971; May 1987).
2.5 Flower number
The number of flowers per branch decreases exponentially from the most proximal to the distal branch, forming a panicle (Pratt 1971). The trend for a decrease in flower number towards the terminus is reflected in the pattern of berry occurrence on the bunch (May 2000). Inflorescence and flower differentiation at budburst is largely controlled by temperature; however there is also evidence to suggest that competition between developing inflorescences and flowers also affects flower number (Dunn & Martin 2000).
Experiments with potted Cabernet Sauvignon vines showed that when the roots of the vines were exposed to temperatures ranging between 11-35°C there were significant differences in the number of berries produced per bunch (Kliewer 1975). Higher root temperatures gave fewer berries per bunch. In similar experiments with potted Cabernet Sauvignon and Merlot vines, Pouget (1981, cited in Dunn & Martin 2000) found that holding the vines at two different temperatures during flower differentiation produced inflorescences with significantly different numbers of flowers. Cabernet vines held at 12°C produced 130% more flowers than those held at 25°C and Merlot produced 29% more flowers at the lower temperature. Ezzili (1993) confirmed this phenomenon in two other vinifera varieties but also found that vines exposed to the lower temperature during flower differentiation produced non-functional flowers.
Dunn and Martin (2000) attempted to simulate the potted vine experiments in the field by pruning at different times to affect the time of budburst thereby exposing developing inflorescences to different temperatures. With observations of individual nodes, it was found that the mean flower number per inflorescence decreased significantly as the soil and air temperatures increased (relative to the time of budburst), to the magnitude of 24 flowers per degree Celsius. They believe that this is consistent with Pouget’s 43 flowers per degree Celsius achieved in the growth cabinet.
This then poses a question: is it time or temperature that influences flower number? Is it possible that earlier bursting buds are preconditioned to have a higher flower number or is it that there is competition for assimilates between early and late bursting buds (Dunn & Martin 2000), or is it, for that matter, the interaction between the growing shoot and the differentiating inflorescence under warm and cool conditions (Pouget 1981, cited in Dunn and Martin 2000).
May (1987) hypothesised that under higher temperatures early flowers are ‘cytokinin enhanced’ and that they dominate and inhibit later flower formation. This may also explain differences in flower number depending on the number and position of inflorescence(s) on a shoot. Shoots with multiple inflorescences tend to have higher flower numbers than single- inflorescence shoots. On two-inflorescence shoots, proximal inflorescences tend to have more flowers than distal inflorescences (May & Cellier 1973; Dunn & Martin 2000).
2.6 Flower opening
Flowering, or anthesis, occurs in response to temperature-stimulated changes in cell turgor of the interlocking cells of the calyptra (Pratt 1971). The calyptra separates at its base, the petals curl up and the cap is shed to release the stamens and expose the stigma (Swanpoel & Archer 1988). This is also known as ‘capfall’.
2.6.1 Timing of flower opening
The timing of flowering occurs at a relatively constant stage of vine development. It is suggested that there is a synchrony between flowering and the formation of leaves on a shoot, observed as a relatively constant number of visible internodes on a shoot at flowering (17-19 internodes, longer than 5 mm) (Coombe 1973; Pratt & Coombe 1978). However, the timing of individual flower opening is dependent on factors that affect their development such as temperature and position on a vine.
Buttrose and Hale (1973) suggested a heat summation of 350 degree days (base 10°C) for flowering to occur. This enables the flower buds to reach a stage of development necessary for opening. The opening of flowers within an inflorescence of V. vinifera follows a period of 24 hours, during which Staudt (1999) observed two distinct times of opening occurring early in the morning and in the evening. Earlier studies by Sharples et al. (1956) and Jawanda et al. (1965) observed flower opening occurring only in the mornings, suggesting that this may vary from variety to variety. The rate of flower opening relies on ambient air temperatures with earlier flowers on an inflorescence tending to open more slowly than those opening later (Jawanda et al. 1965). As flowering proceeds, the highest number of flowers to open usually occurs on day five or six under favourable weather conditions (Staudt 1999).
As mentioned earlier, flowers within an inflorescence do not all develop at the same time and this affects the timing of flower opening. There is a progression of flower opening from the base to the tip of an inflorescence and, flowers on the inner arm of the inflorescence tend to open earlier than those on an outer arm. Flowers on a distal inflorescence of a shoot with two inflorescences tend to open earlier than those on the proximal inflorescence. This provides further evidence to suggest that these flowers are formed earlier than those on proximal inflorescences and are in competition with the later developing proximal inflorescence (Agaoglu 1971; May 2000).
2.7 Cleistogamy
The flowers of some V. vinifera varieties are cleistogamous, that is they have the ability to pollinate prior to flowering. In 1981, Staudt (cited in Staudt 1999) determined that the rate of pollen tube growth was closely correlated with temperature and that by measuring the length of the longest pollen tube the approximate time of pollen germination could be calculated. Applying these measurements to newly-opened flowers of Pinot Noir and Mueller-Thurgau it was found that the anthers of these flowers dehisced and delivered pollen to the stigmas prior to capfall (Staudt 1999). In some varieties cleistogamy may lead to poor fruitset, a problem that may be due to self-incompatibility (Tkačenko 1955).
2.8 Stigma receptivity
The stigmas of grape flowers are receptive to pollen germination when a secretion is released from the stigma giving it a shiny appearance. When the stigma is no longer receptive, it appears dull, dry and brownish black (Randhawa & Negi 1965). Stigma receptivity begins one day prior to anthesis and the stigma remains receptive until one day after, however, receptivity given by highest percent fruitset, has been observed to be greatest on the day of flower opening (Jawanda et al. 1965; Randhawa & Negi 1965).
2.9 Pollination
The most common form of pollination in grapevines is self-pollination (autogamy) however cross-pollination (allogamy), via pollination by insects (entomophily) or wind pollination (anemophily) is also known to occur (Santa Maria et al. 1994).
2.9.1 Insect pollination
The nectaries of grape flowers (perhaps incorrectly named) do not actually secrete nectar but release perfume during flowering (Jawanda et al. 1965; Brantjes 1978). It is thought that this is the reason for sporadic visits by pollinating insects (Brantjes 1978). In an experiment where bees were trapped in the vicinity of flowering vines they were found to be carrying mostly grapevine pollen (Vorwohl 1977 cited in Santa Maria et al. 1994). In experiments with Cardinal grapes, Sharples et al. (1965) found that when insect pollinators were excluded during flowering the mean number of seeds per berry decreased from 1.81 to 1.65. When insect pollinators were allowed to visit the inflorescences during flowering there was a significant decrease in the number of seedless and shot berries present in the bunches, however the authors conceded that this was not enough to be of practical significance.
2.9.2 Wind pollination
Though several experiments have measured the amount of pollen deposited at different distances from vines and the actual distances travelled by pollen, the importance of wind as a transporter of pollen is uncertain. Grapevine pollen can be dispersed over long distances (Panigai & Moncomble 1992, cited in May 2004) suggesting its importance in cross- pollination of some varieties.
2.9.3 Self-pollination and cross-pollination
Many grape varieties self-pollinate successfully and, in a large genetic study, it was found that only 1-2% of seeds are formed through fertilisation with pollen from other sources (Antcliff 1980). When studying pollination, inflorescences are often enclosed inside bags during flowering to exclude pollen from other sources. In self-pollinating varieties, when the bags are removed, flowering and fruitset appear to occur normally (Da Costa e Sousa 1942; Jawanda et al. 1965; Staudt & Kassrawi 1973; May 2000). However, for some varieties inflorescences that are left to pollinate naturally tend to have improved fruitset compared to those that are enclosed inside bags, and a few varieties set no fruit at all when all other sources of pollen are excluded. For these varieties pollen from other sources is necessary for fruitset to occur (Uppal et al. 1975).
Cross-pollination has been found to increase yield by affecting both berry weight and the number of berries per bunch in numerous varieties that are not grown in Australia (Darnay 1954; Karadži & Razumihina 1963; Uppal & Mukherjee 1968; Samaan et al.1981). Darnay (1954) also found that six out of nine pollinator varieties gave differences in bunch size, berry shape, colour, time of ripening and flavour of the berries of Muscat Hamburg grown in Russia while Samaan et al. (1981) found no differences in the composition of berries of cross- pollinated White Roumi in Egypt.
The mode of pollination that gives the greatest yield differs amongst grape varieties. In modern Australian viticulture, varieties that are susceptible to poor fruitset may not have been as conspicuous in traditional vineyards where many different varieties were grown in close proximity to one another.
2.10 Pollen
Pollen grains of V. vinifera are triangular in shape in polar view with radial symmetry, the surfaces are slightly concave, swelling to appear hexagonal in polar view when fully hydrated. The polar axes measure approximately 28 µm (Santa Maria et al. 1994).
2.10.1 Pollen tube growth
Once the pollen grain lands on the stigma there is an initial lag phase after which pollen tube growth may occur (Staudt 1981, cited in Staudt 1999). However there are several factors that may limit the ability for pollen tube germination and growth.
In order for pollen tube growth to occur several conditions must be satisfied. Apart from water, oxygen and a suitable osmotic environment, an adequate supply of nutrients must also be available to the growing pollen tube, supplied from the pollen grain and the surrounding transmitting tissue (Cresti et al. 1975, cited in Cresti and Ciampolini 1999). In vitro studies of a wide range of pollen types have shown that both boron and calcium are essential for pollen tube growth, however in the absence of other ions such as potassium, sodium and magnesium, growth will also be retarded (Dvornic et al. 1970; Brewbaker & Kwack 1963).
Temperature has also been found to limit pollen tube growth both in vitro and in vivo. Winkler (1926) found that pollen growth on a sucrose medium was greatest above 24°C. When excised inflorescences of V. rupestris were grown under controlled conditions it was found that at 10°C pollen germination was delayed and few pollen grains had grown to a satisfactory length. After 15 days under these conditions the stigmas, styles and transmitting tissues were still functional but the ovules showed signs of degeneration and were no longer receptive within 6 days. At 15°C pollen germination was also reduced but some pollen tubes reached the nucellus only after 48 hours. Temperatures between 25 and 28°C provided optimal pollen tube growth with fertilisation occurring after 12 hours (Staudt 1982). Ebadi (1996) also found that exposing flowers of Shiraz and Chardonnay to day/night regimes of 12°/9°C two days before flowering and on the day of flowering significantly decreased the number of pollen tubes reaching the ovules compared to the control that was held at 25°/20°C and that there were differences in the sensitivity of the two varieties.
Using different methods of controlling the shoot vigour of potted vines it was found that competition between the developing transmitting tissue and vigorous shoot growth reduces pollen tube growth in the tetraploid grape cultivar Pione (a hybrid of V. vinifera and V. labrusca). This is thought to occur as a result of changes in levels of endogenous gibberellins, cytokinins and abscisic acid. In lower vigour vines it was found that development of the transmitting tissue was promoted, allowing more pollen tubes to penetrate into the ovary (Okamoto et al. 2001). Similarly the application of the growth retardant chlormequat inhibits shoot growth and organic nutrients are thereby diverted to the developing ovaries (Coombe 1970; Skene 1969, cited in Coombe 1973).
2.10.2 Pollen tube signalling
For fertilisation to occur the pollen tube must grow from the stigma, through the style and central part of the ovary and enter the ovule via the micropyle. Pollen tubes rely on signals from their environment to navigate to their ultimate destination, the egg cell. Under normal conditions the pollen tube grows unidirectionally along the transmitting tract to the ovary. However, once in close proximity to a healthy ovule, it diverges (frequently at 90˚) towards the micropyle (Russell 1996). The pollen tube receives successive directional controls from lipids on the stigma (Wolters-Arts et al. 1998 and Lush et al. 1998 cited in Higashiyama et al. 2001), arabinogalactan proteins in the style (Cheung et al. 1995; Wu et al. 1995; Wu et al. 2000, cited in Higashiyama et al. 2001) and adhesive properties of the transmitting tissue (Wilhelmi & Preuss 1996, cited in Higashiyama et al. 2001). While there remains some conjecture about the exact sources of signalling and their relative importance it is clear that complex mechanisms are responsible for signalling the pollen tube at different points along its route.
Higashiyama et al. (2001) showed that in close proximity (150 µm) to the ovule pollen tubes respond to true chemotropic signals originating from at least one of the two synergid cells based on its ‘…location, appearance and histochemical properties’. In Lilium longiflorum the ubiquitous signalling molecule nitric oxide (NO) has been shown to play a role in the re- orientation response of the pollen tube (Feijó et al. 2004), however the exact directional cues are unknown.
Several studies have successfully shown that in grapes in the absence of the embryo sac, and hence the synergid cells, fertilisation does not occur and fruitset is poor (Fougere-Rifot et al. 1993; Ebadi 1996). Ebadi (1996) found that when Chardonnay flowers were exposed to cold temperatures prior to opening the incidence of ovule abnormalities increased. The predominant defect of the ovules in that study was the absence of the embryo sac.
In some species it has been suggested that the number of pollen tubes in the ovary is limited to the number of receptive ovules (Ebadi 1996; Huelskamp et al. 1995, cited in Russell 1996). It is also suggested that the attraction frequency [from the ovule] depends on the number of pollen tubes (Higashiyama et al. 2001).
Signalling also occurs to inhibit pollen tube growth; for example, once an ovule has been fertilised (Higashiyama et al. 2001). However, there may be variability among species as to the sensitivity of this signalling. In some varieties of V. vinifera it is not uncommon to see multiple pollen tubes entering the same micropyle (Ebadi 1996).
2.11 Fertilisation
Once inside the embryo sac, one of the two haploid nuclei of the pollen tube fuses with the haploid egg nucleus while the other fuses with the diploid polar nucleus to form the triploid primary endosperm nucleus. After successful fertilisation the ovule enlarges and the endosperm begins to develop (May 2004).
The number and degree of development of seeds influences the growth of the pericarp via the production of essential hormones by the seed (Olmo 1946; Coombe 1959). Environmental conditions prior to and at flowering affect both seed developmental characteristics and seed number and hence influence final berry weight (Ebadi 1996).
2.12 Bunch development
In seeded varieties of V. vinifera, berries develop in a number of ways: a) by vegetative parthenocarpy - fruit develops from ovaries with defective embryo sacs such that they develop into seedless berries in the absence of pollination; b) stimulative parthenocarpy - when stimulus in the pistil from pollen tubes or parasitic fungi prompts fruit growth (Stout 1936); c) stenospermocarpy - where the seed aborts after pollination and fertilisation (May 2004); and d) by normal pollination and fertilisation that leads to endosperm and normal fruit development. Both stimulative parthenocarpy and stenospermocarpy can lead to ovule enlargement and partial seed development (Sedgley & Griffin 1989). To differentiate between the two types actual fertilisation must be known to have occurred (Stout 1936).
Once flowering is complete flowers on an inflorescence have several fates. For most grape varieties it is normal for a high proportion of flowers to abscise from the rachis immediately after flowering. Under normal conditions a smaller proportion of flowers will enlarge to form normal functional berries and some will remain on the rachis arrested at different stages during the transition from ovary to functional berry.
2.12.1 Shot berries
May (2004) describes ‘shot’ berries as those that form after pollination but without fertilisation suggesting that they form via stimulative parthenocarpy. However vegetative parthenocarpy cannot be ruled out nor can stenospermocarpy because of the difficulty of establishing the presence or absence of fertilisation. The mechanism for shot berry development may be via a combination of pathways and there may be variation across varieties or under differing endogenous and exogenous conditions. This is yet to be clarified.
The shape of shot berries is described as reminiscent of an ovary, that is, with the style intact, or they may be round in shape (May 2004). Other authors include small (2-3 mm, Cholet et al. 2002), and mid-sized (4-6 mm, Colin et al 2002) (less than 8 mm, Sharma et al. 1995) green, seedless berries in their definition of shot berries. Notwithstanding, some ovule development takes place and the size of shot berries is small relative to other berries on a bunch. Shot berries have no colour, flavour or sugar development.
Shot berries do not fit the definition of a berry, that is, ‘pericarp consisting of skin enclosing a fleshy or pulpy mass with one or more seeds’, nor do they develop into ‘fruit’ whose primary function is to disperse seeds (Sedgley & Griffin 1989). The origin of the term ‘shot’ berry is unknown but it is clear that it does not accurately discriminate between the organs described in this section and other berry variants described below. It has been proposed that ‘live green ovary’ (LGO) be used in place of ‘shot berry’ to distinguish them from true berries (pers. comm. P. Dry 2003).
LGOs do not contribute to yield and should not be included when quantifying fruitset. Anecdotally LGOs may be detrimental to winemaking, possibly due to phenolic extraction, however this is unlikely to be significant.
2.12.2 Chicken berries
‘Chicken’ berries are small (4-7 mm, Cholet et al. 2002) berries that are ripe at harvest (May 2004) and are either seedless or contain seed traces (May 2004). Anecdotally, chicken berries are reported to be beneficial to winemaking because of differences in their composition related to their small size and seedlessness. An example of this is the Mendoza clone of Chardonnay and other varieties susceptible to producing a high proportion of chicken berries. A small, unreplicated study by Cahurel (1999) showed that chicken berries from Gamay Noir were marginally higher in total soluble solids, pH and had a higher concentration of anthocyanins, however the total acidity and phenolic concentration was lower than for the normal berries.
2.12.3 Hen berries
‘Hen’ berries are normal-sized (9-17 mm, Cholet et al. 2002) berries. For most commercial wine grape varieties grown in Australia, normal berries contain one to four seeds. Normal- sized, seeded hen berries usually make up the greatest proportion of berries on a bunch.
2.13 Fruitset
Fruitset is the transition from the ovary to the young fruit (Coombe 1962). In a qualitative sense it is the retention of the ovary on the peduncle after flowering (May 2004). Quantitatively, fruitset describes the proportion of flowers on an inflorescence that develop into berries (May 2004) and, as previously discussed, should not include LGOs.
According to Bessis (1993) normal fruitset in grapevines is around 50% while poor fruitset occurs when fruitset is less than 30%. In much of the literature it is difficult to interpret fruitset data because they are provided solely in terms of the number of berries per bunch, neglecting to acknowledge differences in flower number per inflorescence (May 2004). A measure of ‘normal’ fruitset across different varieties is difficult to establish. As discussed in section 2.5 the number of flowers on inflorescences varies markedly and in experimental situations precautions must be taken to minimise this variability. One way to do this is to select inflorescences from the same position on a shoot (e.g. select all proximal inflorescences) with the same number of inflorescences on it (e.g. only choose from shoots that have two inflorescences).
2.14 Fruitset disorders
2.14.1 Coulure
‘Coulure’ is characterised by infertility of grapevine flowers and their subsequent shedding from the inflorescence resulting in poor fruitset. The term is derived from the French ‘elles coulent’ or ‘they run or flow’ (Chancrin 1908). Excessive flower shedding, such that the inflorescence becomes tendril-like, is sometimes referred to as ‘filage’, from the French ‘becoming a thread’ (May 2004).
2.14.2 Millerandage
‘Millerandage’ is a condition described by the presence of berries arrested at different stages of development and of different sizes on the same bunch. Bunches affected by millerandage tend to be loose at maturity and contain a mixture of normal berries and seedless berries, some of which fail to ripen (Royal Commission on Vegetable Products 1891; Chancrin 1908; Sharma et al. 1995; Colin et al. 2002). Undeveloped berries or bunches with these berries may be described using the term ‘millerand’ berries or bunches (Chancrin 1908).
Berries on bunches affected by millerandage may arise from one or a combination of vegetative or stimulative parthenocarpy, or stenospermocarpy. In the case of vegetative parthenocarpy occurring on bunches with millerandage, the parthenocarpy would be facultative if other ovaries and ovules pollinate and fertilise to form normal berries (Stout 1936).
In Australia the term ‘hen and chicken’ is commonly used interchangeably with millerandage; however its lax usage has perhaps led to its misuse. Firstly, millerandage encompasses shot berries (or LGOs) and chicken berries, however, hen and chicken infers the presence of a high proportion, or at least presence of, seedless chicken berries on a bunch. Therefore, bunches displaying hen and chicken are affected by millerandage. However, bunches affected by millerandage do not necessarily display hen and chicken. This highlights the probable misuse of millerandage and ‘hen and chicken’ as synonyms. Secondly, there may be confusion between poor fruitset and millerandage. In a survey of final-year viticulture students who were presented with two photographs of similar-sized bunches, one affected by poor fruitset and the other by millerandage, most students confused the two disorders (Longbottom 2006, unpublished data). Based on this, viticulturally-inexperienced researchers may also have had difficulty discriminating between the two disorders by visual assessment alone.
The earliest reports attribute millerandage to an excess or lack of vine vigour, faulty plant nutrition, cold, wet weather (Chancrin 1908) and generally ‘unfavourable’ weather which has led to the development of late flowers on an inflorescence. De Malafosse (cited in Royal Commission on Vegetable Products 1891) suggested that the late flowers were thought to be incapable of setting proper fruit but, as a result of other flowers not setting under the unfavourable conditions, the late flowers were able to stay on the bunch at an arrested stage of development. Other authors have reported imperfect pollination as the cause of millerandage (Chancrin 1908; Bioletti 1921, cited in Stout 1936). A closer investigation of yields produced by different modes of pollination, such as cross- or supplementary self-pollination, in varieties susceptible to coulure and millerandage (see section 2.14) under Australian conditions is warranted. In recent years millerandage has been associated with molybdenum deficiency of Merlot. This is discussed in section 2.16.
Polyamines and millerandage Polyamines (PA) have been implicated in arrested growth of berries in bunches affected with millerandage. In normal bunches polyamines accumulate at the beginning of flowering followed by a gradual decrease up to harvest (Colin 2000; Colin et al. 2002). However, in Merlot berries affected by arrested development the composition and concentration of PAs remain similar to those at that stage of development. This was related to the activity of polyamine oxidase (PAO), the enzyme responsible for the catabolism of spermidine and spermine to diaminopropane and hydrogen peroxide (Colin et al. 2002). A study by Cholet et al. (2002) suggested that the production of high levels of PAs in shot berries was a wound response, the increase in PA concentration increasing in proportion to the extent of the trauma. They further postulated that the scar tissue formed at the wound site might inhibit pollen tube growth and fertilisation (Cholet et al. 2002). This suggests that the formation of shot berries on Merlot affected with millerandage is via stimulative parthenocarpy. However, the relative proportions of ovaries affected by the ‘trauma’ and their contribution to the production of LGOs is unknown.
2.15 Other factors that affect fruitset
Factors that can affect fruitset can be grouped into: a) those that are inherently caused by genetic factors, that is, abnormal development of the flower parts, pollen, embryo sac, embryo and endosperm, and failure of pollination or fertilisation to take place (Crane & Lawrence 1929); and b) sterility that occurs as a result of unfavourable conditions during the development of the flower and flower parts or during flowering (Dorsey 1914). In addition to these forms of sterility, incompatibility (that is the failure of pollen tubes to reach and fertilise the ovule) also prevents fruitset from occurring (Crane & Lawrence 1929). Extensive selection and propagation has for the most part eradicated inherent characteristics of certain varieties that negatively affect yield, and it is now widely regarded that environmental conditions are the largest contributing factor to poor fruitset (Staudt 1982; May 2004).
2.15.1 Environmental factors
Environmental conditions that can cause sterility include rain, shading and cool temperatures during flowering while endogenously, variations in adequate levels of growth hormones and nutrients also affect the ability of flowers to set fruit. While it is important to mention each individually the interactions between environmental factors and the influence of the environment on endogenous factors should not be ignored.
Rain Heavy rain at flowering affects fruitset by interrupting the pollination process. In a study of two grape varieties at peak flowering it was observed that 4.9 mm of rain removed between 20-30% of pollen grains from the stigma and after a heavy downpour of 27 mm received over a period of 8.5 hours, only 3% of the pollen remained (Tkačenko 1960).
Solar radiation Light exposure has been shown to affect fruitset of grapevines both in the field and under controlled environment conditions (Coombe 1973; Nuno 1993, cited in May 2004; Ollat 1993 cited in Ebadi 1996). However, it has been suggested that the effect of diminished solar radiation on fruitset is coupled with the effects of low temperature and reduced whole-vine photosynthetic potential (May 2004).
Temperature It is well recognised that low temperatures just before and at the early stages of flowering contribute to poor set in some grape varieties. In potted vine experiments Ebadi (1996) showed that fruitset of Chardonnay decreased by around 30% when vines were exposed to 12°/9°C day/night temperature compared with those held at 17°/14°C. This was a result of detrimental effects on both the pollen and ovules of these vines. Pollen tube growth was slowed by the cool temperature regime resulting in a delay in the pollen tube reaching the ovule. Applying supplementary pollen to the inflorescences had significant positive effects on fruitset of these vines. The effect of cold treatment on the ovules has been previously discussed (section 2.10.2). After further observation of cell numbers in the ovaries and resultant berries of the low temperature treated vines it was found that the effect of low temperature during flowering is greatest via its effects on ovule/seed development and subsequent fruitset and berry growth rather than by the influence on pollen germination and pollen tube growth (Ebadi1996). It should not be discounted that, in addition to causing damage to the ovules by imposing low temperatures on the vines, there may be a predominant interaction, that is an inability of the vines to mobilise nutrients required for complete ovule development, pollen tube growth and cell division of the newly fertilised ovule.
