Abstract
Background and Aims
Demography and spatial distribution of shoots are rarely studied on pruned trees. The present 2-year study deals with the effect of pruning strategies on shoot demography and development, and consequences on the spatial distribution of leaf area in three architecturally contrasted — from type II to IV — apple cultivars: ‘Scarletspur Delicious’, ‘Golden Delicious’ and ‘Granny Smith’.
Methods
All trees were initially subjected during 5 years to Central Leader training with winter heading on all long shoots. For 2 years, half of the trees were further trained with Centrifugal training, where removal of flowering shoots — called extinction pruning — was carried out along the trunk and at the bottom of branches at flowering time. During these 2 years, shoot type (vegetative, inflorescence) and length, and the three-dimensional spatial distribution of all shoots were assessed with an electromagnetic digitizer.
Key Results
Shoot demography, frequency of transitions toward an inflorescence from either an inflorescence (bourse-over-bourse) or a vegetative shoot (trend toward flowering), and the number of bourse-shoots per bourse were strongly affected by cultivar, with little influence of tree manipulation. In contrast, the proportion of vegetative long shoots developing from previous year latent buds was significantly lower in Centrifugal-trained trees for the three cultivars. Canopy volume showed large variations between cultivars, but only that of ‘Granny Smith’ was affected by tree manipulation in the 2 years. Spatial distribution of shoots varied significantly according to cultivar and manipulation. In ‘Scarletspur Delicious’ and, to a lesser extent ‘Golden Delicious’, the distribution of vegetative and flowering shoots in the outer and the inner parts, respectively, was not affected by tree manipulation. In contrast, in ‘Granny Smith’, vegetative shoots were stimulated in the periphery of Central Leader trees, whereas flowering shoots were stimulated in the periphery of Centrifugal-trained trees.
Conclusions
In apple, the variability of responses to contrasted pruning strategies partly depends on the genetically determined growth and flowering habit of the cultivar.
Key words: Malus domestica, apple, architecture, tree ideotype, shoot demography, shoot type, spatial pattern, centrifugal training, central leader, extinction pruning, reiteration
INTRODUCTION
Fruiting habits of apple trees have been investigated in several past studies (Bernhard, 1961; Lespinasse, 1977, 1980; Looney and Lane, 1984). Lespinasse and Delort (1986) classified apple trees according to their fruiting types (e.g. position of flowering on shoots, terminal vs. lateral flowering, shoot length and orientation of growth) and position of branches along the trunk (basitony vs. acrotony). This typology has been used to analyse flowering patterns of existing cultivars (Lespinasse, 1977, 1980; Lauri et al., 1995). It has also been used to describe tree architecture in progeny, leading to the proposal of four ideotypes, from type I or ‘columnar type’ with upright branches, high branching density and lateral flowering, to type IV with pendant branches, low branching density and terminal bearing (Lespinasse, 1992; De Wit et al., 2004). It is likely that the architectural components underlying these ideotypes are complex. Previous studies showed that the shoot growth rhythm (Costes et al., 1995; Lauri and Térouanne, 1995, 1998; Costes et al., 2003) and annual shoot length (Lauri and Trottier, 2004) determine in part flowering in the lateral vs. terminal position, and in the latter case inflorescence size and fruit set (Ferree and Palmer, 1982; Rom and Ferree, 1984; Proctor and Palmer, 1991; Lauri et al., 1996).
Flowering and vegetative growth may be modified by training and pruning (Johnson and Lakso, 1986; Rom and Barritt, 1990). Winter pruning increases vegetative growth and decreases yield in the following season depending on the severity of pruning (Elfving and Forshey, 1976; Elfving, 1990; Barden and Marini, 1998). It also influences the position of inflorescences with regard to vegetative growth, e.g. heading cuts may dissociate flowering from vegetative areas by enhancing vegetative growth in the outer part of the tree canopy and keeping inflorescences inside the canopy (Tustin et al., 1988; Hoying and Robinson, 2000). This may have deleterious effects on fruit quality.