2.15.2 Growth substances
Auxins Auxins are involved in numerous physiological process in plants-most relevant in this review is seed-produced auxin associated with cell division and enlargement in the developing ovary/berry (May 2004).
Cytokinins The function of cytokinins is not well understood; however it has been suggested that they play an important role in flower formation. When applied to potted Muscat of Alexandria vines, tendrils were converted to inflorescences (Srinivasan & Mullins 1980). There also appears to be a relationship with the determination of flower number under different temperature regimes with cytokinin content in xylem sap (Skene & Kerridge 1967). Cytokinins have also been reported to improve fruitset by preventing ovary abscission and are reported to influence berry size (May 2004).
Gibberellins Gibberellins (GA) are produced in the seeds after flowering (May 2004), however when applied exogenously have been shown to promote anlagen formation and growth of tendrils while inhibiting the production of inflorescences (Srinivasan & Mullins 1980). GA has also been used to modify yield by decreasing fruitset and increasing berry size; however there are differences in the sensitivity amongst varieties. High GA has also been associated with reduced pollen viability (May 2004).
Growth inhibitors The growth inhibitor chlormequat has been shown to promote inflorescence formation in otherwise unfavourable conditions such as low temperature. In the same study, chlormequat was found to have an opposite effect on the formation of anlagen and tendril growth by inhibiting their production, probably an indirect effect of its function as an inhibitor of GA synthesis (Srinivasan & Mullins 1980).
2.15.3 Carbohydrates and organic nutrients
Organic nutrients and reserve materials play important roles in determining yield because of their influence at numerous stages of yield determination. The amount of accumulated carbohydrate reserves has been suggested to influence flower differentiation and fruitset of Semillon vines (Smith & Holzapfel 2003) while low levels of reserve nitrogen is detrimental to fruitset without affecting flower number or the number of seeds per berry of Grenache (Duchêne et al. 2001). May (2004) suggested that varieties differ in their dependence on reserve materials, citing studies by Zapata et al. (1999; and 2001, cited in May 2004) in which it was found that Merlot relies more heavily on reserves during the flowering period than Pinot Noir, a characteristic that may partly explain the susceptibility of Merlot to coulure.
The requirement for sufficient nutrient supply at flowering is also essential for adequate fruitset to occur. In several experiments it has been shown that altering the supply of nutrients within the vine by shoot girdling and removing leaves, flowers and shoot tips, fruitset can be affected (Coombe 1962; Caspari et al. 1998). Removing sources of competition between vine parts tended to improve fruitset while removing the source of nutrients, that is the leaves, decreased fruitset. Similarly, when the effects of different pruning regimes on fruitset were examined it was found that it was the influence of leaf number at or near flowering time that affected the ability of the vines to supply the developing flowers with the required balance of nutrients. More severely pruned vines produced fewer leaves per vine at flowering and decreased the mean berry weight, a function of both decrease in the germinability of the pollen and the success of fertilisation (Winkler 1926).
Deficiencies in organic nutrients can be induced by adverse soil conditions such as high or low soil pH (Vunkova-Radeva et al. 1988), waterlogging and compaction, interactions with other nutrients, differences in plant ability to retrieve and mobilise nutrients and via the indirect effect of temperature, light, disease etc. on plant metabolism (Smith 1986). Deficiencies in micronutrients (N, B, Zn, Mo) have been shown to affect yield in some varieties, however the components of yield that are affected are unclear in most cases. There is, however, evidence to suggest that zinc influences yield by influencing seed formation (Hewitt & Jacob 1954).
2.16 Molybdenum as a potential remedial treatment for low-yielding varieties
Apart from its function in nitrate reductase enzymes there is little known about the function of molybdenum (Mo) in higher plants (Agarwala et al. 1979). Molybdenum-deficient plants however, have been found to display symptoms such as the ‘whiptail’ disorder (Fido et al. 1977), decreased biomass (Chatterjee & Nautiyal 2001), small and abnormally formed flowers, delay in or suppression of flowering, low pollen grain numbers and viability of pollen, decreased enzymatic activity in the pollen (Agarwala et al. 1979), small and deformed seeds (Vunkova-Radeva et al. 1988), decreased yield (Yu et al. 1999) and increased susceptibility to low temperature and waterlogging stress (Vunkova-Radeva et al. 1988). It has been suggested that low temperature stress may be due to the breakdown of Mo- containing enzymes (Vunkova-Radeva et al. 1988).
In grapevines there have been numerous attempts to show the yield benefits of molybdenum application, in different forms, to Mo-deficient plants without any reference to the underlying mechanism with which it acts. In the early experiments examining the effects of molybdenum on grapevines that reported positive yield effects, it is unclear which components of yield were affected and the repeatability of results is poor (Hatle & Miculka 1971; Veliksar 1977; Misa 1977; Strakhov & Chazova 1981). More recently Ma et al. (1992) found that spraying sodium molybdate on Kyoho vines twice before flowering tended to increase fruitset and pollen germination in vitro, however these results were not significant.
2.16.1 Effects of molybdenum deficiency on yield of Merlot
Molybdenum was first reported to increase yield of Merlot vines when two spring applications of sodium molybdate (Na2Mo4.2H2O) were applied to the foliage of vines at McLaren Vale, South Australia (pers. comm., C. Williams 2003). The results from those experiments were published in 2004 (Williams et al.). However, the improvements in yield and the different components of yield were not consistent across all experimental sites in all seasons. Gridley (2003) examined the effects of molybdenum sprays on own-rooted Merlot compared to Merlot grown on two different rootstocks. In both studies the greatest increase in yield was on own-rooted Merlot vines that were deficient in molybdenum (0.05-0.09mg/kg measured in the petioles at 80% flowering).
Effects of molybdenum on bunch number per vine Both Gridley (2003) and Williams et al. (2004) reported increases in bunch number per vine with molybdenum treatment, suggesting that application of molybdenum to Mo-deficient vines prevented the loss of inflorescences from the vines. Gridley (2003) suggested that differences in bunch numbers between Mo-treated and untreated vines observed by Williams et al. (2004) may have been due to bunch loss after budburst, potentially as a result of early bunch stem necrosis (EBSN) or filage associated with Mo-deficient conditions.
Effects of molybdenum on berry weight Bunch weight has been reported as the most significant contributor to improved yield in response to Mo-treatment on Mo-deficient Merlot vines, a function of increased berry weight brought about by a decrease in the proportions of shot and chicken berries relative to hen berries, or put another way, the bunches were less affected by millerandage (Gridley 2003; Williams et al. 2004; Phillips 2004). Coupled with this, berries from Mo-deficient vines tended to have fewer seeds, a function of the high proportion of chicken berries (Williams et al. 2004).
Effects of molybdenum on fruitset of Merlot In 2003 Gridley calculated fruitset by constructing a standard curve to estimate flower numbers. She did this by conducting a regression analysis of flower numbers per inflorescence and the combined length of the rachis and the wing of the inflorescence (r2 = 0.91). In that study it was reported that fruitset was better on bunches from vines treated with molybdenum compared to the untreated controls, however those differences were not statistically significant. There are a number of potential explanations for the lack of significant differences in fruitset found by Gridley (2003). Firstly, those experiments were conducted on two different Merlot clones (2093 and 2613) and it has not reported on which clone the regression analysis was conducted. It is possible that there were clonal differences in flower numbers and that this method of estimating flower numbers was not valid for both clones. It is also conceivable that fruitset was not affected on the clones that were studied. Secondly the variability associated with the estimation of flower number may have negated any potential differences in flower number and resultant fruitset between treatments.
In 2003-04 Phillips (2004) enclosed inflorescences in mesh bags during flowering to catch the flower caps in order to calculate fruitset at the same site used by Gridley (2003). In that study Phillips (2004) reported significant differences in fruitset on Merlot clone 2093 (2613 was not studied), however the increase in fruitset was not a result of more berries per bunch but rather due to fewer flowers per inflorescence. This further highlights the danger of using berry number per bunch as a guide to fruitset and also the importance of standardising the selection of inflorescences to minimise variability in flower number. Notwithstanding the significant effect on fruitset, Phillips concluded that the main contributor to improved yield was berry weight and a reduced severity of millerandage. Close scrutiny of that data revealed that fruitset might in fact have been more important than was reported.
In the case of Williams et al. (2004) it is possible that the extent to which millerandage was responsible for the low yields on Mo-deficient vines, compared to the effect of poor fruitset, was overestimated because only berry weight and number were measured, not fruitset. The bunches from the Mo-deficient vines in that study were clearly looser with fewer berries per bunch than those from the Mo-treated vines (pers. comm. C. Williams 2003; Williams et al. 2004). However, the number of berries per bunch increased in response to molybdenum treatment in only one out of nine experiments. It is possible that yield was most affected by poor fruitset in the Williams et al. (2004) study, however the lack of measurements performed does not permit any conclusions.
2.16.2 Potential negative effects of applying molybdenum to Merlot
Gridley (2003) found that above a certain level, the concentration of molybdenum in the petioles of sprayed vines was associated with decreased yield on those vines, suggesting a possible toxicity effect. No previous research into the long-term effects of molybdenum application to grapevines is available at this time.
2.16.3 Areas for further investigation
The existing reports suggest that molybdenum may affect the ability for successful fertilisation to take place, thereby affecting fruitset and subsequent berry development. Possible causes of this may be faulty development of the flowers or flower parts, reduced pollen viability, inhibition of pollen tube signalling and induced seed abortion in bunches from Mo-deficient vines. Further research into the mechanism by which molybdenum affects yield of Merlot is justified, with observations of flower structure, pollen viability and pollen tube growth essential. Observations of developing inflorescences and bunches under Mo- deficient conditions is also warranted to elucidate the possible effect of molybdenum in preventing early inflorescence loss that is suggested to affect the number of bunches per vine. The effect of high molybdenum levels on yield and any long-term effects are further areas for research so that any potential negative yield effects are fully explored.
2.17 Conclusion
Grape yield is determined over two seasons, the components of yield and their general timing of determination are well documented. While the factors that affect yield are numerous, there are few that have been described in detail, and often without reference to the mechanism by which they affect yield. There is disagreement in the literature about pollination of grapevines and few, if any, comprehensive reports about the effects of pollination on yield under Australian viticultural conditions. The effects of temperature on flowering and fruitset are well documented, both anecdotally and scientifically, and the mechanism by which temperature affects flower development has been only recently reported. The role of organic nutrients in yield determination received much attention in the 1950s and 1960s in Australia, however focus on micronutrients and the mechanisms involved in flowering and fruitset has been largely overlooked. The mechanisms by which molybdenum affects yield of Merlot must be more closely examined in order that grapegrowers can apply remedial measures confidently and sustainably.
Chapter 3 General materials and methods
3.1 Experiments
Three main experiments were conducted in commercial vineyards in South Australia. The experiments are described in the various chapters as follows:
Table 3.1. List of experiments Site Season Chapters
Molybdenum (Mo) experiments McLaren Vale & Adelaide Hills 2003-04 4, 5, 6, 7, 8 2004-05 2005-06
Pollination experiments McLaren Vale & Adelaide Hills 2003-04 9 2004-05
Mo-pollination experiments McLaren Vale 2005-06 9
3.2 Experimental field sites
The experimental sites were chosen on the basis of previous reports of Mo-deficiency occurring in the Adelaide Hills and at McLaren Vale, and on their climatic differences. At McLaren Vale the experimental site was located in a Hardy Wine Company vineyard on Bayliss Road. The Adelaide Hills experiment was within the Nepenthe Vineyard on Newman Road at Charleston. Molybdenum deficiency in Merlot is suggested to occur in cool, wet seasons (pers. comm., C. Williams 2003; Williams et al. 2004) where they are grown in low pH soils with high concentrations of adsorbing oxides such as Fe, Mn and Al (Kaiser 2005). This gave the Adelaide Hills site a greater probability of developing Mo-deficiency given its cooler climate (relative to McLaren Vale) and typical Adelaide Hills soil type (Table 3.2), that is, low pH and high Fe concentration (pers. comm., M. Leake 2003). Vines at both experimental sites had not previously received any molybdenum treatment. A description of the experimental vines and vineyard layout is provided in Table 3.3.
Table 3.2. Climatic characteristics of and soil types at experimental sites in the Adelaide Hills and at McLaren Vale.
Site MJTa Annual rainfall (mm) Soil description
0-20 cm sandy loam with shale Adelaide Hills 19.1 1029 b & ironstone Nepenthe Viticulture, Charleston 20-60 cm sandy clay loam 60-140 cm orange/red heavy
clay
McLaren Vale 21.7 656b 0-20cm - sandy clay loam Hardy Wine Company, Bayliss Rd. 20-75cm - light medium clay 75-135cm - clay loam a Mean January temperature (MJT) was derived by calculating the average of the mean daily maximum and mean daily minimum temperatures for January sourced from long term meteorological observations at the Australian Bureau of Meteorology (2002). b Dry and Smart (1988)
Table 3.3. Description of experimental vines and vineyard layout in the Adelaide Hills and at McLaren Vale. Site Variety Clone Rootstock Year Irrigation Row x vine Trellis type planted source spacing (m)
Dam and Adelaide Hills Merlot D3V14 own 1999 2.7 x 1.5 VSP a bore water
Treated McLaren Vale Merlot D3V14 own 1999 2.75 x 1.8 VSP waste water a Vertically shoot positioned (VSP)
From hereon, the experimental sites are referred to in the text as McLaren Vale and Adelaide Hills (Hills).
3.3 Weather data
Daily weather data was collected during the period from budburst to flowering in the Hills and at McLaren Vale using on-site weather stations. In the Hills the weather station was a MEA Uni-Data Starlogger 6004C and, at McLaren Vale, a Davis GroWeather station from Cameron Instruments was used.
Day degrees were calculated using the mean daily temperature base 10°C, summated for each day over the period from budburst to 80% flowering.
3.4 Phenology
Phenological development was recorded using the modified E-L system described by Coombe (1995) (see APPENDIX A).
3.5 Molybdenum sprays
Two pre-flowering foliar sprays were applied to the experimental vines using a hand-held pump sprayer (Figure 3.1). The first spray was applied at modified E-L stage (MELS) 12 (shoots 10cm) and the second was applied approximately 1 week later. The molybdenum was applied in the form of sodium molybdate (Na2Mo4.2H2O) at three different rates: 0 g, 0.101 g and 0.202 g per vine (Table 3.4). The spray rates were calculated on a per vine basis with rate 1 being equivalent to the 300g/ha rate used by Gridley (2003). The sodium molybdate was dissolved in deionised water and delivered at a rate of 250 mL per vine.
Figure 3.1. Molybdenum spray being applied to Merlot vines at modified E-L stage 12 at McLaren Vale.
Table 3.4. Molybdenum treatments applied to experimental vines at McLaren Vale and in the Adelaide Hills.
Treatment Sodium molybdate (Na2Mo4.2H2O) (g/vine)
Rate 0 (control) 0 Rate 1 0.101 Rate 2 0.202
3.6 Experimental design
3.6.1 Vine selection
The vines within all the experiments were selected on the basis of vine uniformity. At McLaren Vale the Mo experimental site was selected using an infrared image of plant cell density. This enabled the selection of an area of vines within the same block that were approximately the same size. In the Hills the experimental sites were selected using a visual assessment of the uniformity of vine trunk circumference. Vines were also assessed for uniformity of bud number after pruning each season.
3.6.2 Molybdenum experiments 2003-04
The treatments were randomly allocated to a block of three panels in seven alternate rows using a completely randomised design. This arrangement allowed a buffer row between each treatment row (Figure 3.2). Within each block the allocated spray treatments were applied to the vines in the central panel (4 vines at McLaren Vale and 3 vines in the Hills) plus two vines either side of the panel, the outer vines acting as buffers to the neighbouring vines. All vines in the experimental area were tagged with coloured flagging tape to identify the treatments.
In 2003-04 additional double-pruned (Dry 1987) / molybdenum treatments were imposed on vines within the experimental block. However, due to a high incidence of abnormal flowers produced on these vines, those treatments were abandoned and will not be discussed here.
X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
X X X X X X X X X X X X X X X Row direction X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
Figure 3.2. Experiment layout at McLaren Vale and in the Adelaide Hills in 2003-04. Each X along a row represents 1 panel of vines (4 vines per panel at McLaren Vale and 3 vines per panel in the Hills). Rate 0 (□), rate 1 (■) and rate 2 (■) molybdenum sprays were applied to vines in the central panel in each block plus two vines either side of the central panel leaving unsprayed buffer vines at either end of the block. Alternate rows between the treatments were unsprayed buffer rows. Blocks coloured with ■, ■ and ■ were double-pruned and sprayed with molybdenum in summer.
3.6.3 Molybdenum experiments 2004-05
Hills In 2004-05 the experimental design remained identical to that in 2003-04 without the double- pruned vines (Figure 3.2).
McLaren Vale In 2004-05 the molybdenum experiment was expanded to include additional vines in the same vineyard, neighbouring the original experiment block that had not previously received any molybdenum sprays. The original experimental vines are referred to in the text as the ‘old block’ and those treated for the first time in 2004-05 are referred to as the ‘new block’. The experimental design was identical to that used in 2003-04 (Figure 3.3).
New block Old block
X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X XXX X X X X X X X X X X X X X X X X X X X X X X X X X X X XXX X X X X X X X X X X X X X X X X X X X X X X X X X X X XXX X X X X X X X X X X X X X X X X X X X X X X X X X X XXXXXX X X X X X X X X X XX X X X X X X X X X X X X X XXXXXX X X X X X X X X X XX X X X X X X X X X X X X X XXXXXX X X X X X X X X X XX X X X X X X X X X X X XX X XXX X X X X X X X X X X X X X
X X X X X X X X X X X XX X XXX X X X X X X X X X X X X X
X X X X X X X X X X X XX X XXX X X X X X X X X X X X X X ction X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X w dire
X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Ro X X X X X X X X X X X XXXX XX X X X X X X X X X X X XX X X X X X X X X X X X XXXX XX X X X X X X X X X X X XX X X X X X X X X X X X XXXX XX X X X X X X X X X X X XX X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
Figure 3.3. Experiment layout at McLaren Vale in 2004-05. Each X within a row represents 1 panel of vines (4 vines per panel at McLaren Vale and 3 vines per panel in the Hills). Vines in the old block were sprayed with rate 0 (□), rate 1 (■) and rate 2 (■) molybdenum sprays in 2003-04 and 2004-05 and vines in the new block were sprayed in 2004-05 only. Sprays were applied to the central eight vines in each block leaving two unsprayed buffer vines at either end of the block. Alternate rows between the treatments were unsprayed buffer rows. 3.6.4 Molybdenum experiments 2005-06
In 2005-06 the molybdenum experiments were scaled down and only used for the collection of nutritional information and observation of flower structure. Rates 0 and 1 molybdenum sprays were applied in a completely randomised design to six replicates in a single row of vines. Replicates were single vines of previously untreated vines; the central vine of three sprayed vines such that each replicate was separated by at least one buffer vine.
X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X
Row direction
Figure 3.4. Experimental layout at McLaren Vale in 2005-06. Each X represents three vines. Vines were sprayed with rate 0 (control) (■) and rate 1(■) molybdenum sprays at modified E-L stage 12 (shoots 10 cm) in 2005-06.
3.7 Flower observations
3.7.1 Pollen tube growth
Selection of inflorescences At the commencement of flowering inflorescences were monitored daily to determine the optimum timing for flower selection. This occurred once all replicates had several inflorescences with at least one flower open on branches 2, 3 and 4. Inflorescences were selected such that they all occurred basally on a shoot with two inflorescences and were all at a similar stage of flowering estimated by visually assessing the proportion of opened flowers on the inflorescence. Three inflorescences per replicate were tagged with coloured tape and the flowers isolated for observations of pollen tube growth as described below.
Selection of flowers for observation On the first day all opened flowers on branches 2, 3 and 4 of the tagged inflorescences were identified. Different methods were used in 2003-04 and 2004-05.
In 2003-04 all opened flowers were removed from the inflorescence on day one using jewellers’ forceps. On the following day, all the unopened flowers on branches 2, 3 and 4, were discarded leaving only the newly opened flowers to be harvested 4 days later (Ebadi 1996). Where the number of flowers to be removed was more than 50%, the newly opened flowers were isolated on the branch using loosely tied cotton thread rather than removing the flowers so that potential compensatory effects on the success of pollen tube growth were avoided.
In 2004-05 the method of flower selection was revised to avoid potential compensatory effects between flowers caused by flower removal. On day one the pedicels of all open flowers on branches 2, 3 and 4 were marked with coloured acrylic ink applied using a Rotring technical pen (Gray 2002). The following day the pedicels of newly opened flowers were marked with a different colour ink to identify them as the flowers for observation.
Sample processing Four days after flower isolation the flowers were harvested and placed into labelled vials filled with Carnoy’s fluid. For the molybdenum experiments, sample processing was continued on a sub-sample of 24 flowers (8 flowers from each of the 3 inflorescences from each replicate) randomly selected from the initial sample. In cases where fewer than 24 flowers were available, no fewer than 16 flowers per replicate were taken through the final stages of processing. Hydration and softening of the flowers was continued in the laboratory using the following procedure:
1. Fix flowers in Carnoy’s fluid (absolute alcohol : chloroform : acetic acid, 6:3:1) for a minimum of 2 hours. 2. Hydrate in 70% alcohol for 10 minutes followed by 30% alcohol for 10 minutes, then twice in distilled water for 10 minutes each. 3. Soften samples in 0.8N NaOH at 60oC for approximately 20 minutes. 4. Stain using 0.1% water-soluble aniline blue in 0.1 NK3PO4 for a minimum of 10 minutes.
The standard method to mount pistils is in 80% glycerol and squashed using a cover slip (Ebadi 1996). However, through experimentation it was found that sectioning the pistil longitudinally down its centre then opening out the two halves to expose the inner components of the pistil before mounting in 80% glycerol gave a more precise and complete observation of the stylar transmitting tissue and the ovules. Pollen and pollen tubes were observed at 520 nm using the Zeiss AX10 PHOT microscope with an exciter filter of 395-440 nm.
The numbers of pollen tubes were recorded at different points along the passage towards the ovule, namely: the lower style, which was considered to be level with the top of the ovules and; the lower ovary, which was between the lower style and the bottom of the ovules (Figure 3.5).
A B a
b
Figure 3.5. Typical images of Merlot flowers taken using fluorescence microscopy. A: This image shows the inner regions of two halves of the stigma and style. (a) Fluorescing pollen grains visible along the edge of the stigma and (b) pollen tubes extend down into the style. B: This image shows the inner region of one half of a pistil. The coloured lines indicate different points along the passage of the pollen tubes towards the ovule. Green: stigma/upper style, yellow: lower style, orange: lower ovary. Arrows indicate pollen tubes entering the ovules.
3.7.2 Ovule structure
On the day prior to flower opening, determined by the change in colour of the calyptra from bright green to green-yellow, flowers were collected from the field and transported back to the laboratory in an insulated container. The flowers were dissected and the pistils separated from the pedicel, calyptra and androecium. Each pistil was scored with a scalpel to facilitate infiltration of the fixative. Samples were prepared using the following procedure:
1. Place scored pistils into 25% glutaraldehyde in 0.025M phosphate buffer (pH 6.8-7.2) for 48 hours 2. Dehydrate for 2 hours in each of two changes of: methoxy ethanol, ethanol, propanol and butanol. 3. Infiltrate for 2 hours in a mix of 1:1 glycol methacrylate (GMA) (93mL GMA, 7mL polyethylene glycol 400, 0.6 g benzoyl peroxide) and butanol 4. Infiltrate for 4 days in GMA, replacing the GMA after 2 days. 5. Embed in gelatine capsules (pharmaceutical) filled with GMA. 6. Polymerise for 2 days at 60°C. 7. Section samples using a microtome fitted with a glass knife and float individual sections onto a drop of water on a microscope slide. 8. Dry slides overnight on a heat bed at 60°C. 9. Slides were stained with periodic acid-Schiff’s reagent and toluidine blue (Ebadi 1996) All slides were observed under a Zeiss AX10 PHOT microscope. 3.8 Yield components
3.8.1 Flower number per bunch
One or two days prior to the onset of flowering 5 uniformly-developed, proximal inflorescences on shoots with two inflorescences were selected from one vine from each replicate. Each inflorescence was enclosed inside a numbered mesh bag and secured with a zip tie. After flowering, when the berries had reached MELS 29 (berries peppercorn size) the bags were removed and the bunches tagged with the corresponding number for later reference. The number of flower caps in each bag was counted to give the number of flowers per inflorescence.
3.8.2 Fruitset
Close to ripeness, the 5 tagged bunches (section 3.8.1) were harvested into individual labelled plastic bags and transported to the laboratory. The total number of coloured berries on each bunch was counted. Percent fruitset for each bunch was calculated using the following formula:
% fruitset = 100 x (no. berries per bunch / no. flowers per bunch)
It could be argued that measuring fruitset using the bagging method is not representative of fruitset on the inflorescences on the vine as a whole given the potential effects of the mesh bags on pollen flow, temperature, humidity and light interception inside the bags; all of which are factors known to affect the success of pollination and fertilisation. However, estimating flower number per inflorescence by other means, for example by measuring rachis length, has a large error associated with it (Gridley 2003). Bagging bunches was considered the best way to accurately quantify flower number in order to calculate fruitset. The relative differences in fruitset between treatments is believed to be valid.
3.8.3 Ovary retention at harvest
The total number of ovaries that remained on bunches at harvest was calculated on all the bagged bunches from McLaren Vale in 2004-05. The calculation is given by the equation below:
% ovary retention = 100 x (no. berries per bunch + no. LGOsa per bunch) / no. flowers per inflorescence a live green ovaries (see section 3.9 for definition) 3.8.4 Mean berry weight
After removal from the rachis, the berries from each bunch in section 3.8.2 were weighed and their total weight divided by the number of berries per bunch to give the average berry weight.
Average berry weight = weight of berries per bunch / no. berries per bunch
3.9 Classification of berry types
Berries were categorised at harvest using the following classifications:
Live green ovaries (LGOs) All green ovaries that remained on the rachis at harvest (Figure 3.6).
Chicken berries All coloured berries that contained no functional seeds (seeds with embryo and endosperm) (Figure 3.6).
Hen berries All coloured berries that contained at least one functional seed (Figure 3.6).
C
B
A
Figure 3.6. A Merlot bunch showing typical hen berries (A), chicken berries (B) and live green ovaries (LGOs) (C). Note that chicken berries were classified as such based on the absence of functional seeds and not on size alone. 3.10 Statistical methods
All data was analysed using the computer program GraphPad PrismTM. The data was analysed using a one-way analysis of variance and where significant differences were found (p<0.05) a Tukey’s Multiple Comparison Test was used to determine separation of treatment means. Where percentage data did not meet normality assumptions the data were arc sine transformed. Chapter 4 Effects of sodium molybdate sprays on vine and soil nutrient status
4.1 Introduction
Molybdenum (Mo) is a trace element essential for plant growth. However, apart from its function in nitrate reductase enzymes there is little known about the function of molybdenum in higher plants (Agarwala et al. 1979).
Molybdenum deficient plants may be smaller than those grown with adequate molybdenum (Chatterjee & Nautiyal 2001), or display symptoms such as the ‘whiptail’ disorder (Fido et al. 1977). The reproductive structures may also be affected by Mo-deficiency. Abnormal flower formation, delay in or suppression of flowering, low pollen grain numbers, decreased viability of pollen and decreased enzymatic activity in the pollen have all been associated with Mo- deficiency (Agarwala et al. 1979). In some plants grown under Mo-deficiency, yield is reduced (Yu et al. 1999), seeds may be small and deformed (Vunkova-Radeva et al. 1988), and the plants have increased sensitivity to low temperature and waterlogging stress (Vunkova-Radeva et al. 1988). In grapevines Mo-deficiency has been suggested as the potential cause of the ‘Merlot problem’ characterised by poor growth of young vines (Robinson and Burne 2000). Molybdenum deficiency is suggested to be more prevalent in cool, wet seasons (pers. comm., C. Williams 2003; Williams et al. 2004) combined with low pH soils that are high in concentrations of adsorbing oxides such as Fe, Mn and Al (Kaiser et al. 2005).
Molybdenum, in various forms, has been applied to grapevines by a number of researchers as a remedial measure to combat low yield. In Australia, Williams et al. (2004) first reported significant yield improvements on Merlot when vines were treated with two foliar sprays of sodium molybdate in springtime and suggested that Mo-deficiency occurred when petiolar molybdenum concentration was between 0.05-0.09 mg/kg at modified E-L stage (MELS) 25 (80% flowering) (Williams et al. 2004). This presents a remediation problem with very practical implications. When tissue analysis is performed at flowering, the results are available only after flowering. If a deficiency is found in a nutrient that affects flower development or flowering and subsequent fruitset, it is too late to remedy the situation in that season. There is a requirement for an early season diagnostic tool to detect molybdenum deficiency prior to flowering.
36 In the previous research examining the effects of molybdenum on Merlot, petiole analysis was conducted at flowering time to quantify the difference in vine molybdenum status between sprayed and unsprayed vines (Gridley 2003; Phillips 2004; Williams et al. 2004). In all cases molybdenum increased in the petioles at flowering time in response to molybdenum sprays that were applied in early spring. However, there was extensive variability in petiolar molybdenum concentration at flowering time, making it difficult to compare data across the different studies. In those studies the leaves and thus petioles were sprayed with sodium molybdate, and were therefore contaminated at the time of sampling. This has raised some concern about the value of petiole analysis for measurement of molybdenum in Mo-treated vines.
In recent years grapegrowers have increasingly applied molybdenum preparations to Merlot as a standard practice as ‘insurance’ against poor yield (pers. comm., B. Miller 2004). Research into the effects of molybdenum on yield of Merlot has been underway in Australia since the late 1990s; however there has been limited, if any, investigation into its carryover in vines from season to season, or the potential residual effects on vine and soil nutrient status.
4.1.1 Aims
The aims of this study were to:
• Evaluate the usefulness of tissue analysis at MELS 12 (shoots 10 cm) as a potential indicator of molybdenum deficiency at flowering time.
• Quantify the effects of sodium molybdate foliar sprays on molybdenum concentration in the vines and on other vine nutrients.
• Investigate whether molybdenum is carried over in the vines from season to season after molybdenum treatment and examine its potential residual effects on other vine nutrients.
• Evaluate the usefulness of petiole analysis after Mo-treatment by examining:
• the relationship between petiole and shoot tip analysis
• the relationship between petiole and inflorescence analysis
• Characterise the soil properties of the two experimental sites.
• Compare soil from beneath treated and untreated vines to examine the effects of sodium molybdate sprays on soil nutritional status.
37 4.2 Materials and methods
4.2.1 Experimental sites
The experiments were conducted on own-rooted Merlot (clone D3V14) vines in commercial vineyards in the Adelaide Hills and at McLaren Vale. A detailed description of the experimental sites, experimental design and spray treatments is provided in Chapter 3.
4.2.2 Weather data
Weather data was collected using on-site weather stations, the details of which are described in Chapter 3.
4.2.3 Timing of tissue nutrient analysis
Tissue was collected for nutritional analysis at two stages of the season in order to compare nutrient levels before and after the application of sodium molybdate. The first tissue sample was collected immediately prior to the first spray application at MELS 12 (shoots 10 cm) and the second tissue sample was collected at MELS 25 (80% flowering) (Coombe 1995).
In 2004-05 vines in the ‘old block’ that received sodium molybdate treatment in the previous season displayed delayed budburst and delayed early spring growth, which persisted until flowering time. This meant that spraying and tissue sampling dates prior to flowering were different for the controls and molybdenum treatments. It was deemed more important to delay these activities on those vines so that all vines received treatment or had samples taken from them at the same growth stage.
4.2.4 Tissue sampling procedure
All samples were collected in the early morning to avoid diurnal nutrient fluxes within the vines and to maintain consistency between samples.
Tissue types sampled Petiole samples were collected in 2003-06 to compare with known flowering time standards. Each sample consisted of 15-20 petioles collected from nodes opposite bunches. Care was taken to avoid collecting petioles subtending bunches on which measurements were being taken. 38
Shoot tips were collected in 2003-05 to enable analysis of fresh, unsprayed tissue at flowering time and therefore avoid potential contamination from the molybdenum sprays (pers. comm., B. Robinson 2003). Shoot tips were cut directly below the second separated leaf on actively growing shoots. Samples contained approximately 20 shoot tips each. Care was taken to avoid removing the tips of shoots on which bunches were being studied.
Inflorescences were collected in 2004-05 only. Each sample contained 15-20 inflorescences.
4.2.5 Replication of tissue samples
MELS 12 tissue samples Due to the lack of available plant tissue at MELS 12, replication of samples was not possible. In 2003-04 the pre-spray tissue samples were collected from across the entire experimental site to gain an overall picture of the nutritional status of the vines. In 2004-05 the MELS 12 tissue samples were collected from across the entire ‘new’ block. In the ‘old’ block, samples were collected from the seven replicates of each treatment applied in the previous season and bulked together to give a single sample for each treatment. In 2004-05 the shoot tip sample for rate 0 was misplaced so for that sample it was assumed that the molybdenum concentration was the same as for the control in the old block.
MELS 25 tissue samples At 80% flowering, tissue samples were collected from each replicate and analysed separately.
4.2.6 Tissue sample processing
All samples were collected and placed into clean, labelled paper bags and stored in a cooled esky during transportation to the laboratory.
Petiole washing prior to analysis Washing of petioles prior to drying and analysis is generally only recommended where there are visible or likely surface contaminants from soil or from known contaminants such as foliar sprays (Reuter et al. 1986). Even with foliar spray contamination, numerous washing formulae have been trialled with inconclusive outcomes and ambiguous recommendations.
39 Given the lack of recommended washing procedures, rinsing with deionised water was suggested as a preliminary treatment for assessing the effect of washing on molybdenum concentration in petioles (pers. comm., B. Robinson 2003).
In 2003-04 an experiment was conducted to evaluate the necessity of washing petioles to remove surface contaminants prior to drying. Half of the petioles from four replicates of each molybdenum treatment in the Hills were washed in deionised water for approximately two minutes then blotted with paper towel to remove excess water. Samples were processed and analysed following the procedures described below.
Results of petiole washing The concentration of molybdenum was significantly lower in the washed petioles from rates 1 and 2 compared to the unwashed petioles. Washing the petioles removed approximately half of the molybdenum. Washing the petioles also affected some of the other nutrients analysed, however there was no significant interaction with molybdenum treatment (Table 4.1).
Given that there was most interest in the molybdenum concentration measured in the controls in order to distinguish between deficient and adequate molybdenum levels, it was decided that washing petioles prior to analysis was unnecessary.
Dust residues on the surface of the petioles may explain the significant difference in iron concentration in the washed petioles from the control vines. The significant differences in Cu, Mn, Zn and Na between the washed and unwashed petioles is likely to have occurred as a result of removal of fungicide and sodium molybdate residues from the surface of the petioles (Table 4.1). These differences are not thought to be of any practical significance.
40 Table 4.1. Nutritional analyses of washed and unwashed petioles from Merlot vines treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays in the Adelaide Hills in 2003-04. Different letters after the values represent significant differences between unwashed and washed values within a treatment (p<0.05). Mo treatment Rate 0 Rate 1 Rate 2 Fe (mg/kg) Unwashed 18 a 19 a 18 a Washed 15 b 16 a 16 a Cu (mg/kg) Unwashed 25 a 29 a 27 a Washed 21 b 24 a 22 a Mn (mg/kg) Unwashed 181 a 176 a 192 a Washed 148 a 148 a 152 b Mo (mg/kg) Unwashed 0.186 a 14.627 a 28.473 a Washed 0.099 a 8.052 b 14.601 b B (mg/kg) Unwashed 37 a 37 a 39 a Washed 36 a 38 a 37 a Zn (mg/kg) Unwashed 80 a 87 a 86 a Washed 65 b 71 a 70 a Ca (mg/kg) Unwashed 18371 a 17471 a 17171 a Washed 16675 a 17300 a 16650 a K (mg/kg) Unwashed 44143 a 44000 a 46000 a Washed 43750 a 41750 a 44250 a Mg (mg/kg) Unwashed 5743 a 5757 a 5429 a Washed 5275 a 5600 a 5425 a P (mg/kg) Unwashed 5371 a 5100 a 5614 a Washed 5100 a 5100 a 5025 a Na (mg/kg) Unwashed 533 a 651 a 703 a Washed 408 b 578 a 553 a S (mg/kg) Unwashed 1730 a 1930 a 1803 a Washed 1568 a 1768 a 1735 a
Tissue drying and grinding All petiole, shoot tip and inflorescence samples were dried overnight in a fan-forced oven at 80°C. Dried samples were ground into particles less than 1mm in diameter using an IKA A10 grinder. In 2003-04 a standard stainless steel blade was used in the grinder, however, in 2004- 05 it was fitted with a carbide metal beater (86% tungsten, 14% cobalt) to prevent potential contamination arising from wear and tear on the blade.
4.2.7 Tissue analyses
Waite Analytical Services performed the analyses of Mo, Fe, Mn, Cu, B, Zn, Ca, Mg, K, P, and Na using Inductively Coupled Plasma Optical Emission Spectrometry (ICPOES) in the first instance and thereafter using Inductively Coupled Plasma Mass Spectrometry (ICPMS) to measure molybdenum concentration below 0.1 mg/kg. Samples for analysis of total and nitrate nitrogen were sent to Analytical Crop Management Laboratory / South Australian Soil and Plant Analysis Service at Loxton, South Australia. These analyses were performed on petiole samples from the controls and rate 1 treated vines from McLaren Vale only.
41 4.2.8 Soil sampling procedure
Soil samples were collected from the experimental sites in July 2005. Samples were collected from two depths, 0-30 cm and 70-100 cm in the vine rows. Soil was sampled from each replicate then the replicates were bulked together, mixed thoroughly and separated into three replicates for analysis.
4.2.9 Soil analyses
Soil samples were analysed for K, Fe, Mo, P, S, nitrate N, ammonium, pH (CaCl2), conductivity and organic carbon. The soil analyses were conducted at CSBP laboratory in Perth, Western Australia.
4.2.10 Statistical analyses
No statistical analyses were performed on the tissue samples collected at MELS 12. Tissue samples collected at MELS 25 were analysed using a one-way analysis of variance and where there were significant differences (p<0.05) a Tukey’s Multiple comparison test was performed to determine separation of treatment means. No statistical analyses were performed on the soil samples.
42 4.3 Results
4.3.1 Phenology and weather data
Both budburst and 80% flowering occurred approximately two weeks later in the Hills compared to McLaren Vale in both 2003-04 and 2004-05. The mean daily minimum temperature tended to be lower in the Hills compared to McLaren Vale, however, the mean daily maximum temperatures tended to be higher in the Hills. The mean daily temperature and day degree summation during the period from budburst to 80% flowering were similar for both the Hills and McLaren Vale (Table 4.2).
Table 4.2. Weather observations from budburst to 80% flowering on Merlot in the Adelaide Hills and at McLaren Vale in 2003-04 and 2004-05 Budburst to 80% No. Mean daily Mean daily Mean daily Day flowering days min. (˚C) max. (˚C) temp. (˚C) degrees a
Adelaide Hills 2003-04 29 Sept. - 9 Dec. 72 7.5 19.8 13.7 308 2004-05 26 Sept. - 29Nov. 65 7.5 21.0 14.3 287
McLaren Vale 2003-04 12 Sept. - 24 Nov. 74 9.9 18.6 14.3 319 2004-05 New block 12 Sept. - 12 Nov. 61 9.6 19.1 14.4 271 Old block 16 Sept. - 16 Nov. 61 9.7 19.5 14.6 284 a Day degrees equals the mean daily temperature minus ten, summed over the period from budburst to 80% flowering.
4.3.2 Molybdenum in vine tissues at modified E-L stage 12
Adelaide Hills The concentration of molybdenum in the petioles from across the experiment at MELS 12 in the Adelaide Hills in 2003-04 was higher than for the control in 2004-05. In 2004-05 the petiolar molybdenum concentration was similar for all treatments (Table 4.3).
At MELS 12 in 2003-04, the concentration of molybdenum in the shoot tips from across the experiment was higher than in the controls in 2004-05. In 2004-05 molybdenum concentration in the shoot tips from control vines was higher than for vines treated with rates 1 and 2 (Table 4.3).
43 In 2004-05 the concentration of molybdenum in the inflorescences at MELS 12 was similar for all treatments (Table 4.3).
McLaren Vale At McLaren Vale in 2003-04, the concentration of molybdenum in the petioles at MELS 12 was higher than any sample collected at the same stage of growth in 2004-05. In the old block in 2004-05, petiolar molybdenum concentration was similar for all treatments and to that in the new block (Table 4.3).
In 2004-05, petiolar molybdenum concentration at MELS 12 was similar at McLaren Vale and in the Hills (Table 4.3).
In 2003-04, the concentration of molybdenum in the shoot tips was high compared to similarly untreated vines in the new block in 2004-05. The molybdenum concentration in the shoot tips from the control vines in the new block was approximately half that in those from Mo-treated vines (Table 4.3).
In 2004-05, molybdenum concentration in the shoot tips from Mo-treated vines at McLaren Vale tended to be higher than in the Hills (Table 4.3).
In 2004-05, molybdenum concentration in the inflorescences at MELS 12 was similar in both the new and old blocks (0.020 mg/kg and 0.015 mg/kg respectively). However, the inflorescences from rates 1 and 2 treated vines tended to be higher in molybdenum (0.185 mg/kg and 0.076 mg/kg respectively) compared to the controls (Table 4.3).
44 Table 4.3. Concentration of molybdenum in the petioles, shoot tips and inflorescences of Merlot at modified E-L stage (MELS) 12 (shoots 10 cm) in 2003-04 and 2004-05. Samples from the Hills in 2004-05 and those from the old block at McLaren Vale were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 in 2003-04. All other samples had not received any molybdenum treatment prior to sampling in that season. Samples were not replicated and statistical analysis was not performed. Molybdenum treatment Rate 0 Rate 1 Rate 2 2003-04
Hills Petioles 0.684 Shoot tips 0.111
McLaren Vale Petioles 2.671 Shoot tips 1.140
2004-05
Hills Petioles 0.054 0.066 0.073 Shoot tips 0.498 0.066 0.077 Inflorescences 0.058 0.056 0.083
McLaren Vale New block Petioles 0.070 Shoot tips 0.054 Inflorescences 0.020
Old block Petioles 0.023 0.056 0.055 Shoot tips 0.054 0.130 0.127 Inflorescences 0.015 0.185 0.076
45 4.3.3 Other nutrients in the vines at modified E-L stage 12
Adelaide Hills In the Hills there were no notable differences between the nutritional analyses from the controls and the Mo-treated vines at MELS 12 in 2003-04 and 2004-05 (Table 4.4).
Table 4.4. Nutritional analyses of petioles, shoot tips and inflorescences collected from Merlot vines in the Adelaide Hills at modified E-L stage (MELS) 12 (shoots 10 cm) in 2003-04 (prior to Mo treatment) and in 2004- 05 after treatment with rate 0 (control), rate 1 and rate 2 at MELS 12 in 2003-04. 2004-05 2003-04 Mo treatment Rate 0 Rate 1 Rate 2 Fe (mg/kg) Petioles 63 27 32 30 Shoot tips 143 111 100 106 Inflorescences 68 78 81 Cu (mg/kg) Petioles 184 23 24 22 Shoot tips 230 21 20 21 Inflorescences 21 22 21 Mn (mg/kg) Petioles 104 78 80 66 Shoot tips 128 122 105 102 Inflorescences 99 111 97 Mo (mg/kg) Petioles 0.684 0.054 0.066 0.073 Shoot tips 0.111 0.498 0.066 0.077 Inflorescences 0.058 0.056 0.083 B (mg/kg) Petioles 36 47 38 39 Shoot tips 46 35 30 33 Inflorescences 40 38 37 Zn (mg/kg) Petioles 118 53 68 56 Shoot tips 77 57 59 62 Inflorescences 56 65 63 Ca (mg/kg) Petioles 11200 10700 11100 10000 Shoot tips 10300 8800 8300 8000 Inflorescences 10200 10600 10700 K (mg/kg) Petioles 42000 44000 45000 43000 Shoot tips 29000 25000 26000 27000 Inflorescences 26000 28000 27000 Mg (mg/kg) Petioles 2000 2500 2500 2400 Shoot tips 2600 2200 2100 2200 Inflorescences 2300 2400 2500 P (mg/kg) Petioles 8600 6900 7300 6700 Shoot tips 9200 7200 7500 7900 Inflorescences 6000 6600 6300 Na (mg/kg) Petioles 1220 470 500 450 Shoot tips 860 460 460 480 Inflorescences 510 560 540 Al (mg/kg) Petioles 31 12 14 13 Shoot tips 128 72 62 67 Inflorescences 34 44 47 S (mg/kg) Petioles 5200 3000 3300 3200 Shoot tips 5000 3700 3800 3900 Inflorescences 3500 3800 3600
46 McLaren Vale At McLaren Vale there were seasonal differences in Fe, Mn, Ca, S and Na concentrations in the petioles and shoot tips at MELS 12, however there were no notable differences in the nutritional analyses between treatments (Table 4.5).
Table 4.5. Nutritional analyses of petioles, shoot tips and inflorescences collected from Merlot vines at McLaren Vale at modified E-L stage 12 (shoots 10 cm) in 2003-04 (prior to Mo treatment) and in 2004-05. Vines in the new block had not received any previous Mo treatment and those in the old block received rate 0 (control), rate 1 and rate 2 molybdenum sprays in 2003-04. 2003-04 2004-05 New 2004-05 Old block block Mo treatment Rate 0 Rate 1 Rate 2 Fe (mg/kg) Petioles 47 29 30 54 47 Shoot tips 135 77 112 360 117 Inflorescences 55 59 117 118 Cu (mg/kg) Petioles 35 25 26 43 91 Shoot tips 38 28 28 91 94 Inflorescences 29 31 70 69 Mn (mg/kg) Petioles 65 21 20 25 23 Shoot tips 108 22 21 33 29 Inflorescences 22 21 27 33 Mo (mg/kg) Petioles 2.671 0.070 0.023 0.056 0.055 Shoot tips 1.140 0.054 - 0.130 0.127 Inflorescences 0.020 0.015 0.185 0.076 B (mg/kg) Petioles 51 62 58 58 56 Shoot tips 37 47 41 52 50 Inflorescences 50 46 55 56 Zn (mg/kg) Petioles 137 88 100 83 72 Shoot tips 94 67 68 67 64 Inflorescences 64 64 64 56 Ca (mg/kg) Petioles 11000 8000 8900 10100 10300 Shoot tips 8200 6067 5700 7300 7400 Inflorescences 6300 6700 7700 7200 K (mg/kg) Petioles 37000 46000 45000 40000 38000 Shoot tips 30000 28333 28000 26000 25000 Inflorescences 29000 29000 26000 25000 Mg (mg/kg) Petioles 1820 2033 2100 2700 2400 Shoot tips 2500 1943 1880 2100 2000 Inflorescences 2060 2000 2200 1910 P (mg/kg) Petioles 6800 5567 5800 5300 5000 Shoot tips 9000 7367 7400 7100 6800 Inflorescences 6133 6100 5700 5300 Na (mg/kg) Petioles 4800 443 460 1090 890 Shoot tips 1180 263 240 2300 1990 Inflorescences 340 360 1170 1010 Al (mg/kg) Petioles 25 11 8 45 36 Shoot tips 85 37 86 380 300 Inflorescences 21 24 97 93 S (mg/kg) Petioles 3900 2667 2700 2700 2500 Shoot tips 4700 3333 3400 3700 3600 Inflorescences 3000 3100 3100 2900
47 4.3.4 Molybdenum in vine tissues at modified E-L stage 25
Adelaide Hills - petioles The concentration of molybdenum in the petioles of the control vines in the Hills was considered adequate in the two seasons of the experiments (0.186 mg/kg in 2003-04 and 0.153 mg/kg in 2004-05) according to the standards proposed by Williams et al. (2004). In 2003-04 application of rate 1 foliar spray at MELS 12 significantly increased the concentration of molybdenum in the petioles at MELS 25 compared to the controls. Doubling the concentration of molybdenum applied to the vines (rate 2) increased petiolar molybdenum concentration approximately two-fold. In 2004-05, when the same experimental vines were left unsprayed in spring, the concentration of molybdenum in the petioles returned to similar levels found in the control vines in 2003-04 and there were no significant differences between treatments (p<0.05) (Figure 4.1 and Table 4.6).
40 0.24 a ) c kg 30 0.18 a a on (mg/ i
rat 20 0.12 b
10 0.06 Mo concent a 0 0.00 Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2
Figure 4.1. Concentration of molybdenum (mg/kg) in Merlot petioles at modified E-L stage (MELS) 25 (80% flowering) in the Adelaide Hills. In 2003-04 (■) vines received rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm). In 2004-05 (■) the vines were left unsprayed. Different letters above the columns indicate significant differences between treatments (p< 0.05). Note the different scales.