Different methods of pruning according to the length of the eliminated organs, their type (flowering vs. vegetative shoots) and the timing of cutting have been described (Barritt, 1992). However, all pruning techniques tend to reduce the excessive vegetative growth, at least in the short term. An architectural analysis of these procedures shows that the effects of pruning also vary with the architectural level at which it is done: the removal of a portion of a long shoot (e.g. heading or shortening cuts sensu Barritt, 1992) usually prompts a strong reiteration process, leading to vegetative shoot growth (Hallé et al., 1978; Lauri, 2002), whereas the complete removal of 1-year-old short lateral shoots tends to stimulate growth of the remaining shoots (Lauri et al., 2004) and fruit set of adjacent inflorescences (Lauri and Térouanne, 1999). These findings led to the proposal of pruning strategies based on the removal of flowering shoots at a young stage of growth, with less pruning of old branches (Lauri et al., 1997a, b, 2004; Lauri, 2002). This procedure is called artificial extinction or, more commonly, extinction pruning. It increases leaf area of remaining shoots and light interception by the tree canopy through a decreased shoot density and a better distribution of shoots in space (Willaume et al., 2004). From a physiological point of view, this procedure would then improve the physiological autonomy of the flowering shoot, especially during the period preceding June drop, by enhancing carbon assimilation necessary for current fruit and bourse-shoot growth and flower initiation in the terminal bud of the bourse-shoot (Lauri et al., 2004). Note that bourse is the usual name of the small and bulging shoot bearing the apple fruit and that bourse-shoots are axillary shoots that develop from this bourse during the same year.
In this study the shoot demography and spatial distribution of leaf area were compared in two contrasted training systems: (a) the Centrifugal training system with extinction pruning carried out in the tree centre and on the underside of branches to improve light penetration within the tree (Willaume et al., 2004); and (b) the Central Leader system with annual heading and shortening cuts of all long shoots situated in the periphery of the tree with the objective of reducing the size of the lateral branches that compete with the leader (Heinicke, 1975). Although the former system allows free growth of branches with a low density of laterals and a more oblique habit, the latter stimulates vegetative and upright shoot growth (Fig. 1). The objective of the study was to characterize the influence of tree manipulation on the architectural features of the tree for a range of apple cultivars. More precisely, the aim was to analyse the effects of cultivar architectural ideotype and manipulation on (a) shoot type (vegetative, inflorescence), demography and shoot length; (b) sequence of lateral development over 2 years; and (c) spatial distribution of shoots.
Fig. 1.
Images of representative trees, and shoot type distribution (green, vegetative shoot; red, bourse; blue, bourse-shoot) according to manipulation (C-trees, Centrifugal training; L-trees, Central Leader), for ‘Scarletspur Delicious’ (A, B), ‘Golden Delicious’ (C, D) and ‘Granny Smith’ (E, F). Images are horizontal views synthesized with VegeSTAR software (Adam et al., 2002).
MATERIALS AND METHODS
Plant material
The experiment was situated in the Bekaa valley, Lebanon, at 900 m above sea level in the American University of Beirut research field (Haouch Sneid; 33°95′N; 36°02′E). Trees were planted in 1999 in a randomized block design and consisted of three two-tree plots for each cultivar, one plot per row. Trees were grafted on M7 rootstock. The orchard layout was 4 × 3·5 m with South–North row orientation. Agricultural practices included irrigation with mini-sprinklers, standard fertilization and spraying.
In winter 2004, three cultivars belonging to contrasted ideotypes according to Lespinasse (1992) typology were selected: ‘Scarletspur Delicious’ (type II), ‘Golden Delicious’ (type III) and ‘Granny Smith’ (type IV).
Training methods
Since plantation, trees were trained to obtain a Central Leader (Heinicke, 1975). Heading of the leader was done for the first year in order to obtain a stronger trunk made of consecutive reiterated complexes. One-third of the 1-year-old lateral shoots was headed back to reduce the competition of branches with the trunk. Only shoots competitive with the trunk were headed back. The same practices were applied to the three cultivars. Tree morphology was contrasted and was highly related to the cultivar (Fig. 1): (a) very compact and small trees for ‘Scarletspur Delicious’ with weak vigour as soon as the tree started to fruit; (b) intermediate tree size for ‘Golden Delicious’ with no obvious dominance of the trunk and most of the inflorescences located in the inner part of the tree; and (c) larger trees with vigorous vertical branches with secondary bending for ‘Granny Smith’.