Adelaide Hills - shoot tips Application of molybdenum sprays to the vines at MELS 12 significantly increased the molybdenum concentration in the shoot tips at MELS 25. In 2004-05, when the same vines were left unsprayed, the concentration of molybdenum in the shoot tips dropped to a level similar to that found in the control in 2003-04 and there were no significant differences between treatments (Figure 4.2 and Table 4.6).
48 1.00 c ) kg 0.75
ion (mg/ b at
r 0.50
0.25 a a Mo concent a a
0.00 Rate 0 Rate 1 Rate 2
Figure 4.2. Concentration of molybdenum (mg/kg) in Merlot shoot tips at modified E-L stage (MELS) 25 (80% flowering) in the Adelaide Hills. In 2003-04 (■) vines received rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm). In 2004-05 (■) the vines were left unsprayed. Different letters above the columns indicate significant differences between treatments and seasons (p< 0.05).
Adelaide Hills - inflorescences There was no significant difference in molybdenum concentration in the inflorescences between treatments in 2004-05 (Figure 4.3 and Table 4.6).
0.3 a ) kg a 0.2 on (mg/ i a rat
0.1 Mo concent
0.0 Rate 0 Rate 1 Rate 2
Figure 4.3. Concentration of molybdenum (mg/kg) in the inflorescences at modified E-L stage (MELS) 25 (80% flowering) in the Adelaide Hills in 2004-05. In 2003-04 vines received rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm). In 2004-05 (■) the vines were left unsprayed. Different letters above the columns indicate significant differences between treatments (p< 0.05).
McLaren Vale- petioles At McLaren Vale the concentration of molybdenum in the petioles from the control vines at MELS 25 fell within the suggested deficiency range of 0.05-0.09 mg/kg (Williams et al. 2004) in all blocks in all seasons. In 2003-04 the molybdenum concentration in the petioles from vines sprayed with rate 1 increased significantly compared to the control and it approximately doubled from rate 1 to rate 2. In 2004-05 the petiolar molybdenum 49 concentration also increased significantly in the Mo-treated vines compared to the controls (Table 4.6).
There were no significant differences in petiolar molybdenum concentration between the control treatments in the new and old blocks and the 2004-05 controls were not significantly different to the control in 2003-04. The petioles from rate 1 in the new block had significantly more molybdenum than those in the old block and a similar trend was observed for rate 2; however, this was not significant (Figure 4 and Table 4.6).
In 2005-06, when the vines were left unsprayed there was no significant difference in petiolar molybdenum concentration between treatments, and the molybdenum concentration in the petioles fell to similar levels to those in the control vines in all years (Figure 4.4 and Table 4.6).
a a 0.10 8 16 a )
kg aa b ab 0.08 a 6 12 a b 0.06 on (mg/ i 4 8 rat 0.04
2 4 0.02 c c Mo concent 0.00 0 0
Rate 0 Rate 1 Rate 2
Figure 4.4. Concentration of molybdenum (mg/kg) in Merlot petioles at modified E-L stage (MELS) 25 (80% flowering) at McLaren Vale. In 2003-04 (■) vines received rate 0 (left), rate 1 (centre) and rate 2 (right) molybdenum sprays at MELS 12 (shoots 10 cm). In 2004-05, vines in the new block (■) were sprayed at MELS 12 in 2004-05 only and vines in the old block (■) had been sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. In 2005-06 (■) vines from the new block were left unsprayed. Different letters above the columns indicate significant differences between seasons (p<0.05). For significant differences between treatments within a season, see Table 4.6.
McLaren Vale - shoot tips When the vines were sprayed with molybdenum at MELS 12, its concentration in the shoot tips increased significantly at 80% flowering. In 2003-04 the shoot tips from the control had 0.039 mg/kg of molybdenum compared to 0.235 mg/kg for rate 1 and 0.486 mg/kg for rate 2. Similar responses to the molybdenum sprays occurred in the shoot tips in 2004-05; however, the molybdenum concentration in the shoot tips from the Mo-sprayed vines in the new block was significantly less than in those from the old block (Figure 4.5 and Table 4.6). 50
0.05 0.4 0.6
) a a kg a a b 0.04 0.3 ab b a 0.4
on (mg/ 0.03 i a 0.2 rat 0.02 0.2 0.1 0.01 Mo concent 0.00 0.0 0.0
Rate 0 Rate 1 Rate 2
Figure 4.5. Concentration of molybdenum (mg/kg) in Merlot shoot tips at modified E-L stage (MELS) 25 (80% flowering) at McLaren Vale. In 2003-04 (■) vines received rate 0 (left), rate 1 (centre) and rate 2 (right) molybdenum sprays at MELS 12 (shoots 10 cm). In 2004-05, vines in the new block (■) were sprayed at MELS 12 in 2004-05 only and vines in the old block (■) had been sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. Different letters above the columns indicate significant differences between seasons (p<0.05). For significant differences between treatments within a season, see Table 4.6.
McLaren Vale - inflorescences At McLaren Vale there was a significant increase in molybdenum concentration in the inflorescences between treatments at 80% flowering. There were no significant differences between the new and old blocks (Figure 4.6 and Table 4.6).
16 c
) c
kg 12 on (mg/ i b b
rat 8
4 Mo concent
a a 0 Rate 0 Rate 1 Rate 2
Figure 4.6. Concentration of molybdenum (mg/kg) in Merlot inflorescences at modified E-L stage (MELS) 25 (80% flowering) at McLaren Vale. In 2004-05 vines in the new block (■) received rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm). Vines in the old block (■) had been sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. Different letters above the columns indicate significant differences between treatments and seasons (p<0.05).
51 Table 4.6. Concentration of molyb denum (mg/kg) in t he petioles, shoo t tips and inflo rescences of Merlot in the Adelaide Hills and at McLaren Vale at modified E-L stage (MELS) 25. In 2003-04 all vines were treated with rate 0 (control), rate 1 and rate 2 molybdenum sp rays at MELS 1 2 (shoots 10 cm). In 2004-05 vines in the Hills were left uns prayed. At McLaren Vale vines in th e new block were sprayed in 20 04-05 only and vines in the old block had been sprayed at MELS 12 in 2003-04 a nd again at ME LS 12 in 2004-0 5. In 2005-06 vines in the new block at M cLaren Vale were left u nsprayed. Differen t letters after the values within a n experiment indicate sig nificant differences between Mo-treatments within rows (p<0.05). Mo trea tme nt Rate 0 Rate 1 Rate 2 2003-04
Hills Petioles 0.186 a 14.62 7 b 28.47 3 c Shoot tips 0.076 a 0.402 b 0.860 c
McLaren Vale Petioles 0.060 a 7.333 b 14.391 c Shoot tips 0.039 a 0.235 b 0.486 c
2004-05
Hills Petioles 0.153 a 0.123 a 0.135 a Shoot tips 0.063 a 0.077 a 0.058 a Inflorescences 0.185 a 0.143 a 0.121 a
McLaren Vale New block Petioles 0.058 a 7.504 b 11.649 c Shoot tips 0.023 a 0.208 b 0.394 c Inflorescences 0.035 a 6.661 b 11.904 c
Old block Petioles 0.066 a 5.429 b 8.647 c Shoot tips 0.038 a 0.303 b 0.508 c Inflorescences 0.032 a 7.161 b 14.358 c
2005-06
McLaren Vale (2004-05 new Petioles 0.060 a 0.049 a 0.066 a block)
52 4.3.5 Total and nitrate nitrogen in the vines at modified E-L stage 25
There was no significant interaction between Mo-treatment and total or nitrate nitrogen in the petioles at 80% flowering at McLaren Vale in 2003-04 or in 2004-05 (Table 4.8).
4.3.6 Other nutrients analysed in the vines at modified E-L stage 25
2003-04 In the Hills in 2003-04 petiolar sodium and sulphur concentrations tended to be higher in response to molybdenum treatment (Table 7). At McLaren Vale calcium concentration in the shoot tips was significantly lower relative to Mo-treated vines. There were no other significant interactions between Mo-treatment and any of the nutrients analysed (Table 4.7 and Table 4.8).
Table 4.7. Nutrient analyses of petioles and shoot tips from Merlot vines at modified E-L stage (MELS) 25 (80% flowering) in 2003-04 in the Adelaide Hills. Vines were sprayed with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04. Different letters after the numbers indicate significant differences between treatments (p<0.05). Adelaide Hills Molybdenum treatment Rate 0 Rate 1 Rate 2
Fe (mg/kg) Petioles 18 a 19 a 18 a Shoot tips 73 a 72 a 76 a Mn (mg/kg) Petioles 181 a 176 a 192 a Shoot tips 82 a 78 a 83 a Zn (mg/kg) Petioles 80 a 87 a 86 a Shoot tips 54 a 52 a 52 a Ca (mg/kg) Petioles 18371 a 17471 a 17171 a Shoot tips 8543 a 8071 a 8343 a K (mg/kg) Petioles 44143 a 44000 a 46000 a Shoot tips 24000 a 23429 a 23286 a S (mg/kg) Petioles 1730 a 1930 b 1803 ab Shoot tips 2943 a 2943 a 2943 a P (mg/kg) Petioles 5371 a 5100 a 5614 a Shoot tips 5443 a 5386 a 5300 a Mg (mg/kg) Petioles 5743 a 5757 a 5429 a Shoot tips 2243 a 2129 a 2200 a Cu (mg/kg) Petioles 25 a 29 a 27 a Shoot tips 27 a 27 a 26 a B (mg/kg) Petioles 37 a 37 a 39 a Shoot tips 66 a 64 a 59 a Na (mg/kg) Petioles 533 a 651 ab 703 b Shoot tips 200 a 203 a 218 a
53 Table 4.8. Nutrient analyses of petioles and shoot tips from Merlot vines at modified E-L stage (MELS) 25 (80% flowering) in 2003-04 at McLaren Vale. Vines were sprayed with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04. Different letters after the numbers indicate significant differences between treatments (p<0.05). McLaren Vale Molybdenum treatment Rate 0 Rate 1 Rate 2
Fe (mg/kg) Petioles 24 a 23 a 22 a Shoot tips 81 a 79 a 78 a Mn (mg/kg) Petioles 76 a 71 a 76 a Shoot tips 34 a 31 a 33 a Zn (mg/kg) Petioles 99 a 95 a 93 a Shoot tips 58 a 55 a 56 a Ca (mg/kg) Petioles 22429 a 21814 a 21514 a Shoot tips 7214 a 6686 b 6671 b K (mg/kg) Petioles 27000 a 21543 a 23543 a Shoot tips 24143 a 23143 a 23143 a S (mg/kg) Petioles 2271 a 2300 a 2246 a Shoot tips 3186 a 3243 a 3157 a P (mg/kg) Petioles 4643 a 4629 a 4343 a Shoot tips 6286 a 6243 a 6043 a Mg (mg/kg) Petioles 4400 a 5043 a 4657 a Shoot tips 1723 a 1656 a 1661 a Cu (mg/kg) Petioles 21 a 21 a 22 a Shoot tips 26 a 25 a 26 a B (mg/kg) Petioles 51 a 47 a 50 a Shoot tips 84 a 87 a 86 a Al (mg/kg) Petioles 13 a 12 a 13 a Shoot tips 26 a 24 a 23 a Na (mg/kg) Petioles 3000 a 2571 a 2687 a Shoot tips 148 a 127 a 135 a Total N (%) Petioles 1.0 a 1.1 a
Nitrate N (mg/kg) Petioles 1033 a 907 a
2004-05 In the Hills in 2004-05 calcium concentration in the shoot tips from Mo-sprayed vines was significantly lower than the control (Table 4.9). At McLaren Vale the nutrient analyses from Mo-sprayed vines in the new block tended not to differ from the control however, in the old block, vines treated with rates 1 and 2 tended to have higher potassium and lower magnesium concentrations in the petioles, and higher sodium and lower sulphur, phosphorus and copper concentrations in the shoot tips compared to all other treatments in 2004-05 (Table 4.10). There were no other significant interactions between Mo-treatment and other nutrients analysed.
54 Table 4.9. Nutrient analyses of petioles, shoot tips and inflorescences from Merlot vines at modified E-L stage (MELS) 25 (80% flowering) in 2004-05 in the Adelaide Hills. Vines were sprayed with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and left unsprayed in 2004-05. Different letters after the numbers indicate significant differences between treatments (p<0.05). Adelaide Hills Molybdenum treatment Rate 0 Rate 1 Rate 2
Fe (mg/kg) Petioles 15 a 16 a 15 a Shoot tips 66 a 66 a 65 a Inflorescences 57 a 58 a 58 a Mn (mg/kg) Petioles 96 a 93 a 98 a Shoot tips 64 a 68 a 71 a Inflorescences 122 a 129 a 123 a Zn (mg/kg) Petioles 93 a 101 a 98 a Shoot tips 50 a 52 a 50 a Inflorescences 74 a 70 a 74 a Ca (mg/kg) Petioles 16586 a 16543 a 16843 a Shoot tips 8186 a 8329 a 8186 a Inflorescences 9786 a 9729 a 9214 a K (mg/kg) Petioles 57857 a 56857 a 57857 a Shoot tips 24571 a 24857 a 25143 a Inflorescences 32857 a 32286 a 32429 a S (mg/kg) Petioles 1743 a 2160 b 1969 ab Shoot tips 3029 a 3071 ab 3157 b Inflorescences 2514 a 2671 b 2614 ab Al (mg/kg) Petioles 6.3 a 5.6 a 6.3 a Shoot tips 14.7 a 14.2 a 12.4 a Inflorescences 18.1 a 18.5 a 18.0 a P (mg/kg) Petioles 3729 a 4371 a 3543 a Shoot tips 5514 a 5643 a 5743 a Inflorescences 4329 a 4471 a 4214 a Mg (mg/kg) Petioles 4243 a 3929 a 4586 a Shoot tips 2143 a 2157 a 2129 a Inflorescences 2900 a 2786 a 2743 a Cu (mg/kg) Petioles 77 a 85 a 81 a Shoot tips 48 a 47 a 44 a Inflorescences 104 a 113 a 108 a B (mg/kg) Petioles 36 a 36 a 35 a Shoot tips 50 a 53 a 48 a Inflorescences 41 a 41 a 40 a Na (mg/kg) Petioles 877 a 976 a 914 a Shoot tips 54 a 54 a 45 a Inflorescences 464 a 447 a 450 a
55 Table 4.10. Nutrient analyses of petioles, shoot tips and inflorescences from Merlot vines at modified E-L stage (MELS) 25 (80% flowering) in 2004-05 at McLaren Vale. Vines in the new block were sprayed with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and vines in the old block were sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. Different letters after the values indicate significant differences between treatments (p<0.05).
McLaren Vale Molybdenum treatment Rate 0 Rate 1 Rate 2 New block Old block New block Old block New block Old block
Fe (mg/kg) Petioles 13 a 14 a 13 a 15 a 14 a 15 a Shoot tips 84 a 114 a 88 a 98 a 93 a 96 a Inflorescences 52 a 60 a 60 a 63 a 58 a 62 a Mn (mg/kg) Petioles 22 a 20 a 19 a 18 a 32 a 19 a Shoot tips 30 a 28 a 27 a 28 a 32 a 27 a Inflorescences 27 a 25 a 26 a 30 a 33 a 24 a Zn (mg/kg) Petioles 43 a 49 a 48 a 49 a 46 a 45 a Shoot tips 54 a 53 a 52 a 46 b 54 a 47 b Inflorescences 29 a 31 a 33 a 34 a 31 a 33 a Ca (mg/kg) Petioles 16757 a 17629 a 17571 a 16686 a 16986 a 16229 a Shoot tips 9314 a 9086 a 9129 a 9500 a 9314 a 9129 a Inflorescences 10114 a 10371 a 9900 a 10557 a 10729 a 10314 a K (mg/kg) Petioles 31286 a 32571 a 31714 a 35857 b 30714 a 36571 b Shoot tips 25714 a 25286 a 25429 a 24857 a 25429 a 25286 a Inflorescences 24429 ab 24571 ab 24143 a 26143 ab 24857 ab 26429 b S (mg/kg) Petioles 1571 a 1677 a 1597 a 1660 a 1607 a 1537 a Shoot tips 3357 a 3357 a 3314 a 3057 b 3343 a 3000 b Inflorescences 2343 a 2386 a 2343 a 2414 a 2414 a 2371 a Al (mg/kg) Petioles 6 a 7 a 6 a 8 a 7 a 8 a Shoot tips 32 a 43 b 38 ab 44 b 40 ab 41 b Inflorescences 21 a 23 ab 26 ab 28 b 24 ab 24 ab P (mg/kg) Petioles 2199 a 2227 a 2279 a 2071 a 2091 a 1803 a Shoot tips 6171 a 6129 a 6000 a 5586 b 5857ab 5471 b Inflorescences 3386 a 3486 a 3443 a 3557 a 3429 a 3443 a Mg (mg/kg) Petioles 5371 a 5071 a 5314 a 4086 b 5500 a 4129 b Shoot tips 2300 a 2186 a 2257 a 2300 a 2329 a 2286 a Inflorescences 2586 a 2471 a 2557 a 2643 a 2714 a 2529 a Cu (mg/kg) Petioles 33 a 38 a 33 a 36 a 32 a 38 a Shoot tips 69 a 74 a 76 a 44 b 79 a 43 a Inflorescences 95 a 103 a 110 a 101 a 64 a 64 a B (mg/kg) Petioles 49 ab 52 ab 52 a 45 b 51 ab 45 b Shoot tips 76 a 73 a 82 a 69 a 79 a 65 a Inflorescences 63 a 60 a 70 a 59 a 62 a 54 a Na (mg/kg) Petioles 1424 a 1776 b 1563 ab 1667 b 1763 b 1630 ab Shoot tips 1067 a 1091 a 1147 a 1686 b 1156 a 1631 b Inflorescences 1023 a 1104 a 1130 ac 1391bc 1151ac 1596 b Total N (%) Petioles 0.83 a 0.87 ab 0.82 a 0.89 b
Nitrate N Petioles 483 ab 741a 377 b 463 ab (mg/kg)
56 4.3.7 Adelaide Hills and McLaren Vale soil properties
Soil was more acidic in the Hills than at McLaren Vale at 0-30 cm and 70-100 cm. Soil conductivity was lower in the Hills than at McLaren Vale. Molybdenum treatment did not affect either soil pH or conductivity at either site (Table 4.11).
4.3.8 The effects of sodium molybdate foliar sprays on molybdenum and other soil nutrients
Molybdenum The concentration of molybdenum was higher in the soil at McLaren Vale compared to the Hills (Table 4.11).
In the Hills the concentration of molybdenum in the soil at 0-30 cm was marginally higher under the Mo-treated vines compared to the soil beneath untreated control vines. There was no difference in soil molybdenum concentration between the treatments at 70-100 cm (Table 4.11).
At McLaren Vale molybdenum tended to be higher in the upper soil profile beneath Mo- treated vines compared to soil sampled from under unsprayed control vines. At 70-100 cm there was little difference in molybdenum concentration in the soil profile between treatments (Table 4.11).
Nitrogen The levels of nitrate and ammonium in the soil were similar at both McLaren Vale and in the Hills.
In the Hills the concentration of nitrate N tended to be higher under Mo-treated vines than the untreated controls at both depths. The concentration of ammonium in the soil in the Hills was similar for both treatments at both 0-30 cm and 70-100 cm (2mg/kg) (Table 4.11).
At McLaren Vale the concentration of nitrate N in the soil at 0-30 cm was lower under the control vines than vines sprayed with molybdenum. A similar trend was observed at 70-100 cm. The concentration of ammonium in the soil was not significantly different beneath control vines compared to soil under vines treated with molybdenum at either depth (Table 4.11). 57 Iron There was a notable difference in iron concentration in the soil between the Hills and McLaren Vale. However, molybdenum treatment did not affect the concentration of iron in the soil at either experimental site (Table 4.11).
Table 4.11. Analyses of soil properties performed in winter 2005 from 0-30 cm and 70-100 cm. Vines in the Hills were sprayed with rate 0 (control) and rate 1 Mo sprays in 2003-04 only. At McLaren Vale vines in the new block were sprayed in 2004-05 only and vines in the old block were sprayed in 2003-04 and 2004-05. Adelaide Hills McLaren Vale Rate 0 Rate 1 Rate 0 Rate 1 New block Old block 0-30 70-100 0-30 70-100 0-30 70-100 0-30 70-100 0-30 70-100 cm cm cm cm cm cm cm cm cm cm Molybdenum 6 6 8 6 13 12 23 15 39 15 (µg/kg) Nitrate N 7 2 9 10 9 11 11 12 14 15 (mg/kg) Ammonium 2 2 2 2 2 3 3 4 2 3 (mg/kg) Phosphorus 30 11 29 19 28 13 22 13 32 9 (mg/kg) Potassium 196 132 204 234 533 385 425 369 534 323 (mg/kg) Sulphur 16 30 22 64 9 46 18 34 58 29 (mg/kg) Organic 1.33 0.61 1.32 0.79 1.35 1.15 1.28 0.98 1.67 1.09 carbon (%) Iron (mg/kg) 1055 627 1147 750 485 471 467 496 488 547 Conductivity 0.069 0.082 0.106 0.205 0.228 0.251 0.226 0.308 0.266 0.297 (dS/m) pH (CaCl2) 5.4 5.4 5.3 5.9 7.5 7.7 7.6 7.8 7.5 7.7
58 4.4 Discussion
4.4.1 Molybdenum in vine tissues at modified E-L stage 12
The aim of performing tissue analysis at MELS 12 was to evaluate its usefulness as an earlier indicator of potential molybdenum deficiency than the more conventional time of MELS 25.
In 2003-04 the molybdenum concentration in both petioles and shoot tips at MELS 12 was higher than that at MELS 25 for both McLaren Vale and in the Hills. However, high molybdenum levels did not eventuate at flowering (Table 4.12). At McLaren Vale, the concentration of molybdenum in both petioles and shoot tips was much higher at MELS 12 than the Hills. The difference did not persist until flowering and the MELS 12 data could not have been used to predict neither the deficiency situation at McLaren Vale nor the adequate level of molybdenum in the Hills at flowering time.
In 2004-05 when tissue analysis at MELS 12 showed lower levels of molybdenum in the petioles (lower than the suggested standard at MELS 25) compared to 2003-04, by flowering time the petiolar molybdenum concentration in the Hills had increased to a level considered to be adequate. However, at McLaren Vale it remained within the suggested deficiency range (Table 4.12). The pattern of molybdenum increase from MELS 12 to MELS 25 was similar to that observed by Williams (pers. comm., 2005) and may be a more typical pattern of molybdenum flux in Merlot during spring and early summer.
Molybdenum appeared to be equally partitioned between the developing tissues at MELS 12. This suggests that, at MELS 12, a highly vegetative stage of growth, the demand for molybdenum is equal amongst petioles, shoot tips and inflorescences.
The data collected at MELS 12 over the two seasons of experimentation do not sufficiently describe the course of molybdenum accumulation or decline in Merlot during the period leading up to flowering. In the Hills, molybdenum tended to increase in the vines from MELS 12 to MELS 25, however in most cases at McLaren Vale there was little change in vine molybdenum status over the same period. It is not possible to draw any firm conclusions about the changes in molybdenum concentration during the season because of the lack of replication and possible error in the data. The differences in the data between the two experimental sites suggest the existence of another factor that may influence vine molybdenum status.
59 Table 4.12. Concentration of molybdenum (mg/kg) in the petioles, shoot tips and inflorescences of Merlot in the Adelaide Hills and at McLaren Vale at modified E-L stages 12 (shoots 10 cm) and 25 (80% flowering) in 2003- 05. Vines in the Hills were sprayed with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 in 2003-04 and left unsprayed in 2004-05. At McLaren Vale vines in the new block were sprayed MELS 12 in 2003-04 and vines in the old block were sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. Molybdenum treatment Rate 0 Rate 1 Rate 2 E-L 12 E-L 25 E-L 12 E-L 25 E-L 12 E-L 25 Petioles Hills 2003-04 0.684 0.186 14.627 28.473 2004-05 0.054 0.153 0.066 0.123 0.073 0.135
McLaren Vale 2003-04 2.671 0.060 7.333 14.391 2004-05 new 0.070 0.058 7.504 11.649 2004-05 old 0.023 0.066 0.056 5.429 0.055 8.647
Shoot tips Hills 2003-04 0.111 0.076 0.402 0.860 2004-05 0.498 0.063 0.066 0.077 0.077 0.058
McLaren Vale 2003-04 1.140 0.039 0.235 0.486 2004-05 new 0.054 0.023 0.208 0.394 2004-05 old - 0.038 0.130 0.303 0.127 0.508
Inflorescences Hills 2004-05 0.058 0.185 0.056 0.143 0.083 0.121
McLaren Vale 2004-05 new 0.020 0.035 6.661 11.904 2004-05 old 0.015 0.032 0.185 7.161 0.076 14.358
4.4.2 Molybdenum in the vines at modified E-L stage 25
In the Adelaide Hills the petiolar molybdenum concentration was considered adequate at flowering time in both seasons. However, at McLaren Vale the concentration of molybdenum in the petioles of control vines fell within the suggested deficiency range at flowering (0.05- 0.09 mg/kg) (Williams et al. 2004) in all experiments. This was not expected and does not support the theory that Merlot vines grown in the Adelaide Hills are more likely to be predisposed to Mo-deficiency as a consequence of climate and soil type.
The phenological and climatic data show that vines in the Hills burst later than McLaren Vale in both seasons. However, as a consequence, the mean daily temperatures in the period between budburst and flowering were similar at both sites. The length of the period from budburst to flowering was also similar for the Hills and McLaren Vale suggesting that both
60 sites experienced similar conditions during that period. Therefore, temperature was not responsible for the differences in vine molybdenum status between the two sites.
4.4.3 The effect of sodium molybdate foliar sprays on molybdenum concentration in the inflorescences and petioles at modified E-L stage 25
Molybdenum concentration increased significantly in the inflorescences and petioles at flowering time in response to molybdenum foliar sprays applied at MELS 12. Nutritional analysis is not normally performed on inflorescences and this is the first known record of the effects of Mo-treatment on molybdenum concentration in inflorescences. The increase in petiolar molybdenum concentration was expected and is in agreement with the findings of Gridley (2003), Phillips (2004) and Williams et al. (2004). When rate 2 was applied to the vines the concentration of molybdenum measured in the petioles was approximately double that in the petioles from rate 1-treated vines showing that the vines treated with rate 2 maintained a proportionally higher level of molybdenum from the time of spraying until tissues were analysed.
The absolute molybdenum values were not consistent between the sites in the current study or other previous reports (Gridley 2003; Phillips 2004; Williams et al. 2004) despite the fact that the same rate of molybdenum was applied in all cases. Petiole washing prior to analysis, or rain, would account for lower molybdenum values in the Williams et al. (2004) study. However, this does not account for all the discrepancies since higher levels were also reported in that study. This further highlights the need for an alternative to petiole analysis to overcome disparities caused by contamination.
4.4.4 The effect of sodium molybdate foliar sprays on molybdenum concentration in the shoot tips at modified E-L stage 25
The increase in molybdenum concentration in the shoot tips at MELS 25 after treatment with sodium molybdate foliar sprays at MELS 12 confirmed that the spray applied to the vines in spring penetrated the vine tissue and molybdenum was mobile within the vines. This is in agreement with Brady (2004) who found that when molybdenum was applied to single leaves on pot grown Mo-deficient Merlot vines, Mo-dependant enzymatic activity was activated in leaves that did not receive molybdenum treatment.
61 In the Hills and at McLaren Vale, the responses to sodium molybdate application, measured by the concentration of molybdenum in the shoot tips at flowering, were of a similar magnitude ranging from approximately five times (in the Hills) to nine times (in the 2004-05 new block at McLaren Vale) that of the control. Compared to rate 1, the shoot tips from rate 2-treated vines consistently contained approximately double the concentration of molybdenum. This demonstrated not only the mobility of molybdenum but also quantifiable differences in molybdenum concentration in untreated parts of the vine in response to different rates of application.
4.4.5 The relationship between petiolar molybdenum concentration and the concentration of molybdenum in the shoot tips and inflorescences at modified E-L stage 25
The relationship between petiolar molybdenum concentration and the concentration of molybdenum in the inflorescences from McLaren Vale over the two seasons of experimentation was strong (r2 = 0.7745) (Figure 4.7). This demonstrates that molybdenum applied in the springtime foliar sprays remained on the inflorescences at flowering and that petiolar molybdenum concentration can be correlated with molybdenum in the inflorescences. The effects of molybdenum sprays on the reproductive structures are discussed in detail in Chapter 7.
20
2
) R = 0 .774 5 g
g/k 15 (m o
e M 10 enc c
5 ores Infl
0 0481216
Petiolar Mo (mg/kg)
Figure 4.7. The relationship between molybdenum concentrations in the petioles and inflorescences from Merlot vines at modified E-L stage 25 (80% flowering) at McLaren Vale in 2004-05.
62 There was a strong relationship between petiolar molybdenum concentration and the concentration of molybdenum in the shoot tips at flowering time (r2 = 0.7912) (Figure 4.8). This information may be beneficial to grapegrowers who routinely perform petiole analysis to monitor molybdenum levels in Merlot in conjunction with the application of ‘preventative’ molybdenum sprays. Rather than analysing petioles that have been contaminated by the foliar spray, shoot tip analysis could be performed in its place, with confidence in the absolute values. Moreover, given the strong correlation between molybdenum concentration in the two tissues types, the existing standards for petioles could be adapted for shoot tips so that only one tissue type need be sampled, saving both time and money.
1. 5
R2 = 0 .79 12
1. 2 ) kg
0.9 p Mo (mg/ i
t 0.6 Shoot 0.3
0.0 0 10 20304050
Petiolar Mo (mg/kg)
Figure 4.8. The relationship between molybdenum concentrations in the petioles and shoot tips from Merlot vines at modified E-L stage 25 (80% flowering) at McLaren Vale in 2003-05.
4.4.6 Molybdenum in the vines in the season after sodium molybdate treatment
Data from the Hills and McLaren Vale suggest that there is not a strong carry over of molybdenum in vines from season to season. When tissue analysis was performed on vines in the season after treatment with sodium molybdate the concentration of molybdenum decreased to levels similar to those in the controls in all cases. This suggests that at sites where Mo-deficiency is consistently a problem, remedial foliar sprays need to be applied every year.
63 4.4.7 Effects of applying sodium molybdate foliar sprays in consecutive seasons
Molybdenum in the vines The concentration of molybdenum in the petioles at flowering time tended to decrease after consecutive seasons of treatment with sodium molybdate. When previously untreated Mo- deficient vines were sprayed with rate 1 (for example, in 2003-04 and the new block in 2004- 05 at McLaren Vale) the petiolar molybdenum concentration at flowering was similar (7.333 mg/kg and 7.504 mg/kg respectively). However, where the same vines were treated with rate 1 in two consecutive seasons (in the old block in 2004-05) the molybdenum concentration in the petioles at flowering was significantly lower (5.429 mg/kg).
A similar observation was made in field experiments by Phillips (2004). He reported lower molybdenum concentrations in petioles from previously treated vines than untreated controls. Furthermore, the data presented by Williams et al. (2004) over a three-year period also showed a successive decline in petiolar molybdenum concentration (Table 4.13). However, in that study no statistical analyses were provided comparing the flowering time petiolar molybdenum concentrations between the seasons nor was there any reference to, or explanation for, the decline in vine molybdenum status over the period of the study. Given the data from the two experiments at McLaren Vale in 2004-05, this trend toward declining petiolar molybdenum concentration after successive seasons of Mo-treatment appears to be significant.
When Phillips (2004) treated pot-grown, own-rooted Merlot vines with sodium molybdate, after two weeks the greatest proportion of molybdenum was translocated to the old wood. One hypothesis to explain the decline in molybdenum in the petioles after multiple seasons of treatment with sodium molybdate is that once the starvation caused by the deficiency is satisfied by the first treatment, the petioles become weaker sinks for molybdenum (pers. comm., B. Kaiser 2006). Excess molybdenum may be translocated and stored in the permanent structures (trunk and roots) of the vine and remobilised in the following spring to oth er organs for example, the shoot tips. This would also explain the increase in molybdenum concentration that occurred in the shoot tips after consecutive seasons of Mo-treatment.
64 Table 4.13. Concentration of molybdenum in Merlot petioles at modified E-L stage (MELS) 25 (80% flowering) from vines sprayed with rate 0 (control) and rate 1 molybdenum sprays at MELS 12 (shoots 10 cm) in 2000, 2001 and 2002 at site 1 (Lower Hermitage), site 2 (Meadows) and site 3 (Kuitpo). Adapted from Williams et al. (2004)
NOTE: This table is included on page 65 of the print copy of the thesis held in the University of Adelaide Library.
Other vine nutrients Application of sodium molybdate to the vines in successive seasons significantly affected levels of potassium, magnesium and boron in the petioles and, in the shoot tips, sodium, sulphur, phosphorous and copper were affected. The differences in petiolar potassium concentration is not thought to have any practical significance, however, additional increase after future Mo-treatment may potentially affect juice composition. Phosphorus was marginal the petioles from all treatments in 2004-05 and the decrease in its concentration in the shoot tips from the old block may be a concern considering that phophorus is readily retranslocated from old to new leaves. Magnesium and copper were well above the suggested adequate range in the petioles from all treatments probably due to contamination from fungicides. The decreases in magnesium and copper after repeated applications of molybdenum are not thought to have any practical significance. Sodium tended to increase after consecutive concentration of sodium in the petioles from all treatments fell within the deficiency range (Robinson 1986). Applying sodium molybdite may be beneficial to the vines in this case. Molybdenum treatment in consecutive seasons affected the molybdenum status of Merlot. The effects of spraying vines with molybdenum on vine nutrient status in the long-term are unknown and this is an area for future investigation.
65 4.4.8 Soil properties at the Adelaide Hills and McLaren Vale experimental sites
The soil results were as expected and reflect those typical for the regions. However, the soil properties at the two experimental sites did not affect the molybdenum status of the vines in the expected way. In the Hills the molybdenum concentration in the soil was lower, and the soil was more acidic and higher in iron compared to McLaren Vale, which should have predisposed the Hills’ vines to molybdenum deficiency. However, as discussed, this was not the case.
Low soil pH (pH<5.5) was cited as one of the causes of Mo-deficiency in Merlot (Williams et al. 2004). However, in that study no strong relationship between soil pH and Mo-deficiency was found. This is in agreement with the current findings. In glasshouse experiments Brady (2004) showed that Merlot vines grown in a nutrient solution (containing sodium molybdate) of pH 6.5 were unable to utilise available molybdenum suggesting that other factors affect the ability of Merlot to extract molybdenum from the soil.
High levels of sulphur and phosphorus in the soil are both associated with the inhibition of molybdenum uptake by plants (Zimmer & Mendel 1999, cited in Brady 2004), however no notable differences between these two elements were found in the current study. The most notable difference between the Hills and McLaren Vale soils, apart from those already mentioned, was conductivity. No previous reports linking soil conductivity and inhibition of molybdenum uptake have been reported. The mechanism for soil conductivity affecting molybdenum uptake is beyond the scope of the present study.
4.4.9 Effects of sodium molybdate sprays on molybdenum accumulation in the soil
This study provides evidence to suggest that molybdenum may accumulate in the soil under vines that are sprayed with sodium molybdate. This has several implications for grapegrowers contemplating spraying molybdenum preparations on their vines on an annual basis. On the one hand it may appear to be beneficial to add molybdenum to the soil in this manner in areas where the soil is low in molybdenum. However, as it has been demonstrated, the presence of moly bd enum in the soil alone does not guarantee its uptake by the vines. On the other hand, accumulation of molybdenum may potentially have detrimental effects. In situations where molybdenum is applied to other grape varieties as ‘insurance’ against low yield, molybdenum accumulation over the long-term may have toxic effects on those varieties. The effect of high 66 levels o f molybdenum on soil micro-organisms and resultant soil interactions should also be considered, however, the current level of information about these interactions is low.
4.5 Conclusions
• Tissue analysis at MELS 12 did not reveal any clear pattern of molybdenum accumulation or decline in treated or untreated vines and its use as an early indicator of molybdenum deficiency requires more research.
• Molybdenum concentration was higher in the petioles, shoot tips and inflorescences of Merlot at MELS 25 when the vines were sprayed with sodium molybdate at MELS 12.
• There were strong relationships between petiolar molybdenum concentration and the concentration of molybdenum in the shoot tips and the inflorescences. Shoot tip analysis may be a useful tool for monitoring molybdenum in vines at MELS 25.
• There was no carryover of molybdenum in the vines from season to season and vines treated with sodium molybdate in consecutive seasons had less molybdenum in the petioles than vines treated for the first time.
• Soil analysis alone may not adequately reflect the tendency of Merlot to develop molybdenum deficiency.
• Molybdenum may accumulate in the soil beneath vines sprayed with sodium molybdate. The consequences of molybdenum accumulation for soil and vine health are unknown.
67 Chapter 5 Effects of sodium molybdate sprays on the vegetative growth of Merlot
5.1 Introduction
Molybdenum (Mo) deficiency of Merlot has been implicated as the possible cause of the ‘Merlot problem’. The symptoms of this disorder include small leaves that roll downwards; burnt, papery leaf margins; and stunted, highly lateralised shoots that are sometimes zigzagged or distorted in habit and have a dark, inky epidermis (Robinson and Burne 2000). Field experiments at Renmark, South Australia showed that these symptoms could be improved within a few weeks by applying sodium molybdate to the foliage. The results from those experiments were probably instrumental in the commencement of the work by Williams et al. (2004) who later showed that yield of Merlot could also be improved by treating vines with sodium molybdate. The effects of molybdenum on yield of Merlot are discussed in detail in Chapter 6.
Gridley (2003) assessed the effects of molybdenum on field-grown molybdenum deficient Merlot vines by comparing the length of the 5th internode on sprayed and unsprayed vines. In that experiment molybdenum did not significantly affect internode length. When molybdenum was applied to pot-grown Mo-deficient Merlot it had no significant effect on dry mass of those vines compared to the controls (Phillips 2004). Phillips also compared the number of new leaves separated and shoot length in a two-week period between Mo-treatment and analysis of the potted vines and found no significant differences between Mo-treated and untreated vines. In the Williams et al. (2004) study, in which sodium molybdate was applied to Merlot vines at three field sites over three seasons, vegetative growth was not measured.
Despite previous reports that Mo-treatment had no effect on vegetative growth of Merlot, in the present study it was considered important to measure vegetative components and relate them to yield in order to assess potential effects on vine balance. Moreover, it was desirable to monitor potential early-season residual effects that were not necessarily apparent in tissue nutrient analysis at MELS 25 (Chapter 4).
68 5.1.1 Aims
The aims of this experiment were to:
• Monitor the effects of Mo-treatment on the vegetative growth of Merlot in the season that the treatments were applied.
• Assess the effects of Mo-treatment on vine balance
• Quantify any residual effects of Mo-treatment on vine growth in the season after treatment.
5.2 Materials and methods
5.2.1 Experimental design
All measurements were performed on experimental vines in the Adelaide Hills and at McLaren Vale. Details of the treatments and experimental design are provided in Chapter 3.
5.2.2 Phenological data
Phenological stages were recorded for the modified E-L stages (MELS): 4 – budburst, 12 – shoots 10 cm, 19 – start of flowering, 25 – 80% caps off, and 27 – setting (Coombe 1995). In 2004-05 phenology in the Hills was recorded on control vines only.
5.2.3 Veraison canopy measures
At veraison in 2003-04, canopy measurements were performed on three vines per replicate in the Hills and at McLaren Vale. In 2004-05 canopy measurements were performed on one vine per replicate. The following measurements were performed on each vine:
• Cordon length • Number of shoots per vine • Mean length of 10 randomly selected shoots per vine a • Mean number of main leaves per shoot a • Mean number of lateral leaves per shoot b • Mean main leaf area b • Mean lateral leaf area a Counted on the 10 randomly selected shoots per vine b Mean leaf area was calculated from seven randomly selected main and lateral leaves per vine. Leaves were scanned with an Area Meter AM200 from ADC BioScientific Ltd. to
69 determine leaf area. Due to the absence of significant differences in leaf area in 2003-04, leaf area was not measured in 2004-05.
Total leaf area per vine was calculated using the following formula:
Total leaf area/vine = # shoots/vine x ((# main leaves/shoot x area per main leaf) + (# lateral leaves/shoot x area/lateral leaf))
5.2.4 Pruning weight
Pruning weight was measured when the vines were dormant in 2004 at McLaren Vale and in 2005 in the Adelaide Hills and at McLaren Vale. One vine from each replicate was pruned to two-node spurs and the prunings were collected and weighed. The average cane weight was calculated by dividing the weight of the prunings by the number of canes pruned.
5.2.5 Yield to pruning weight ratio
The ratio of yield to pruning weight (Y/P), referred to as the Ravaz Index (Champagnol 1984), was used as an indicator of balance between fruit and vegetative growth. This was calculated by dividing the yield per vine (Chapter 3 and Chapter 6) by the total pruning weight per vine in the following winter.
5.2.6 Observations at budburst
In 2004 general observations of vine development were made during routine inspections of the experiments. In 2005 the phenology of the vines was assessed at McLaren Vale. On one vine from each replicate each bud was assigned an appropriate growth stage. The proportions of buds on each vine were calculated for each of the following categories:
• < MELS 5 (rosette of leaf tips visible)
• MELS 7-9 (1-3 leaves separated)
• > MELS 11 (4 leaves separated)
70 5.3 Results
None of the vines in the Hills or at McLaren Vale displayed symptoms of molybdenum deficiency or the ‘Merlot problem’.
5.3.1 Phenological data
Vines at McLaren Vale reached budburst and flowering earlier than those in the Adelaide Hills in both 2003-04 and 2004-05. In 2004-05, vines reached MELS 12 earlier than in 2003- 04 at both sites (Table 5.1).
In 2003-04 there were no visible differences in phenology between the controls and Mo- treated vines in the Hills and at McLaren Vale. In spring 2004-05, vines in the new block at McLaren Vale were phenologically the same as the controls in the old block. However, vines in the old block that had been treated with rates 1 and 2 exhibited delayed development compared to the controls. This delay of 4 days persisted until MELS 27 (Table 5.1).
Table 5.1.Phenological data for Merlot vines at experimental sites in the Adelaide Hills and at McLaren Vale in 2003-06. Stage 4 Stage 12 Stage 19 Stage 25 Stage 27 Budburst Shoots 10 cm Flowering begins 80 % caps off Setting
Adelaide Hills 2003-04 29 Sept. 3 Nov. 03 28 Nov. 03 9 Dec. 03 14 Dec.
2004-05 Rate 0 26 Sept. 14 Oct. 04 24 Nov. 04 29 Nov. 04 2 Dec. 04
McLaren Vale 2003-04 12 Sept. 14 Oct. 03 12 Nov. 03 20 Nov. 03 24 Nov.
2004-05 New block 12 Sept. 30 Sept. 04 29 Oct. 04 12 Nov. 04 20 Nov.
Old block Rate 0 12 Sept. 30 Sept. 04 29 Oct. 04 12 Nov. 04 20 Nov. Rate 1 16 Sept. 9 Oct. 04 1 Nov. 04 16 Nov. 04 20 Nov. Rate 2 16 Sept. 9 Oct. 04 1 Nov. 04 16 Nov. 04 20 Nov.
5.3.2 Veraison canopy measurements
In 2003-04 in the Hills, rate 1-treated vines had significantly shorter shoots and fewer main leaves than the controls and vines treated with rate 2. Leaf area was not significantly affected 71 by Mo-treatment. Mo-treatments applied in the previous season (2003-04) did not affect any vine canopy components measured in the Hills in 2004-05 (Table 5.2).
Table 5.2. Canopy measures of Merlot vines performed at veraison in the Adelaide Hills. Vines were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage 12 (shoots 10 cm) in 2003-04 and were unsprayed in 2004-05. Different letters within a column indicate significant differences (p<0.05) between treatments within a season. # Shoots / Shoot # Main leaves # Lateral Main leaf Lateral leaf Total calculated Hills m length (cm) / shoot leaves / shoot area (mm2) area (mm2) leaf area / vine (m2)
2003-04 Rate 0 26 a 100 a 17 a 7 a 14529 a 2743 a 7.04 a Rate 1 29 a 77 b 14 b 5a 13942 a 2766 a 6.22 a Rate 2 28 a 101 a 17 a 6 a 14600 a 2525 a 5.87 a
2004-05 Rate 0 26 a 101 a 19 a 12 a - - - Rate 1 25 a 95 a 18 a 9 a - - - Rate 2 27 a 97 a 18 a 10 a - - -
At McLaren Vale in 2003-04 shoots on vines treated with rate 1 were significantly longer than shoots on controls and vines treated with rate 2. However, there were no significant differences between treatments for any other canopy components measured (Table 5.3).
In the new block at McLaren Vale in 2004-05, vines treated with rates 1 and 2 tended to have fewer main leaves per shoot compared to the control; however this was only significant for rate 1. Mo-treatment did not affect any of the other canopy components measured in the new block (Table 5.3).
In the old block Mo-treated vines tended to have fewer shoots per metre compared to the controls. Mo-treated vines also had significantly fewer main and lateral leaves per shoot; however there was no significant difference in shoot length between treatments (Table 5.3).
72 Table 5.3 Canopy measures of Merlot vines performed at veraison at McLaren Vale. In 2003-04 vines were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm). Vines in the new block were sprayed at MELS 12 in 2004-05 only and vines in the old block were sprayed at MELS 12 in 2003-04 and again in 2004-05. Different letters within a column indicate significant differences (p<0.05) between treatments within a season or experiment. McLaren # Shoots / Shoot # Main leaves # Lateral Main leaf Lateral leaf Total calculated 2 2 2 Vale m length (cm) / shoot leaves / shoot area (mm ) area (mm ) leaf area / vine (m )
2003-04 Rate 0 21 a 91 a 19 a 18 a 12932 a 2615 a 6.09 a Rate 1 20 a 103 b 21 a 19 a 14802 a 2858 a 7.24 a Rate 2 21 a 91 a 20 a 12 b 14821 a 3016 a 6.70 a
2004-05 New block Rate 0 18 a 99 a 20 a 17 a - - - Rate 1 19 a 86 a 18 b 17 a - - - Rate 2 20 a 91 a 18 ab 14 a - - -
2004-05 Old block Rate 0 23 a 97 a 23 a 21 a - - - Rate 1 20 ab 99 a 20 b 15 b - - - Rate 2 17 b 89 a 20 b 15 b - - -
5.3.3 Pruning weight and vine balance
In the Hills there was no significant difference in either pruning weight per vine or mean cane weight between treatments (Table 5.4).
Table 5.4. Dormant vine measures performed on Merlot vines in the Adelaide Hills in 2005. Yield per vine was measured at harvest. Vines were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage 12 (shoots 10 cm) in 2003-04. Yield / vine (kg) # Canes / vine Pruning weight / Average cane Y/Pa Hills vine (kg) weight (g)
Rate 0 3.7 a 28 a 0.62 a 22 a 6.0 a Rate 1 3.4 a 28 a 0.69 a 24 a 6.1 a Rate 2 4.5 a 32 a 0.68 a 21 a 7.1 a a Y/P is the ratio of yield to pruning weight
At McLaren Vale in 2003-04 pruning weight on Mo-treated vines tended to be lower than the controls. This was a result of fewer canes per vine, which also weighed less than the control. The Y/P of Mo-treated vines was significantly higher than the control, a function of both a significantly higher yields and lower pruning weights (Table 5.5).
In the new block at McLaren Vale in 2004-05, the pruning weight from Mo-treated vines was significantly lower than the controls, a function of significantly lighter canes. The lower
73 pruning weight and significantly higher yield on Mo-treated vines gave a significantly higher Y/P (Table 5.5).
In the old block at McLaren Vale pruning weight tended to be lower on Mo-treated vines compared to the control due to fewer and lighter canes. The Y/P was significantly higher on Mo-treated vines compared to the controls (Table 5.5).
Table 5.5. Dormant vine measures performed on Merlot vines at McLaren Vale in 2004 and 2005. Yield per vine was measured at harvest. In 2004 vines had been treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in 2003-04. In the new block vines were sprayed at MELS 12 in 2004-05 only and vines in the old block were sprayed at MELS 12 in 2003-04 and again in 2004- 05. Yield / vine (kg) # Canes / vine Pruning weight / Average cane Y/Pc McLaren Vale vine (kg) weight (g)
2004 Rate 0 3.11 a 35 a 0.91 a 27 a 3.5 a Rate 1 6.29 b 31 a 0.75 a 25 a 8.6 b Rate 2 5.94 b 30 a 0.65 a 22 a 9.7 b
2005 New block Rate 0 4.6 a 29 a 0.90 a 31 a 5.3 a Rate 1 6.0 b 28 a 0.52 b 19 b 11.8 b Rate 2 6.9 b 28 a 0.64 b 23 b 11.1 b
Old block Rate 0 4.0 a 30 a 0.81 a 27 a 5.4 a Rate 1 6.8 ab 26 a 0.69 a 27 a 10.1 b Rate 2 7.3 b 26 a 0.60 a 23 a 12.3 b c Y/P is the ratio of yield to pruning weight
5.3.4 Observations at budburst
In early spring 2004 (the spring following the first season of Mo-treatment) there were visible differences in vine growth between the treatments in the Hills (Figure 5.1) and at McLaren Vale (Figure 5.2). At McLaren Vale the control vines were at MELS 9 (2-3 leaves separated) while the Mo-treated vines were at MELS 4-5 (first leaf or rosette of leaf tips visible). During the next visit the differences between treatments were less obvious; however the delay in their development persisted until flowering.