For each cultivar, six healthy trees (two trees per row) were selected. In each pair, one tree was trained as previously described to a Central Leader (L), and a Centrifugal training system (C) was applied to the other one.
Centrifugal training included the following steps. In the first year (2004), all the shoots and buds along the trunk and on the underside of branches were removed to make a light well. Extinction pruning led to an average of 30–35 flowering buds per branch to approximate the expected number of fruits on Central Leader trees. All branches along the trunk were kept. In the second year (2005), there was almost no re-growth at sites where extinction pruning was carried out. Complementary extinction pruning was carried out if needed to reach the same crop load as in 2004.
Extinction pruning on C trees and shortening cuts on L trees were performed before full bloom and before bud burst, respectively. In both treatments, fruit thinning was carried out to keep one fruit per inflorescence.
Shoot demography assessment and sequence analysis
A demographic approach was carried out at harvest from August to October in 2004 and 2005, depending on the cultivar, using the methodology developed by Sinoquet and Rivet (1997) and adapted to apple by Willaume et al. (2004). An electromagnetic digitizing device (Fastrak®, Polhemus; Colchester, VT, USA) was used with the adapted software POL95 (Adam, 1999). This methodology allowed three-dimensional (3D) spatial positioning of each individual shoot by recording the spatial coordinates of shoot proximal and distal ends and deriving shoot length and orientation angles. For each shoot category (see below), a sample of ten shoots per cultivar showing a large range of size was randomly collected from the trees of each treatment. It was used to compute allometric relationships between shoot length (obtained from digitizing), and (a) the number of leaves and (b) shoot leaf area. Shoot leaf area was estimated by scanning leaves and measuring their area with image processing software Scion Image. Since there were no significant differences between the two treatments, data were merged, giving rise to 20 shoots per shoot type and per cultivar.
In 2004, annual shoots were divided into two types according to previous literature (Lauri and Kelner, 2001; Willaume et al., 2004): flowering shoot (FS), which was split into bourse (B) and associated bourse-shoot(s) (BS), and vegetative shoot (VS). BS and VS were split into short and long shoots, denoted by SBS and LBS, and SVS and LVS, respectively. For BS and VS, the threshold of 4 cm was used to discriminate long from short shoots. In the second year of the experiment, 2005, the same description was used on each 2005-shoot and on the parent 2004-shoot which bore it through either terminal growth or the branching process, or below pruning cuts. In the latter case, this was mainly on L-trees. When there was no FS or VS subtending the 2005-shoot, it was assumed that the shoot issued from a 2004-latent bud (L).
Shoot sequences over 2 years were defined to characterize shoot dynamics. Parameters of bourse-over-bourse (β) and trend toward flowering (ϕ) were used to quantify the tendency to make an FS in year i + 1 from either an FS or a VS produced in year i, respectively (Lauri et al., 1995, 1997a):
 |
1 |
where N[FSi → FSi + 1] denoted the number of FS in year i giving an FS in the terminal position in year i + 1, and N[FSi] denoted the total number of FS in year i.
 |
2 |
where N[VSi → FSi +1] denoted the number of VS in year i giving an FS in the terminal position in year i + 1, and N[VSi] denoted the total number of VS in year i.
Two other sequences were used to characterize awakening of latent buds (Lauri et al., 1997a). Parameter ϖ estimated the proportion of latent buds in year i that gave rise to VS in year i + 1 among all latent buds in year i broken in year i + 1:
 |
3 |
where N[Li → VSi +1] is the number of latent buds in year i leading to a VS in year i + 1, and N[Li → (VSi +1 + FSi +1)] is the number of latent buds in year i leading to a VS or an FS in year i + 1.
Parameter λ computed the proportion of VS and FS in year i + 1 coming from latent buds in year i among all VS and FS in year i + 1:
 |
4 |
Finally, the number of bourse-shoots per bourse was analysed.