74
Figure 5.1.Delayed budburst on Merlot vines (inside the orange oval) in the Adelaide Hills on 18 November 2004. These vines were treated with rate 2 molybdenum sprays in spring 2003. Note the stage of growth of the unsprayed vines in the same row outside the oval for comparison.
Figure 5.2. Delayed budburst on Merlot vines (inside the orange oval) at McLaren Vale on 15 September 2004. These vines were sprayed with rate 1 molybdenum in spring 2003. Note the stage of growth of the unsprayed vines in the same row outside the oval and those in the buffer row in the background for comparison.
In 2005 a similar pattern of delayed development was noted at McLaren Vale. In the new block the controls had a significantly higher proportion of nodes that had reached MELS 7-9 compared to Mo-treated vines which had a higher proportion of nodes at stages less than MELS 5. In the old block the Mo-treated vines had a significantly higher proportion of nodes 75 at stages less than MELS 5 compared to the control which tended to have more nodes at MELS 7-9 (Table 5.6).
Table 5.6. Growth stage observations made in September 2005 on Merlot vines at McLaren Vale. Vines in the new block were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in 2004-05 only. Vines in the old block were treated at MELS 12 in 2003-04 and again in 2004-05. Different letters after the values in rows represent significant differences between treatments within an experiment (p<0.05). Figures below the columns show the proportions of the nodes < MELS 5 (purple), at MELS 7-9 (red) and >/= MELS 11 (yellow). New block Old block Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2 % Nodes < MELS 5 62 a 84 a 85 a 48 a 77 b 83 b
% Nodes @ MELS 7- 36 a 16 b 15 b 44 a 22 ab 17 b 9
% Nodes > MELS 11 1 a 0 a 0 a 8 a 1 a 0 a
Total # nodes 31 a 31 a 31 a 37 a 31 a 32 a
5.4 Discussion
The effects of Mo-treatment on the vegetative growth of the vines were not apparent at veraison in either 2003-04 or in 2004-05. In the Hills and at McLaren Vale the shoot lengths and leaf areas fell within the moderate vigour range suggested by Smart and Robinson (1991). The small but significant differences in shoot length and consequent leaf numbers observed at both sites are not thought to be of any practical significance.
The vegetative growth of vines in the Hills was not affected by Mo-treatment probably because they were not deficient in molybdenum prior to treatment (see Chapter 4). However, at McLaren Vale where the controls were Mo-deficient, pruning weight tended to be lower for vines treated with both rates 1 and 2 in 2003-04 and in 2004-05. This was significant when the mean cane weight was also significantly lower (2004-05 new block). However, the mean cane weight was considered to be in the moderate vigour range in all cases (Smart and Robinson 1991).
76 Smart and Robinson (1991) suggest that vines with ‘moderate’ vigour have an Y/P between 5- 10; below this vines are considered to have ‘high’ vigour. The higher Y/P of Mo-treated vines shifted them from the high to the moderate vigour categories or, put another way, the control vines were out of balance. The higher yield and lower pruning weight on Mo-treated vines improved vine balance, albeit at the higher end of the scale.
5.4.1 Effects of Mo-treatment on vine vegetative growth in the season after spraying
This is the first report that sodium molybdate foliar treatment can delay budburst and vine development in the spring following treatment and that Mo-treatment decreases pruning weight.
The delay in budburst observed on vines in the spring after the first season of experimentation was at first thought to be a consequence of the improved yield on those vines. Yield is negatively correlated with vine carbohydrate reserves in the following spring (Smith and Holzapfel 2003). However, Mo-treatment did not affect yield in the Hills, and yet a similar delay in vine development was also observed in the Hills. Vines with low reserves also tend to have lower yield in the following season. However the Mo-treated vines had improved yield compared to the controls (Chapter 6).
An alternative mechanism to explain the effect of Mo-treatment on spring growth of vines is not available at this time and is outside the scope of the present study.
Delayed spring growth may be considered beneficial in some grapegrowing regions of Australia. In regions where vines are at risk of damage by spring frost, Mo-treatment in the previous season may reduce the risk of yield loss. Alternatively Mo-treatment may be used to delay vine development such that flowering occurs later in the season when temperatures are more favourable (as seen in the Hills and discussed in Chapter 4). However, as discussed in Chapter 4, the long-term effects of Mo-treatment on vines and soil are unknown and therefore Mo-treatment for this purpose should be applied with caution.
77 5.5 Conclusions
• The effects of Mo-treatment on the vegetative growth of Merlot were not apparent at veraison.
• Dormant canopy measurements revealed a tendency for lower pruning weights on vines treated with sodium molybdate. In combination with improved yield on Mo- treated vines, the lower pruning weight caused a shift in the yield to pruning weight ratio such that Mo-treated vines were better balanced than Mo-deficient controls.
• In the spring following Mo-treatment budburst was delayed on the vines that received sodium molybdate sprays in the previous season. The delay in vine development continued until flowering.
78 Chapter 6 Effects of sodium molybdate sprays on the yield components of Merlot
6.1 Introduction
Grapevine yield comprises a number of components as outlined in the equations below.
Yield / vine = no. bunches per vine x total weight of bunches per vine
No. bunches / vine (fruitfulness) = no. shoots per vine x no. bunches per shoot
Bunch weight = (no. berries per bunch x berry weight) + rachis weight
Berries / bunch = no. flowers per inflorescence x percent fruitset
% Fruitset = 100 x (no. berries per bunch / no. flowers per inflorescence)
There have been numerous attempts to show the yield benefits of molybdenum (Mo) application to grapevines (Hátle and Mičulka 1971; Veliksar 1977; Misa 1977, Dobrolyubskii et al. 1981; Strakhov & Chazova 1981; Staudt & Kassemeyer 1981; Ma et al. 1992). However, in most cases, there has been no clear indication as to which components of yield are affected. Most recently, Gridley (2003), Phillips (2004) and Williams et al. (2004) studied the effects of molybdenum on yield of Merlot under Australian conditions. In each study a ‘standard’ rate of sodium molybdate (300 g/ha) was applied to the vines twice in spring. In each of the studies, yield increased when Mo-deficient vines were treated with sodium molybdate. The improvement in yield was attributed to both heavier bunches and, in some cases, more bunches per vine. The increase in bunch weight was reported to be a function of a higher proportion of larger berries, or put another way, the effects of millerandage (and ‘hen and chicken’) were reduced. In the previous research, fruitset was either not quantified (Williams et al. 2004) nor found to be a major contributor to increased yield (Gridley 2003; Phillips 2004).
In 2003, Gridley reported that when petiolar molybdenum concentration was high in Merlot, yield might be detrimentally affected. To date, guidelines for applying molybdenum preparations to treat Mo-deficiency in grapevines are anecdotal. There are also no published data regarding the effects of successive seasons of molybdenum treatment on yield. Despite this, Australian grapegrowers are being advised to apply molybdenum preparations to Merlot on a seasonal basis.
79 In the current study molybdenum sprays were applied to own-rooted Merlot vines (clone D3V14) in commercial vineyards in the Adelaide Hills in 2003-04 and at McLaren Vale in 2003-04 and 2004-05. Yield components were measured at both sites over the two seasons.
6.1.1 Aims
The aims of the experiment were to:
• Quantify the effects of Mo-deficiency on the yield components of Merlot and determine which components were most affected by Mo-deficiency.
• Investigate the effects of applying double the recommended rate of sodium molybdate on the yield components of Merlot.
• Examine the effects of consecutive seasons of sodium molybdate application on yield of Merlot.
• Identify the critical level of molybdenum at flowering for optimum yield of Merlot.
6.2 Materials and methods
6.2.1 Experimental design
Details of the experimental sites, spray treatments and the experimental design are provided in Chapter 3.
6.2.2 Nutritional analysis
Details of the sampling procedures and analyses performed are outlined in Chapter 4.
6.2.3 Yield components
The procedures for the measurement and calculation of fruitset, ovary retention, berry number and berry weight are provided in Chapter 3. All yield component data were collected from one vine in each replicate. Fruitset, ovary retention, berry number and berry weight were measured on five bunches per replicate.
6.2.4 Total yield
Total yield per vine was calculated using two different methods. Method A (below) involved multiplying the number of bunches per vine by the average bunch weight of the five bagged bunches per vine. Method B (below) involved multiplying the bunch number per vine by the 80 average weight of five randomly selected unbagged bunches from the proximal position on a shoot and five from the distal position. This latter method accounted for potential differences between the weight of bagged versus non-bagged bunches and the disparate weight of proximal and distal bunches.
Method A Total yield = no. bunches per vine x average bunch weight of 5 bagged bunches
Method B Total yield = no. bunches per vine x average bunch weight of (5 proximal bunches + 5 distal bunches)
6.2.5 Seed number
Seeds were only counted when significant differences in berry weight were measured. All of the berries from three of the five bagged bunches per vine were dissected and the seeds removed. The seeds from each bunch were then washed and dried with paper towel to remove any remaining pulp before being counted and weighed. The total number of seeds per bunch was divided by the number of berries per bunch to give the mean number of seeds per berry.
Average no. seeds per berry = no. seeds per bunch / no. berries per bunch
6.2.6 Berry shrivel
In 2003-04 berry shrivel was assessed by visual appraisal just prior to commercial harvest. A small sample of bunches was collected from the Mo-treated and Mo-deficient vines and photographed. However, a lack of time prohibited the collection of any data. In 2004-05 bunches were monitored visually as harvest approached; however no differences were observed between treatments.
In 2005-06 berry weight was measured on bunches from the Mo-pollination experiment (experimental design described in Chapter 1) on two separate dates using the method described in Chapter 3. Four bagged bunches from each replicate were collected on February 1 and eight weeks later, on March 28, five bunches were harvested at random from each replicate.
81 6.3 Results
6.3.1 Nutritional analysis
In the Hills, vines from all treatments had adequate levels of molybdenum at flowering (Chapter 4). However, at McLaren Vale the control vines were deficient in molybdenum and spraying vines with rates 1 and 2 molybdenum sprays increased the petiolar molybdenum concentration to above the suggested deficiency range (0.05-0.09 mg/kg; Williams et al. 2004). A comprehensive review of the molybdenum status of vines in this experiment can be found in Chapter 4.
6.3.2 Fruitset
In the Hills, flower number per inflorescence and the number of berries per bunch and fruitset were not affected by Mo-treatment, either in the season that it was applied or in the season after Mo-treatment (Table 6.1 and Table 6.2).
At McLaren Vale, flower number per inflorescence was not affected by Mo-treatment in either 2003-04 or 2004-05. However, there was a significant increase in the number of berries per bunch and hence percent fruitset with Mo-treatment. In 2003-04 and in the ‘new’ block in 2004-05, fruitset improved by 40% when vines were treated with rate 1. In the ‘old’ block in 2004-05, fruitset increased by 37%. There was no significant difference in fruitset between rates 1 and 2 in any of the experiments (Table 6.1 and Table 6.2).
Fewer berries per bunch gave bunches from the control treatment a loose appearance compared to the well-filled bunches from the Mo-treated vines (Figure 6.1).
82 Table 6.1.Yield components on Merlot vines treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in the Adelaide Hills and at McLaren Vale in 2003-04. Different letters after the values represent significant differences between treatments within an experiment (p<0.05). # Flowers / # Berries Fruitset Berry Seeds / Bunch # Bunches / Total yield / Total yield / 2003-04 inflorescence / bunch (%) weight (g) berry weight (g) vine vine d (kg) vine e (kg) Adelaide Hills Rate 0 524 a 145 a 28 a 0.88 a - 141 a 81 a 11.42 a 11.13 a Rate 1 523 a 140 a 27 a 0.87 a - 130 a 81 a 10.53 a 11.02 a Rate 2 469 a 138 a 31 a 0.93 a - 141 a 77 a 10.86 a 9.56 a
McLaren Vale Rate 0 300 a 51 a 18 a 0.71 a 1.04 a 41 a 60 a 2.46 a 3.11 a Rate 1 337 a 99 b 30 b 0.91 b 1.37 b 95 b 59 a 5.61 b 6.29 b Rate 2 367 a 110 b 31 b 0.92 b 1.52 b 103 b 56 a 5.77 b 5.94 b d Yield calculated using Method A e Yield calculated using Method B
Table 6.2. Yield components on Merlot vines in 2004-05 in the Adelaide Hills and at McLaren Vale. Vines in the Hills were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in 2003-04 only. Vines in the new block at McLaren Vale were treated at MELS 12 in 2004- 05 only and vines in the old block at McLaren Vale were treated at MELS 12 in 2003-04 and again in 2004-05. Different letters after the values represent significant differences between treatments within an experiment (p<0.05). # Flowers / # Berries / Fruitset Berry Bunch # Bunches Total yield Total yield / 2004-05 inflorescence bunch (%) weight (g) weight (g) / vine / vine d (kg) vine e (kg) Adelaide Hills Rate 0 240 a 74 a 31 a 0.93 a 72 a 43 a 3.10 a 3.72 a Rate 1 243 a 82 a 35 a 0.99 a 93 a 37 a 3.44 a 3.43 a Rate 2 267 a 81 a 32 a 0.97 a 84 a 48 a 4.03 a 4.49 a
McLaren Vale New block Rate 0 303 a 76 a 24 a 1.00 a 80 a 57 a 4.56 a 4.60 a Rate 1 316 a 120 b 40 b 0.97 a 120 b 57 a 6.84 b 6.04 b Rate 2 333 a 119 b 36 b 0.99 a 123 b 57 a 7.01 b 6.92 b
Old block Rate 0 351 a 84 a 24 a 0.97 a 83 a 61 a 5.06 a 4.04 a Rate 1 315 a 121 b 38 b 1.05 a 130 b 56 a 7.28 a 6.80 ab Rate 2 316 a 130 b 43 b 1.06 a 141 b 56 a 7.90 a 7.28 b d Yield calculated using Method A e Yield calculated using Method B
83 A1 A2
B
C
Figure 6.1. Extreme examples of poor fruitset on Merlot bunches from control vines in 2003-04 (A1), a typical control bunch in 2004-05 (A2) and typical bunches from vines sprayed with rate 1 (B) and rate 2 (C) molybdenum sprays at McLaren Vale in 2003-04.
84 6.3.3 Ovary retention
Bunches from control vines in the new block at McLaren Vale tended to have more live green ovaries (LGOs) per bunch compared to those from the Mo-treated vines. The difference in LGOs between the control and rate 1 was not significant, however, there was a significant difference between the control and rate 2.
In the old block, control bunches had significantly more LGOs than bunches from Mo-treated vines. There was no significant difference in the percentage of ovaries retained on bunches (description provided in Chapter 3) at harvest between treatments at McLaren Vale in 2004- 05 (Table 6.3).
Table 6.3. Flower number per inflorescence, number of berries per bunch, number of live green ovaries (LGOs) per bunch, percent fruitset and percent ovary retention on bunches from Merlot vines at McLaren Vale in 2004- 05. Vines in the new block were treated with rates 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in 2004-05 only and vines in the old block received molybdenum sprays at MELS 12 in 2003-04 and again in 2004-05. Different letters after the values represent significant differences between treatments within an experiment (p < 0.05). # Flowers per # Berries per # LGOs per Fruitset Ovary
inflorescence bunch bunch (%) retention (%)
New block Rate 0 303 a 76 a 91 a 24 a 57 a Rate 1 316 a 120 b 77 ab 40 b 63 a Rate 2 333 a 119 b 58 b 36 b 54 a
Old block Rate 0 351 a 84 a 113 a 24 a 55 a Rate 1 315 a 121 ab 64 b 38 b 58 a Rate 2 316 a 130 b 61 b 43 b 61 a
6.3.4 Berry weight
Mo-treatment did not affect berry weight in the Hills in 2003-04 or in 2004-05. At McLaren Vale in 2003-04 berries from Mo-treated vines were significantly heavier than the controls. The increase in berry weight was a result of a significant increase in seeds per berry (Table 6.1).
In 2004-05 there were no significant differences in berry weight between treatments at McLaren Vale (Table 6.2). There was no significant difference in the proportion of chicken
85 berries per bunch between treatments at McLaren Vale. Chicken berries contributed less than 1% to the total berry weight per bunch for all treatments (Table 6.4).
Table 6.4. The contributions of chicken berries per bunch as a function of total berry number per bunch, and chicken berry weight as a proportion of the total berry weight per bunch on Merlot at McLaren Vale in 2004-05. Vines in the new block were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in 2004-05 only and those in the old block were treated at MELS 12 in 2003-04 and again in 2004-05. Different letters after the values indicate significant differences between treatments within an experiment (p < 0.05). # Chicken berries as a % of Chicken berry wt. as a % of total berries per bunch total berry wt. per bunch
New block Rate 0 9 a 0.96 a Rate 1 7 a 0.39 ab Rate 2 8 a 0.34 b
Old block Rate 0 9 a 0.94 a Rate 1 7 a 0.40 a Rate 2 6 a 0.31 a
6.3.5 Bunch number per vine
There were no significant differences in bunch number per vine between treatments in the Hills or at McLaren Vale in 2003-04 or in 2004-05 (Table 6.1 and Table 6.2).
6.3.6 Total yield
In the Hills total yield per vine was not affected by molybdenum treatment in 2003-04 or 2004-05. At McLaren Vale in 2003-04, Mo-treatment significantly increased yield compared to the controls. In 2004-05 Mo-treated vines in the new block also had significantly higher yields than the controls. Vines in the old block tended to have higher yields than the controls, however they were not significantly different (Table 6.2).
6.3.7 The relationship between molybdenum concentration in the vines at flowering and percent fruitset on bunches from Mo-treated vines
The strongest relationships between molybdenum concentration in the vines at MELS 25 (80% flowering) and fruitset were found when data from the same site was analysed within the same season. The strongest of these relationships was found at McLaren Vale in 2004-05 (Figure 6.2, Figure 6.3 and Figure 6.4). The tissue type that best reflected the relationship
86 between molybdenum concentration and percent fruitset was the petioles (r2 = 0.786) (Figure 6.2). Percent fruitset increased as the concentration of molybdenum increased in the petioles, shoot tips and inflorescences. However, when the concentration of molybdenum exceeded 11 mg/kg in the petioles, fruitset tended to decrease (Figure 6.2).
50
40
R2 = 0.786 (degrees) et s 30 Fruit
20 0 5 10 15 20 Petiolar Mo (mg/kg)
Figure 6.2. The relationship between molybdenum concentration in the petioles (mg/kg) at modified E-L stage (MELS) 25 (80% flowering) and fruitset (arc sine transformed) on bunches of Merlot vines treated with rates 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2004-05 at McLaren Vale.
50
R2 = 0.7069
40 (degrees) et s 30 Fruit
20 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 Shoot tip Mo (mg/kg)
Figure 6.3. The relationship between molybdenum concentration in the shoot tips (mg/kg) at modified E-L stage (MELS) 25 (80% flowering) and fruitset (arc sine transformed) on bunches of Merlot vines treated with rates 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2004-05 at McLaren Vale.
87 50
40 (degrees)
et 2
s R = 0.7305 30 Fruit
20 0 5 10 15 20
Inflorescence Mo (mg/kg)
Figure 6.4. The relationship between molybdenum concentration (mg/kg) in the inflorescences at modified E-L stage (MELS) 25 (80% flowering) and fruitset (arc sine transformed) on bunches of Merlot vines treated with rates 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2004-05 at McLaren Vale.
6.3.8 Berry shrivel
During a routine inspection of the experimental block at McLaren Vale on the day prior to commercial harvest in 2003-04, berries on bunches from Mo-deficient vines appeared to be more shrivelled than those from the Mo-treated vines (Figure 6.5). In 2004-05 no differences were observed between berries from the different treatments.
In 2006, berry shrivel was first observed on March 14, however, there were no visible differences between Mo-treated and Mo-deficient vines. Two weeks later, on March 28, the average berry weight from the control vines tended to be less than that on the Mo-treated vines (1.03 g compared to 1.10 g respectively), however this was not significant. Comparing the berry weights measured on March 28 to those measured on February 1, on average the weight of the Mo-treated berries decreased 7.6% due to shrivel compared to a decrease of 8.4% for the controls (Figure 6.6). There was no significant interaction between Mo-treatment and berry weight on either date.
88
Figure 6.5. Typical bunches from rate 0 (control) (left) and rate 1 (right) Mo-treated Merlot vines at McLaren Vale collected on March 3 2004. Berries from the Mo-deficient vines appeared shrivelled compared to berries from Mo-treated vines which appeared turgid and in good condition.
1.40
1.20
1.00
0.80
0.60 Berry wt (g) 0.40
0.20
0.00 Rate 0 Rate 1
Figure 6.6. Mean berry weights from Merlot bunches collected on February 1 (■) and March 28 (■) from vines treated with rate 0 (control) and rate 1 Mo sprays at McLaren Vale in 2005-06.
89 6.4 Discussion
The increase in total yield per vine on vines treated with a standard rate of molybdenum compared to Mo-deficient controls at McLaren Vale is in agreement with the findings of Gridley (2003), Phillips (2004) and Williams et al. (2004). The application of molybdenum sprays did not affect yield in the Hills, probably because those vines were not deficient in molybdenum at flowering (Chapter 4).
6.4.1 Fruitfulness
In the current study, molybdenum foliar spray did not affect the number of bunches per vine in the season that it was applied. This result was as expected since the inflorescence primordia were initiated in the spring prior to Mo-treatment, hence it could not be responsible for any lack of production of bunches. However, Williams et al. (2004) reported that Mo-treatment increased bunch numbers at one of their 3 experimental sites in the second year of experiments. In that experiment, no background was provided about pruning uniformity of the experimental vines and it is possible that the increase in bunches per vine resulted from more shoots per vine brought about by a higher node number per vine retained at pruning. Alternatively, bunches on Mo-deficient vines may have been affected by early bunch stem necrosis or ‘filage’ (Gridley 2003). Neither of these disorders was observed in the current study.
6.4.2 Bunch weight
The significant increase in total yield on vines treated with molybdenum at McLaren Vale was due to a significant increase in bunch weight. This was a result of a higher percent fruitset and, in 2003-04, an increase in mean berry weight on bunches from Mo-treated vines. The higher berry weight, was not due to excessive numbers of small or seedless berries. In 2004- 05 the heavier bunches on Mo-treated vines was a function of improved fruitset alone.
The lack of significant differences in total yield between treatments in the old block in 2004- 05 was due to a high level of variability in bunch weights across all treatments.
6.4.3 Millerandage
In both seasons of the current study, berries on bunches from Mo-deficient vines appeared uniform in size and development and bunches had very few chicken berries. Millerandage and
90 hen and chicken are commonly used synonymously (May 2004). Millerandage refers to the condition where berries on bunches are arrested in development. This may include ‘shot’ berries (or LGOs) and chicken berries (Royal Commission on Vegetable Products 1891; Chancrin 1908; Sharma et al. 1995; Colin et al. 2002). However, hen and chicken infers the presence of a high proportion, or at least presence of, seedless chicken berries on a bunch. Therefore, bunches displaying hen and chicken are affected by millerandage, however, bunches affected by millerandage do not necessarily display hen and chicken (see Chapter 1 for a more detailed discussion).
In the strictest sense, there was a significant difference in millerandage between bunches from Mo-treated and Mo-deficient vines in the current study, given that there were differences in the number of LGOs on bunches at harvest. Bunches from Mo-deficient vines may have been afflicted by millerandage however they did not suffer from hen and chicken.
In the studies of Gridley (2003), Phillips (2004) and Williams et al. (2004), bunches from Mo- deficient vines had higher proportions of chicken berries. This suggests that the ovaries were arrested in development at a later stage than in the current study. This may be due to differences in the susceptibility of clones to millerandage (Gridley studied clones 2093 and 2315, and Phillips studied clone 2093) or it is possible that the mechanism for ovary development was different. This is discussed in further detail in Chapter 7.
6.4.4 Effects of consecutive seasons of Mo-treatment on yield components of Merlot
The effect of sodium molybdate application in consecutive seasons on Merlot yield was the same as the effect gained by one season of Mo-treatment. Yield components that are determined in the season prior to their emergence, that is, inflorescence number per vine and the degree of branching of the inflorescences, were not affected. Percent fruitset on bunches from the old block was the same as in the new block and the degree of improvement in fruitset on Mo-treated bunches compared to control bunches was also similar for the two blocks. These data demonstrate that Mo-treatment does not have a cumulative effect on yield when applied to Mo-deficient vines in consecutive seasons.
91 6.4.5 Effects of high molybdenum concentration on yield
The results presented here are in agreement with previous studies that reported that treatment of Mo-deficient Merlot vines with sodium molybdate can increase yield. Also, application of sodium molybdate to vines that already have an adequate level of molybdenum, and increasing the standard application rate two-fold, did not detrimentally affect yield either in the season that it was applied or in the following season.
Based on the relationship between petiolar molybdenum concentration and fruitset it appears that at McLaren Vale, fruitset tended to decrease at high molybdenum concentrations. However, when concentration of molybdenum was highest, fruitset was higher on average than fruitset on bunches from Mo-deficient vines (Figure 6.7). This was also reported by Mulder (1954, cited in Gupta 1997) who found that applying a higher than normal rate of molybdenum to wheat gave a smaller yield response than for the standard rate, but the yield was still greater than for the untreated controls.
As discussed in Chapter 4, petiole analyses must be interpreted with caution, because the petioles were subject to possible contamination from the sodium molybdate sprays. The relationship between fruitset and the concentration of molybdenum in the shoot tips does not show the decline in fruitset at high molybdenum concentrations (Figure 6.8). These data suggest that high levels of molybdenum are not detrimental to fruitset.
50
40 (degrees) et s 30 Fruit
20 0 5 10 15 2 0
Petiolar Mo (mg/kg)
Figure 6.7. The relationship between petiolar molybdenum concentration (mg/kg) at modified E-L stage (MELS) 25 (80% flowering) and fruitset (arc sine transformed) on bunches from Merlot vines treated with rates 0 (control) (○), rate 1 (●) and rate 2 (●) molybdenum sprays at MELS 12 (shoots 10 cm) in 2004-05 at McLaren Vale. Points inside the red circle represent those that have lower fruitset values associated with high molybdenum concentration in the petioles.
92
50
40 (degrees) et s 30 Fruit
20 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70
Shoot tip Mo (mg/kg)
Figure 6.8. The relationship between molybdenum concentration (mg/kg) in the shoot tips at modified E-L stage (MELS) 25 (80% flowering) and fruitset (arc sine transformed) on bunches from Merlot vines treated with rates 0 (control) (○), rate 1 (●) and rate 2 (●) molybdenum sprays at MELS 12 (shoots 10 cm) in 2004-05 at McLaren Vale.
6.4.6 What is the critical level of molybdenum for optimum yield of Merlot?
Application of rate 1 or rate 2 sodium molybdate sprays to Mo-deficient vines consistently improved fruitset compared to controls. Considering the relationship between shoot tip molybdenum concentration and fruitset at McLaren Vale in 2003-04 and in 2004-05, molybdenum application reduced the variability in fruitset (coefficient of variation [CV] = 10%) compared to fruitset on Mo-deficient controls (CV = 19%) (Figure 6.9). That is, bunches tended to be more uniform in fruitset when adequate molybdenum was available in the vines. Furthermore, in 2003-04 and 2004-05, 98% of bunches from Mo-treated vines at McLaren Vale had fruitset greater than 30%. On the Mo-deficient vines fruitset was less than 30% on 76% of bunches. Put another way, application of molybdenum to Mo-deficient vines increased the likelihood of achieving greater than 30% fruitset.
When the concentration of molybdenum in the shoot tips was greater than 0.1 mg/kg, there was a significant increase in yield — primarily a function of improved fruitset on bunches (Figure 6.9). Given the strong relationship between petiole and shoot tip molybdenum concentrations (Chapter 4), it is suggested that the critical level of molybdenum for optimal fruitset in Merlot is 0.1 mg/kg in the shoot tips at modified E-L stage 25 (80% flowering).
93
50
40
30 (degrees) et s 20 Fruit
10
0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Shoot tip Mo (mg/kg)
Figure 6.9. The relationship between molybdenum concentration (mg/kg) in the shoot tips at modified E-L stage (MELS) 25 (80% flowering) and fruitset on bunches from Merlot vines treated with rates 0 (control) (white), rate 1 (yellow) and rate 2 (red) molybdenum sprays at McLaren Vale. In 2003-04 (■) vines were sprayed at MELS 12 (shoots 10 cm). In the new block (▲) vines were sprayed at MELS 12 in 2004-05 only and in the old block (●) vines were sprayed at MELS 12 in 2003-04 and again in 2004-05. The red lines indicate the suggested critical level of molybdenum (0.1 mg/kg) in the shoot tips to increase fruitset to 30%.
6.4.7 Berry shrivel
Molybdenum has been associated with improved drought resistance of plants by ‘…maintaining a higher degree of hydration of cell colloids’ (Vasil’eva and Startseva 1959, cited in Gupta 1997) and by decreasing the transpiration rate in pea leaves (Zaslonkin 1968, cited in Gupta 1997). The relationship between molybdenum and grape berry shrivel is one that deserves further investigation. Anecdotally, berry shrivel of Merlot is a major problem in Chile (pers. comm., P. Dry 2006) and it has also been observed on low-yielding vines with highly-exposed bunches in Coonawarra, Australia.
Results from recent experiments showed that when sodium molybdate was applied to Shiraz at flowering time it shifts the berry developmental process such that berry shrivel and the resultant changes in berry composition occur later on Mo-treated vines than untreated controls (pers. comm., J. Tilbrook 2006). This may have implications for grapegrowers whose aim is to maintain optimal berry composition while maximising grape yield.
94 6.5 Conclusions
• Treatment of Mo-deficient Merlot vines with sodium molybdate improved bunch weight. This was a function of improved fruitset, not increased mean berry weight as previously reported.
• Molybdenum treatment did not affect bunch number per vine nor did it affect flower number per inflorescence in the season following treatment.
• Treatment of Mo-deficient Merlot vines with sodium molybdate did not have a cumulative effect on yield.
• Treatment of vines with adequate levels of molybdenum did not detrimentally affect yield nor did application of two times the standard rate of sodium molybdate.
• The critical level of molybdenum for optimal fruitset on Merlot is suggested to be 0.1 mg/kg in the shoot tips at modified E-L stage 25 (80% flowering)
95
Chapter 7 Effects of sodium molybdate foliar sprays on flowering and flower characteristics of Merlot
7.1 Introduction
Nutrient availability is one of the critical factors governing the success of flowering and fruitset, either directly or via the concomitant effects of temperature, hormone levels and environmental conditions.
Pollen germination requires nutrient supply from the pollen grain and the surrounding transmitting tissue (Cresti et al. 1975, cited in Cresti & Ciampolini 1999). In maize, molybdenum (Mo) deficiency affects pollen production and viability and also affects enzymatic activity in the pollen (Agarwala et al. 1979). In an in vitro study of grapevine pollen, Ma et al. (1992) reported that for the variety ‘Kyoho’, a higher proportion of pollen germinated on media supplied with sodium molybdate than on Mo-deficient media. They associated the enhanced pollen germination rate with improved fruitset on vines sprayed with sodium molybdate at flowering time.
Pollen tube growth can also be affected by plant hormones, which are thought to affect pollen tube growth indirectly via their control on competition between vegetative and reproductive organs for metabolites (Coombe 1970; Skene 1969, cited in Coombe 1973; Okamoto et al. 2001). As discussed in Chapter 5, application of sodium molybdate to vines in spring did not affect vegetative growth of Merlot in the season that it was applied. Mo-deficiency may, however, have an indirect affect on hormone levels via its effect on plant hormone biosynthesis. In apricots, a relationship has been shown between polyamine (PA) levels and ovule development which is suggested to reduce pollen tube growth (Albuquerque et al. 2006).
As discussed in Chapter 1, pollen tube signalling is governed by complex mechanisms at different points along the route from the stigma to the ovule. Pollen tube signalling may be directly affected by changes in the chemical composition of the style and transmitting tissue, or indirectly via anatomical or cytological changes brought about by Mo-deficiency.
At modified E-L stage 12 (shoots 10 cm), that is prior to Mo-treatment, ovules are undifferentiated cell masses in the placenta, and embryo sacs do not form until just prior to 96 flowering. Various aberrations of ovule development have been associated with poor pollen tube growth and fruitset in grapevines and other fruit crops (Carraro et al. 1979; Vallania et al.1987) including: absence of a functional embryo sac (Pearson 1933; Kassemeyer & Staudt 1982; Reiser & Fischer 1993; Ebadi 1996); degeneration of the embryo sac (Kassemeyer & Staudt 1982; Beppu 1997); degeneration of the egg nucleus (Wang et al. 1993); undeveloped embryo sac (Wang et al. 1993); misshapen ovules (Pearson 1933; Wang et al. 1993); blockages at the placenta or in the funiculus (Polito 1999); callose deposition (or lack of) (Reiser & Fischer 1993; Pimienta & Polito 1982, cited in Polito 1999); asynchronous ovule development (Albuquerque et al. 2006); reduced ovule longevity (Staudt 1986); and ovule abortion (Pratt & Einsett 1961; Wang et al 1993; Barritt 1970).
In Chapter 4, Chapter 5 and Chapter 6 the effects of sodium molybdate application to Mo- deficient Merlot were discussed in detail. Mo-treated vines had a higher concentration of molybdenum in the petioles, inflorescences and shoot tips at flowering time compared to Mo- deficient controls. Mo-treatment did not affect vegetative growth in the season that it was applied, however, yield was significantly higher on Mo-treated vines. Contrary to previous reports, Mo-treatment consistently improved yield of Merlot via its effect on fruitset, not on berry size. Improved fruitset is indicative of a higher proportion of successfully fertilised ovaries that develop into berries, whilst berry size is governed by seed number and thus the number of fertilised ovules per ovary. There are no previous reports describing the mechanism by which Mo-deficiency affects yield of Merlot.
7.1.1 Aims
The aims of this experiment were to:
• Compare the molybdenum concentration in the male and female flower parts in response to foliar application of sodium molybdate to Merlot vines.
• Assess the effects of Mo-treatment on Merlot pollen vitality and pollen tube growth.
• Compare the anatomy of ovules from Mo-deficient and Mo-treated Merlot vines.
97 7.2 Materials and methods
7.2.1 Experimental design
The treatments and experimental designs are described in detail in Chapter 3.
7.2.2 Weather data
Daily weather data was collected using on-site weather stations described in Chapter 3. Daily temperature data was averaged over the period from the beginning of flowering to 80% flowering and during the four-day period of pollen tube observations. Rainfall was totalled over the same periods of time. Due to a malfunction of the Hills weather station in 2004, rainfall data presented was obtained from the Bureau of Meteorology weather station at Lenswood, South Australia which was situated approximately 12 km from the experimental site.
7.2.3 Nutritional analyses
Vegetative tissue In 2003-04 and 2004-05 material was collected from the molybdenum experiments in the Hills and at McLaren Vale. In 2005-06 samples were collected from the molybdenum experiment and the Mo-pollination experiments at McLaren Vale. A detailed description of the sampling, processing and analysis procedures for petioles, inflorescences and shoot tips and a discussion of the results from 2003-04 and 2004-05 are provided in Chapter 4.
Reproductive tissue Flower samples were analysed to gain an overview of the nutritional status of the male and female flower parts. In 2004-05 samples were collected from the molybdenum experiments in the Hills and at McLaren Vale. In 2005-06 samples were collected from the molybdenum experiment at McLaren Vale.
In the laboratory a sample of newly opened flowers was sub-sampled from the inflorescence branches collected for flower observations (section 7.2.5). The flowers were dissected and separated into male (stamens) and female (pistils) parts and left to dry on clean paper overnight at room temperature. Replicates from the individual treatments were batched
98 together to enable sufficient material for analysis. Therefore, results were not replicated and statistical analysis was not performed. The molybdenum analysis for the stamens from the control vines in the old block at McLaren Vale was excluded from the dataset (Figure 7.5) due to sample contamination.
Analysis Waite Analytical Services performed the analyses of Mo, Fe, Mn, Cu, B, Zn, Ca, Mg, K, P, and Na using Inductively Coupled Plasma Optical Emission Spectrometry (ICPOES) in the first instance and thereafter using Inductively Coupled Plasma Mass Spectrometry (ICPMS) if the molybdenum concentration measured below 0.1 mg/kg.
7.2.4 Pollen vitality
At approximately 80% flowering, a sample of branches 2, 3 and 4 was collected from across the inflorescences on the vines in each replicate. From these branches flowers that were close to opening (determined by size and the colour of the cap having faded to pale green / yellow) were dissected and the anthers removed and allowed to dry overnight at room temperature. The following morning the pollen grains were separated from the anther sacs using a needle, stained using fluorescin diacetate (FDA) (Pinney & Polito 1990) and observed at 520 nm using an exciter filter of 395-440 nm. Using this technique pollen grains with active enzymes fluoresce and non-vital pollen does not. Thus, it was assumed that the vital stain reflected viability. Several fields of view, each containing approximately 100 pollen grains, were observed and counts of fluorescing versus non-fluorescing pollen grains were made. The percentage of vital pollen grains was calculated for each treatment.
Due to a procedural irregularity approximately 40% of samples registered an extremely low percentage of vital pollen grains. Samples that had a vitality of less than 4% were excluded from the mean value.
Pollen counts In 2003-04 counts were made of pollen grains visible on the edge of the longitudinal cross- section of the stigma (Figure 3.5). This gave an indication of pollen numbers, however, it should not be considered a complete or precise pollen count. Nevertheless, comparisons between treatments are to be considered valid.
99
7.2.5 Flower structure
The size and structural integrity of flowers were assessed in the field with the naked eye.
Ovule structure Observations of ovule structure were performed on pistils collected from the molybdenum experiment at McLaren Vale in 2005-06. A detailed description of the procedures for sample collection and processing is outlined in Chapter 3.
Ten pistils from each replicate were embedded in GMA-filled capsules. Those in which the pistil was optimally oriented for sectioning were sub-sampled. In order to make general observations of ovule anatomy, every third section of the pistils were mounted and the intermediate sections were discarded. From the remaining sections it was possible to identify the outer- and inner integuments, nucellus and a central embryo sac, when present (Figure 7.1).
Ovules were initially scored for the presence of an elongated embryo sac in the central nucellus. However, because the series of sections was incomplete, the classification of embryo sacs as functional, based on the presence of an egg cell, two synergids and the two polar nuclei or a polar fusion nucleus, could not be confirmed.
Each ovule was then scored on the following anatomical features:
• Normal outer and inner integuments and nucellus. • Elongation of central cells of the nucellus. • Presence of embryo sac. • Complete absence of an embryo sac.
From this information the percentages of ovules with each of these features could be calculated. Each pistil was also scored for the presence of at least one ovule with an embryo sac. The differences between treatment means were analysed using an unpaired, two-tailed t- test with a confidence interval of 95%.
Because entire pistils were sectioned rather than individual ovules, ovules within a single pistil were not all oriented in the same direction. This meant that in some cases ovules within the same pistil were sectioned longitudinally and other sections tended to be transverse. In 100 ovules that were sectioned transversely, evidence of an embryo sac could be observed, however it was impossible to determine whether the full complement of cells that compose a functional embryo sac were present.
A B ii oi n
C D
es
es
m
E F
es
Figure 7.1. Serial sections of a Merlot ovule showing the outer- and inner integuments (oi, ii), nucellus (n), embryo sac (es) and micropyle (m). A: visible inner and outer integuments; B: cells in the central nucellus are
101 elongated; C: first sign of an embryo sac; D: complete embryo sac; E: remnant of the embryo sac remains visible; F: embryo sac no longer visible, central nucellar cells tend to be elongated. The bar represents 100 µm.
7.3 Results
7.3.1 Weather observations during flowering
In 2003-04 the mean daily temperature was approximately 2˚C higher at McLaren Vale than in the Hills during flowering. However, McLaren Vale received substantially more rainfall during this period than the Hills. At McLaren Vale in 2004-05 the flowering period was cooler than in 2003-04 however in the Hills it was warmer in 2004-05. The difference in mean daily temperature between the Hills and McLaren Vale was mainly attributable to differences in mean daily maximum temperatures (Table 7.1).
In 2004-05 McLaren Vale experienced major storms during the flowering period. Rain and hail caused considerable damage to opened flowers (Figure 7.2) and many caps on unopened flowers failed to properly abscise (Figure 7.3).
Table 7.1. Weather observations from the beginning of flowering to 80% flowering on Merlot in the Adelaide Hills and at McLaren Vale in 2003-04 and 2004-05 Start – 80% Mean daily Mean daily Mean daily Rainfall (mm) flowering min (˚C) max (˚C) temp (˚C)
Adelaide Hills 2003-04 2 - 9 Dec. 9.1 26.5 17.8 0.8 2004-05 24 - 29Nov 13.3 32.1 22.7 0 a
McLaren Vale 2003-04 12 - 20 Nov. 13.8 25.1 19.5 9.8
2004-05 New block 29 Oct – 12 Nov. 9.7 19.4 14.6 37.4 Old block 1 Nov – 16 Nov 9.5 18.4 14.0 38.8 a rainfall data from Lenswood, South Australia
102
Figure 7.2. Physical damage on Merlot flowers after a severe storm at McLaren Vale in November 2004. Arrows indicate necrosing and reflexed stamens.
Figure 7.3. Merlot flowers after a severe storm at McLaren Vale in November 2004. The arrows indicate flower caps that only partially abscised.
In 2003-04 the mean daily temperature during the period of pollen tube observation was higher at McLaren Vale than in the Hills due to the higher minimum temperature at McLaren Vale. Total rainfall during the pollen tube observation period was also higher at McLaren
103 Vale than in the Hills. In 2004-05 the mean daily temperature in the Hills was 7.5˚C higher than at McLaren Vale, a function of both higher minimum and maximum temperatures. No rainfall was recorded at either site in during the period of pollen tube observation in 2004-05 (Table 7.2).
Table 7.2. Weather observations during the period of observed pollen tube growth (PTG) in Merlot flowers in the Adelaide Hills and at McLaren Vale in 2003-04 and 2004-05. Period of PTG Mean daily Mean daily Mean daily Rainfall (mm) min (˚C) max (˚C) temp (˚C)
Adelaide Hills 2003-04 3 – 6 Dec. 8.6 21.7 15.2 0.6 2004-05 27 – 30 Nov. 13.0 31.5 22.3 0 a
McLaren Vale 2003-04 19 - 22 Nov. 13.9 24 18.2 7.4 2004-05 7 – 10 Nov. 9.8 19.8 14.8 0 a Rainfall data for Lenswood
7.3.2 Nutritional analysis
Molybdenum in the vegetative tissue The concentration of molybdenum in the petioles from the controls was within the suggested deficiency range (0.05-0.09 mg/kg). Petioles from vines sprayed with sodium molybdate at MELS 12 in 2005-06 had a significantly higher concentration of molybdenum at MELS 25 (Table 7.3).
Molybdenum in the reproductive tissues In the Hills and at McLaren Vale the concentration of molybdenum in the stamens tended to be higher than in the pistils at 80% flowering in 2004-05. In the Hills, where the vines were not sprayed in 2004-05, the concentration of molybdenum in the stamens was higher for the rate 1 and 2 treatments than the controls while the residual effect of the molybdenum treatment was not evident in the pistils (Figure 7.4 and Table 7.3).
104 12 )
kg 9 on (mg/ i 6 rat
3 Mo concent
0 Rate 0 Rate 1 Rate2
Figure 7.4. Molybdenum concentration in the stamens (■) and pistils (■) of Merlot flowers at modified E-L stage (MELS) 25 (80% flowering) in the Adelaide Hills in 2004-05. Vines were treated with rate 0 (control) rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and were left untreated in 2004-05.
At McLaren Vale the concentration of molybdenum tended to be higher in flowers from Mo- treated vines. This response was greater in the new block compared to the old block. The molybdenum concentration tended to be higher in the stamens compared to the pistils in both the new and old blocks (Figure 7.5 and Table 7.3).
12 12
9 9 )
6 6 Mo (mg/kg 3 3
0 0 Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2
New block Old block
Figure 7.5. Molybdenum concentration in the stamens (■) and pistils (■) of Merlot flowers at modified E-L stage (MELS) 25 (80% flowering) in 2004-05 at McLaren Vale. Vines in the new block (left) were treated with rate 0 (control) rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2004-05 only and vines in the old block (right) were sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05.
In 2005-06 flowers collected from the Mo-treated vines at McLaren Vale tended to have a higher concentration of molybdenum than those from the unsprayed control vines (Figure 7.6 and Table 7.3). 105
6
) 4
Mo (mg/kg 2
0 Rate 0 Rate 1
Figure 7.6. Molybdenum concentration in the stamens (■) and pistils (■) of Merlot flowers at modified E-L stage (MELS) 25 (80% flowering) in 2005-06 at McLaren Vale. Vines were treated with rate 0 (control) and rate 1 molybdenum sprays at MELS 12 (shoots 10 cm) in 2005-06.
Table 7.3. Concentration of molybdenum (mg/kg) in the stamens, pistils, petioles, shoot tips and inflorescences of Merlot in the Adelaide Hills and at McLaren Vale at modified E-L stage (MELS) 25 (80% flowering) in 2004- 05 and 2005-06. Vines in the Hills were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and were left untreated in 2004-05. At McLaren Vale vines in the new block were sprayed in 2004-05 only and vines in the old block were sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. In 2005-06 vines were sprayed at MELS 12 in that season only. Different letters after the values within an experiment and within a row indicate significant differences between treatments (p<0.05). Where there are no letters after the values samples were not replicated and statistical analysis was not performed. Mo treatment Rate 0 Rate 1 Rate 2 2004-05 Hills Stamens 0.86 1.1 2.3 Pistils 0.26 0.24 0.31 Petioles 0.153 a 0.123 a 0.135 a Shoot tips 0.063 a 0.077 a 0.058 a Inflorescences 0.185 a 0.143 a 0.121 a
McLaren Vale New block Stamens 3.4 4.9 10 Pistils 0.86 3.0 4.7 Petioles 0.058 a 7.504 b 11.649 c Shoot tips 0.023 a 0.208 b 0.394 c Inflorescences 0.035 a 6.661 b 11.904 c
Old block Stamens - 2.7 4.6 Pistils 0.83 1.8 2.5 Petioles 0.066 a 5.429 b 8.647 c Shoot tips 0.038 a 0.303 b 0.508 c Inflorescences 0.032 a 7.161 b 14.358 c
2005-06 McLaren Vale Stamens 0.40 5.3 - Pistils 0.83 3.3 - Petioles 0.062 a 6.296 b -
106 Other nutrients in the reproductive tissue The most notable difference in the nutrient analyses was the higher concentration of sodium in the flowers from McLaren Vale compared to those from the Hills (Table 7.4 and Table 7.5).
Table 7.4. Nutrient analyses (mg/kg) of Merlot stamens at modified E-L stage (MELS) 25 (80 % flowering) in 2004-05. In the Hills vines were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and were left untreated in 2004-05. At McLaren Vale vines in the new block were sprayed at MELS 12 in 2004-05 only and vines in the old block were sprayed at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. In 2005-06 the vines were sprayed at MELS 12 in that season only. Samples were not replicated and statistical analyses were not performed on the data. Hills McLaren Vale 2004-05 2004-05 2005-06 New block Old block Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Fe 99 93 86 101 109 70 98 94 73 70 107 Mn 31 31 32 21 25 24 20 18 18 27 23 B 43 39 37 99 102 102 91 76 70 122 109 Cu 51 36 34 54 40 38 45 28 33 25 21 Zn 58 56 169 48 50 49 47 52 53 64 57 Ca 10900 10500 10800 10000 10100 12200 11100 11800 11200 10300 9900 Mg 3800 3800 3900 2600 2500 2800 2600 2700 2800 3300 3000 Na 97 89 110 360 480 220 210 370 310 210 270 K 21000 33000 20000 14100 13600 14100 14000 13900 14500 19000 16800 P 4400 4300 4400 4300 4500 4700 4600 4100 4100 5000 4300 S 2800 2800 2800 2700 2800 2500 2500 2200 2200 2900 2500
Table 7.5. Nutrient analyses (mg/kg) of Merlot pistils collected at modified E-L stage (MELS) 25 (80 % flowering) in 2004-05. In the Hills vines were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and were left untreated in 2004-05. At McLaren Vale vines in the new block were sprayed at MELS 12 in 2004-05 only and vines in the old block were sprayed at MELS 12 in 2003- 04 and again at MELS 12 in 2004-05. In 2005-06 vines were sprayed at MELS 12 in that season only. Samples were not replicated and statistical analyses were not performed on the data. Hills McLaren Vale 2004-05 2004-05 2005-06 New block Old block Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Rate 2 Rate 0 Rate 1 Fe 48 56 60 63 187 59 158 75 75 51 55 Mn 11 13 14 17 19 17 16 13 13 18 18 B 23 25 24 72 73 59 58 50 45 89 86 Cu 21 25 24 28 25 25 36 20 24 20 20 Zn 26 30 31 35 37 40 42 40 38 44 41 Ca 8000 9200 9000 11100 11300 11100 10400 10500 9800 10000 10000 Mg 1620 1830 1890 2600 2700 2800 2800 2500 2400 2500 2200 Na - 47 51 480 610 360 360 600 480 360 310 K 18000 21000 21000 18600 17500 22000 21000 23000 23000 23000 24000 P 3400 4000 3800 4500 4300 4500 4600 4200 4200 5800 5600 S 1820 2200 2200 2700 2700 2600 2800 2500 2500 2900 3000
107 7.3.3 Pollen vitality
In the Hills there was a significant difference in pollen vitality between the rate 1 and rate 2 treatments, however neither was significantly different to the control. At McLaren Vale there were no significant differences in pollen vitality between treatments (Table 7.6).
Table 7.6. Vitality (%) of pollen collected from Merlot vines in the Adelaide Hills and at McLaren Vale at modified E-L stage (MELS) 25 (80% flowering) in 2004-05. Vines in the Hills were treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at MELS 12 (shoots 10 cm) in 2003-04 and were left untreated in 2004-05. At McLaren Vale vines in the new block were treated at MELS 12 in 2004-05 only and vines in the old block were treated at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. Different letters after the values represent significant differences between treatments within an experiment (p<0.05). Molybdenum treatment Rate 0 Rate 1 Rate 2
Hills 73 ab 68 a 78 b
McLaren Vale New block 58 a 73 a 73 a
Old block 80 a 78 a 75 a
7.3.4 Pollen tube growth
In the Adelaide Hills molybdenum treatment did not significantly affect the number of pollen grains on the stigma or pollen tube growth in 2003-04 or 2004-05. The results were similar for both seasons. Both the number of penetrated ovules per ovary and the percentage of ovaries with at least one penetrated ovule were higher in the Hills than at McLaren Vale (Table 7.7 and Table 7.8).
At McLaren Vale in 2003-04 there were no significant differences in the number of pollen grains on the stigma or the number of pollen tubes in the lower ovary, however, the number of penetrated ovules per ovule tended to be higher for Mo-treated pistils than the controls. Rate 1-treated pistils had significantly more penetrated ovules per ovary and a higher percentage of ovaries with at least one penetrated ovule than the control (Table 7.7).
108 Table 7.7. Observations made on Merlot flowers four days after opening in 2003-04. Flowers were collected from vines treated with rate 0 (control), rate 1 and rate 2 molybdenum sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in the Adelaide Hills and at McLaren Vale. Different letters after the values represent significant differences between treatments (p<0.05). # Pollen grains # Pollen tubes # Penetrated Ovaries >/= 1 2003-04 on stigma in lower ovary ovules / ovary penetrated ovule (%) Hills Rate 0 93 a 26 a 1.9 a 88 a Rate 1 96 a 28 a 2.0 a 93 a Rate 2 93 a 26 a 1.9 a 87 a
McLaren Vale Rate 0 68 a 5 a 0.6 a 44 a Rate 1 54 a 6 a 1.1 b 68 b Rate 2 59 a 6 a 1.0 ab 57 ab
In the new block in 2004-05, pistils from Mo-treated vines had significantly more pollen tubes in the lower ovary than the controls. In both the new and old blocks the number of penetrated ovules per ovary was significantly higher for rates 1 and 2 compared to the control. In both blocks the percentage of ovaries with at least one penetrated ovule per ovary was significantly higher for pistils from Mo-treated vines than for the controls (Table 7.8).
At McLaren Vale there was a higher proportion of penetrated ovules per ovary and ovaries with at least one penetrated ovule in 2003-04 than in 2004-05 (Table 7.7 and Table 7.8).
Table 7.8. Observations made on Merlot flowers in 2004-05. In the Hills vines were treated with rate 0 (control), rate 1 and rate 2 Mo sprays at modified E-L stage (MELS) 12 (shoots 10 cm) in 2003-04. At McLaren Vale vines in the new block were treated at MELS 12 in 2004-05 only and vines in the old block were treated at MELS 12 in 2003-04 and again at MELS 12 in 2004-05. Different letters after the values represent significant differences between treatments within an experiment (p<0.05). # Pollen tubes # Penetrated Ovaries >/= 1 2004-05 in lower ovary ovules / ovary penetrated ovule (%)
Hills Rate 0 5.1 a 1.9 a 86 a Rate 1 7.6 a 2.2 a 93 a Rate 2 5.3 a 1.7 a 80 a
McLaren Vale New block Rate 0 1.1 a 0.03 a 3 a Rate 1 3.5 b 0.4 b 26 b Rate 2 3.2 b 0.4 b 27 b
Old block Rate 0 1.1 a 0.0 a 4 a Rate 1 2.3 ab 0.4 b 26 b Rate 2 3.5 b 0.3 b 27 b
109 7.3.5 Flower structure
On visual inspection flowers were hermaphrodite with a single slender-necked pistil and five stamens, which is typical for Vitis vinifera. There were few deviations from the normal flower structure and there were no significant differences between flowers from control and Mo- treated vines. However, a high incidence of structural abnormalities was noted for pistils prepared for pollen tube observation from Mo-deficient vines at McLaren Vale. The presence of a hollow or diverted stylar transmitting tissue was usually detrimental to the success of pollen tube progression down the style (Figure 7.7). Many of the stigmas were necrotic and it was not uncommon to see necrotic areas on the surface of the pistil (Figure 7.8).
250 µm
Figure 7.7. A Merlot pistil with a diverted stylar transmitting tissue. Pollen tubes fail to proceed to the lower ovary.
A B
100 µm
Figure 7.8. A: A Merlot pistil with a hollow and diverted stylar transmitting tissue, necrotic stigma and necrosis at the edges of the section (indicated by arrows) and B: a healthy pistil with pollen tubes growing in a continuous, straight style.
110 It was not uncommon to see convoluted pollen tubes in the presence of penetrated ovules in the Hills (Figure 7.9), however there was also a high incidence of convoluted pollen tubes in pistils from McLaren Vale where no ovule was penetrated (Figure 7.10).
100 µm
Figure 7.9. A Merlot pistil from the Adelaide Hills with two penetrated ovules. The redundant pollen tubes appear convoluted (indicated by the arrows).
7.3.6 Ultrastructure of ovules
100 µm
Figure 7.10. A convoluted pollen tube in the lower ovary of a Merlot flower from McLaren Vale.
111 7.3.7 Ovule structure
The outer and inner integuments appeared normally developed for ovules from Mo-deficient and Mo-treated vines. Only one ovule out of the 136 from Mo-deficient vines did not have elongated cells in the central nucellus. The percentage of ovules without an embryo sac in the nucellus was approximately double for those from Mo-deficient vines than from Mo-treated vines, however this was not statistically significant. The incidence of pistils containing at least one ovule without an embryo sac was significantly higher from Mo-deficient vines than Mo- treated ones (Table 7.9).
Table 7.9. Percentages of ovules with elongated central nucellus cells and ovules with an embryo sac in the nucellus from Merlot vines at McLaren Vale in 2005-06. Flowers were collected and fixed on the day prior to opening from vines treated with rate 0 (control) and rate 1 molybdenum sprays at modified E-L stage 12 (shoots 10 cm). Different letters after the values indicate significant differences between treatments (p<0.05). Ovules with elongated Ovules with an embryo Ovaries with >/= 1 ovule with an central nucellus cells (%) sac in the nucellus (%) embryo sac in the nucellus (%)
Rate 0 99.3 a 83.6 a 50 a Rate 1 100 a 91.5 a 77 b
112 7.4 Discussion
7.4.1 Weather conditions at flowering
Molybdenum deficiency has been suggested to be more prevalent in cool, wet seasons (pers. comm., C. Williams 2003; Williams et al. 2004). However, as discussed in Chapter 4, the weather and phenological data show that, as a consequence of later budburst in the Hills, the mean daily temperatures in the period between budburst and flowering were similar in the Hills and at McLaren Vale. The length of the period from budburst to flowering was also similar for the Hills and McLaren Vale suggesting that both sites experienced similar conditions during that period and that temperature was not responsible for the differences in vine molybdenum status between the two sites. The weather data during the flowering period supports this with similar mean daily temperatures throughout that period. However, McLaren Vale received significantly more rainfall than the Hills during flowering in both seasons. Heavy rain (at least 27 mm) received over a period of 8.5 hours can remove up to 97% of pollen from the stigma (Tkačenko 1960) potentially reducing the success of fertilisation. The rain and hail damage to the flowers at McLaren Vale during flowering in 2004-05 and the 7.4 mm of rain received during the period of pollen tube growth may all have contributed to poor pollen tube growth, however it does not explain differences in yield between Mo-treated and Mo-deficient vines. The large number of flowers with partial cap abscission, which also failed to enlarge, suggests that Merlot may not be capable of cleistogamy.
7.4.2 Nutritional analyses of the flowers
Despite the lack of replication of samples submitted for nutrient analysis there was good conformity in the trends across the experiments. In all cases the concentration of molybdenum increased in both the male and female flower parts in response to rate 1 and rate 2 molybdenum sprays in the season that they were applied. Given that the flowers were closed when the spray treatments were applied, the tendency for molybdenum to concentrate in flowers of Mo-treated vines confirmed that the sodium molybdate penetrated the tissue.
Molybdenum in the stamens and pistils In each experiment the concentration of molybdenum was consistently higher in the stamens than the pistils. Molybdenum concentration in the stamens approximately doubled when rate 2 was applied compared to rate 1. The pistils did not always register a response of the same magnitude as that of the stamens, however, this may be a result of analytical error. 113
Given that the male and female flower parts were undifferentiated at the time of Mo-treatment the differences in molybdenum concentration between the stamens and pistils show that there was partitioning of molybdenum between the male and female flower parts during flower development.
Compared to the petioles, shoot tips and pistils, the stamens from the control vines had relatively high levels of molybdenum. This is in agreement with Fregoni (1980, cited in Phillips 2004) who also recorded relatively high molybdenum concentrations (1 mg/kg) in grapevine pollen compared to other parts of the vine. No comparative data could be found for molybdenum concentration in the pistils.
Further investigation into the distribution of molybdenum in the male and female flower parts and the relationship with fruitset could be achieved by sampling larger, replicated samples of flowers from Mo-treated and Mo-deficient vines.
The effect of sodium molybdate foliar sprays applied to vines in consecutive seasons on molybdenum concentration in the flowers The concentration of molybdenum in the flowers was similar in blocks where molybdenum sprays were applied for the first time (i.e. the McLaren Vale new block in 2004-05 and in 2005-06). However, flowers from the old block, which was treated in consecutive seasons, tended to have a lower concentration of molybdenum. This response reflects that observed in the petioles (Chapter 4) and supports the theory that, after the initial Mo-deficiency is satisfied, the demand for molybdenum in the annual vine structures decreases.
Sodium concentration in the flowers The high concentration of sodium in the flowers at McLaren vale may play a significant role in the fecundity of the flowers from that site. While Sun et al. (2004) ruled out the ionic effects of Na+ as the cause of ovule abortion in salt-stressed wheat plants, cell homeostasis may also be disrupted by high concentrations of Na+ causing oxidative damage to the reproductive organs (Da et al. 2000, cited in Sun et al. 2004). Sodium also plays a supporting role in the uptake or binding of calcium, an element found to be critical for pollen tube growth (Brewbaker & Kwack 1963). Therefore high concentrations of sodium may potentially have a secondary deleterious effect on flower fecundity.
114 7.4.3 Pollen vitality
Despite the higher molybdenum levels in the Mo-treated stamens, there was no change in pollen vitality compared to the control, and the number of pollen grains on the stigma was unaffected. There were no differences in the appearance of pollen viewed under the compound microscope. This is in contrast to Agarwala et al. (1979) who found that Mo- deficient maize plants produced fewer and smaller pollen grains that appeared shrivelled and had poor viability. That pollen also had altered enzymatic activity.
Other structural anomalies that prevent germination of Mo-deficient pollen cannot be ruled out. Several authors report ‘acolporation’, or a lack of germination pores in the pollen surface, of some poor producing varieties (Lombardo et al. 1978; Kimura et al 1997; Abreu 2006). In acolporated pollen the continuous exine layer presents a mechanical impediment that prevents pollen from germinating (Lombardo et al. 1976). In the variety ‘Loureiro’, Abreu (2006) found that despite acolporation, the pollen was viable. Further ultrastructural observations may reveal differences in pollen morphology not observed in this study. These are areas for future experimentation.
7.4.4 Pollen tube growth
In the Hills, where the level of molybdenum was considered adequate in both seasons, sodium molybdate sprays did not affect the ability of pollen tubes to reach and penetrate the ovules. In fact, the percentage of ovaries with at least one penetrated ovule was far greater than at McLaren Vale where the Mo-deficiency was corrected.
The effect of Mo-treatment on pollen tube growth is in agreement with Ma et al. (1992). In that study the proportion of Kyoho pollen germinating on agar improved by 10% when the growth medium was supplemented with sodium molybdate. In the current study, Mo- deficiency inhibited pollen tube growth, however, the mechanism that was affected remains unclear. A number of potential inhibitors of pollen tube growth are discussed in section 7.4.5.
Mo-treatment increased the frequency of ovule penetration. However, there was more improvement in the incidence of ovaries with at least one penetrated ovule than in the number of penetrated ovules per ovary. This is in agreement with the effects of Mo-treatment on the yield components of Merlot (Chapter 6). A higher number of penetrated ovules per ovary can potentially produce more seeds per berry and hence larger berries. While there was a small increase in berry size at McLaren Vale in 2003-04, berry size was not affected in 2004-05. 115
In 2003-04 and in 2004-05 the biggest contributor to improved bunch weight was fruitset. To have one penetrated ovule per ovary gives the ovary the potential to develop into a berry. Therefore it was expected that the frequency of ovaries that have at least one penetrated ovule should be strongly related to fruitset (Figure 7.11).
50
R2 = 0.9446
) 40 (% et s t
Frui 30
20 0102030
Ovaries with at least 1 penetrated ovule (%)
Figure 7.11. The relationship between the percentage of ovaries with at least one penetrated ovule and percent fruitset of Merlot at McLaren Vale in 2004-05.
7.4.5 Potential causes of poor pollen tube growth in Mo-deficient flowers
Effects of Mo-deficiency on pollen As discussed in section 7.4.3, Mo-deficiency may have affected pollen development and morphology, or its enzymatic activity, which may have inhibited its ability to germinate. Mo- deficiency may also affect the vine’s sensitivity to environmental conditions. For example, some plants are reported to be more sensitive to cold stress under Mo-deficient conditions (Vunkova-Radeva et al. 1988). Low temperatures affect both flower development and pollen tube growth in grapes (Ebadi 1996). It is possible that the effects of short-term low temperatures associated with storms during flowering at McLaren Vale in 2004-05 may have exacerbated the effects of Mo-deficiency.
Effects of Mo-deficiency on conditions in the style The structure of the style is important to the success of pollen tube growth. In the low yielding variety Poulsard, the style is hollow, a feature that is suggested to inhibit pollen tube growth
116 (May 2004). The hollow sections and diversions in the transmitting tissue of the style observed in Mo-deficient Merlot pistils may also be a contributing factor to poor pollen tube growth.
The chemical composition in the style is critical to the success of pollen tube growth because of its role to supply nutrients as well as providing favourable conditions for growth and signalling to the developing pollen tube. The presence of inhibiting substances in the pistil of flowers has also been shown to suppress pollen tube growth in the lower style or upper ovary, leading to poor set (Okamoto et al. 1989; Okamoto et al. 1995). It is unlikely that molybdenum plays a direct role in either of these functions; however, under Mo-deficient conditions the absence of molybdenum may be responsible for the interruption of metabolic pathways required to support favourable conditions in the style.
In a study by Cholet et al. (2002) a necrotic region of the septum was associated with high polyamine (PA) levels and shot berry formation. Polyamine synthesis is a stress response linked with nitrogen metabolism and while no such browning was observed in this study it is possible that with Mo-deficiency, biosynthesis of polyamines may provide an alternative nitrogen source (Pfosser et al. 1992, cited in Geny & Broquedis 2002). Mo-deficiency may also be a direct trigger for PA metabolism given that it responds to external conditions, especially mineral nutrient deficiencies (Flores et al. 1985, cited in Geny & Broquedis 2002). Preliminary experiments investigating the role of molybdenum in polyamine biosynthesis are currently under investigation by Cassandra Collins at the University of Adelaide.
Interrupted pollen tube signalling There is currently no available literature about the direct function of molybdenum in pollen tube signalling, however, there are a number of possible ways molybdenum may indirectly affect signalling. Molybdenum is a critical component of the enzyme nitrate reductase (NR), which produces nitric oxide (NO), a ubiquitous signalling molecule (Feijó et al. 2004). Under Mo-deficient conditions, NO synthesis is likely to decline, which may cause disruption to the pollen tube guidance system.
Phillips (2004), citing earlier work by Fido et al. (1977) who investigated the effects of Mo deficiency in cauliflower, hypothesised that elevated superoxide production associated with Mo-deficiency may cause seed abortion resulting in lower berry weights and a high incidence of millerandage in his Merlot experiments. In fact the cellular damage on Mo-deficient
117 cauliflower reported at or near flowering by Fido et al. (1977) may be a further cause of inhibited pollen tube signalling in Merlot.
Polyamines have previously been cited as potential inhibitors of pollen tube growth. Conjugated PAs have been implicated in molecular signalling in plant-pathogen interactions (Martin-Tanguy 2001) and thus may also serve a function in pollen tube signalling.
Abnormal ovule development Mo-deficiency may have affected ovule structure in a number of ways that were not examined in the current study. Blockages at the placenta or in the funiculus, or varying access to metabolite supplies through differences in the size of vascular connections, may all contribute to faulty ovule development (Ebadi 1996; Polito 1999). The absence of callose deposition that protects the embryo sac from unfavourable environmental conditions may be detrimental to ovule development and longevity (Abramova et al. 2003), however callose deposition may also be detrimental to the ovule by inhibiting its vascular supply (Pimienta & Polito 1982).
It has been widely reported that pollen tubes receive directional signals along the route to the ovule (Russell 1996). In grapevines the presence of a functional embryo sac, and hence the synergid cells, has been shown to be critical to attract a pollen tube and for fertilisation and fruitset to take place (Fougere-Rifot et al. 1993; Ebadi 1996). More recently it has been reported that in close proximity to the ovule (150 µm) pollen tubes receive specific chemotropic signals from at least one of the two synergid cells in the embryo sac (Higashiyama et al. 2001).
When the ovules in the current study were initially scored the results indicated that a higher incidence of ovules from Mo-treated Merlot vines contained an elongated, normal-looking embryo sac. However, as previously discussed (section 7.2.5), those data did not allow for a definitive conclusion regarding the functionality of the embryo sac to be drawn because the egg cell, two synergids and the two polar nuclei or a polar fusion nucleus could not always be identified. Using a more conservative approach, by scoring ovules as defective on the basis of the absence of an embryo sac, a higher proportion of ovules from Mo-deficient vines tended to be scored as defective. Despite a lack of a significant difference between the treatments, the results suggest that molybdenum may play a role in ovule development and that under Mo- deficient conditions ovules are more likely to develop without an embryo sac.
118 In a study of two apricot cultivars, one with good and the other with poor fruitset (‘Bergeran’), Albuquerque (2006) found that the two cultivars differed in the proportions of functional ovules at flowering time. In ‘Bergeran’, the proportion of functional ovules at flowering was increased by exogenous putrescine application. The authors suggested that a certain level of PAs could favour either pollen tube growth, or ovule development, which could stimulate pollen tube growth. In the current study the conservative approach taken to score the functionality of the ovules may have concealed the existence of asynchronous ovule development in ovaries from Mo-deficient vines. It is possible that molybdenum is critical to PA biosynthesis and that this is the mechanism by which molybdenum affects pollen tube growth in Merlot.
Pollen tube attraction and repulsion The presence of equal numbers of pollen tubes in the ovary, and ovules, has led researchers to presume that each ovule exerts control over one pollen tube (Russell 1996). In the current study it was common for abundant pollen tubes to enter the ovary. This suggests that the attractive signal from each ovule can potentially attract more than one pollen tube.
In most plant species it is also unusual for more than one pollen tube to enter or approach a faulty ovule or one in which fertilisation has taken place (Russell 1996; Higashiyama et al. 2001). However, there is some conjecture in the literature about why this occurs. Fertilised ovules may produce an inhibitory signal or, the attraction signal may be degraded after fertilisation (Weterings & Russell 2004). Therefore, it is conceivable that faulty ovules may potentially send an inhibitory signal to pollen tubes. In the current study convoluted pollen tubes were observed both in the presence and absence of ovule penetration suggesting that in the absence of ovule penetration either the attractive signal was nonexistent or the pollen tube received an inhibitory signal. The significantly higher frequency of ovaries that contained at least one defective ovule under Mo-deficient conditions may indicate that faulty ovules are responsible for repulsion of pollen tubes. If an inhibitory signal is emitted from a faulty ovule then the existence of a single faulty ovule per ovary, and its capacity to control multiple pollen tubes, may be as detrimental to pollen tube growth as an ovary with multiple faulty ovules.
A more comprehensive examination of ovules is warranted to confirm the effect of Mo- deficiency on ovules, specifically, on embryo sac development and pollen tube signalling.
119 7.4.6 Possible mechanisms for ovary development under Mo-deficient conditions
In previous studies, Mo-treatment reduced the effects of ‘millerandage’ and ‘hen and chicken’ on bunches (Gridley 2003; Phillips 2004; Williams et al. 2004). Put another way, bunches from Mo-deficient vines had a high proportion of ovaries arrested at different stages of development producing a combination of live green ovaries (LGOs), chicken berries and hen berries. In the current study, Mo-treatment reduced the effects of millerandage, that is the production of LGOs; however, Mo-deficient bunches were not affected by hen and chicken (Chapter 6).
The differences in stage of arrested development of ovaries between the current study and previous reports may be due to the existence of different proportions of healthy and faulty ovules. Stout (1936) categorised fruit into three classes: seeded, stenospermic and parthenocarpic. Berries within a bunch may be all alike or a combination of types depending on the ability of the ovules to function in fertilisation and develop into seeds.
It is hypothesized that:
• When molybdenum levels are adequate, the proportion of healthy ovules is maximised and so is the potential for fertilisation and normal seeded fruit development.
• When vines are moderately Mo-deficient, ovaries contain fewer functional ovules and berries contain fewer seeds and are smaller. However, small berry size alone is not necessarily indicative of a berry’s classification as a chicken. True chicken berries do not contain functional seeds which in turn governs their size. Chicken berries may contain seed traces, they are larger than LGOs and they develop fleshy pulp. These features are indicative of chicken berries formed by stenospermocarpy.
• With increasing Mo-deficiency the incidence of faulty ovules reaches a maximum. If none of the ovules in an ovary are functional, the ovary can only develop into a seedless berry (chicken) or an LGO via parthenocarpy (either obligate vegetative or stimulative) (Figure 7.12).
This hypothesized sequence suggests that:
• Vines in the current study were subject to a higher degree of Mo-deficiency than in previous studies.
• In previous studies, berries may not have been accurately classified. That is, the classification of chickens may have been based on small berry size alone. In the current
120 study it was found that classification of berries based on their size was not sufficient because in many cases, small berries that would typically have been classed as chickens were dissected, a seed was present.
100 Parthenocarpic -
) 80 LGOs % ( hes 60
Stenospermic -
on to bunc 40 chickens buti i r nt o 20 Seeded - C hens
0 100 75 50 25 Increasing Mo-deficiency Faulty ovules (%)
Figure 7.12. Diagrammatic representation of proposed proportions of ovary and berry types relative to the incidence of faulty ovules. The percentage contribution to bunches can be interpreted on the basis of either weight or incidence.
121 7.5 Conclusions
• The concentration of molybdenum tended to be higher in the stamens than the pistils confirming that molybdenum was partitioned to the male and female parts of the flowers during its development. This may be the reason for the dominant ovule effect.
• Sodium molybdate foliar sprays did not affect pollen vitality.
• Mo-treatment increased the number of penetrated ovules per ovary and the frequency of ovaries with at least one penetrated ovule.
• There was a strong relationship between the proportion of ovaries that had at least one penetrated ovule and percent fruitset.
• Inhibition of pollen tube growth in ovaries of Mo-deficient vines may be a result of changes to pollen morphology or enzymatic activity in the pollen.
• Hollow sections and diversions in the style of ovaries from Mo-deficient vines may have inhibited pollen tube growth.
• Mo-deficiency may have affected pollen tube growth via interruptions of metabolic pathways required to support favourable conditions in the style.
• Mo-deficient vines tended to produce a higher proportion of ovules without an embryo sac and there was a significantly higher incidence of ovaries that contained at least one ovule without an embryo sac. These data suggest that pollen tube signalling may be interrupted in ovaries from Mo-deficient Merlot vines.
• Different levels of Mo-deficiency in Merlot may result in differing proportions of healthy and faulty ovules. This may explain arrested development of berries occurring at different stages and the resultant production of different proportions of hen and chicken berries and LGOs.
122