Spatial distribution of leaf area
As leaves were not digitized, they were generated by using the method of Sonohat et al. (2006), which allows reconstructing the foliage of trees digitized at shoot scale. The method uses allometry relationships and empirical distribution functions to reconstruct tree foliage properties. The method output is the spatial distribution of leaves, where size, spatial coordinates and orientation angles are generated for each leaf. The method has been carefully assessed in the case of peach trees (Sonohat et al., 2006), where the spatial distribution of the foliage was shown to be correctly simulated from the reconstruction method. The method has also been previously used for apple trees (Willaume et al., 2004). From the leaf distribution, further computations were used to summarize the spatial information. First the minimum and maximum values of leaf coordinates along each axis (X, Y, Z) were computed, in order to define the bounding box of the tree. The box was then virtually divided into cubes, the sides of which were 0·2 m. In computer graphics jargon, the cubes are called voxels, i.e. a nickname for ‘volume element’ (e.g. Reche et al., 2004). Leaves were assigned to voxels according to their spatial coordinates. The cumulated volume of vegetated voxels was assumed to quantify canopy volume (Phattaralerphong and Sinoquet, 2005). Canopy volume was computed for the whole foliage, and for FS and VS, separately. The sum of canopy volumes occupied by the FS and VS may be greater than tree volume, because FS and VS may share some common canopy space, which is counted as a contribution to both FS and VS volumes. Consequently, the difference between the sum of FS and VS volumes and tree volume is the canopy volume occupied simultaneously by FS and VS. Spatial coordinates of leaves were also used to compute vertical and radial profiles of leaf area per 0·2 m layer. The vertical profile was computed from the base of the bounding box (zmin). The radial profile was computed from the crown base (zmin) at the trunk location (assumed to be xmean and ymean of the bounding box). All profiles of leaf area were computed separately for FS and VS.
Statistical analyses
All statistical analyses were carried out using SPSS software (SPSS version 11·5, SPSS Inc., 2002). The following variables were subjected to analysis of variance (ANOVA): (a) proportions of shoot types, i.e. LVS and SVS, bourse, LBS and SBS, and number of BS per B; (b) parameters of bourse-over-bourse, trend toward flowering, latent buds leading to vegetative shoots and proportion of growing shoots coming from latent buds; and (c) volume of the tree, fraction of the tree volume occupied by VS and FS, and simultaneously occupied by FS and VS. Eventually the spatial distribution of leaf area within each tree canopy was summarized as follows: the cumulated leaf area of VS and FS as a function of height (‘vertical profile’ indexed by the distance from the orchard floor) and as a function of the distance from the trunk at the base of the crown (‘radial profile’) planes was computed. To avoid extreme values due to isolated branches, especially in C-trees of ‘Granny Smith’, statistical tests were carried out on distances corresponding to 50 % of cumulative leaf area. Cultivar and tree manipulation were considered as main factors. Since there was no significant tree effect, it was discarded from the model, as were the interactions between these factors. Duncan's multiple range test was computed when the cultivar effect was significant. The comparison between tree manipulations was performed with a two-sample t-test. Comparison for dependent samples, such as the LVS length borne by pruned parent shoots on the same L-tree for the three cultivars, was performed with a paired t-test (Table 6). Results with P <0·05 were considered to be statistically significant.
Table 6.
Length of long vegetative shoots (LVS) in 2005 borne on non-pruned and pruned shoots on L-trees of the three cultivars
| Tree manipulation | n | Vegetative shoot length (cm) |
|---|---|---|
| L-trees pruned shoots | 9 | 19·59a |
| L-trees non-pruned shoots | 9 | 16·87b |
Superscript letters indicate significant differences between pruned and non-pruned 2004-shoots of the merged three cultivars, paired samples t-test at P < 0·05.
RESULTS
Shoot demography
Tree manipulation did not significantly affect the proportion of shoot types in both years (Table 1). The proportion of B according to the number of BS per bourse was significantly influenced by the cultivar, except the ‘zero BS’ class in 2005, but was not influenced by tree manipulation (Table 2). ‘Scarletspur Delicious’ had a significantly higher proportion of B with more than one BS, and a lower proportion of B with only one BS in both years. ‘Golden Delicious’ and ‘Granny Smith’ showed similar figures in 2004, but in 2005 ‘Granny Smith’ had the lowest proportion of B with one BS while ‘Golden Delicious’ had the lowest proportion of B with more than one BS.
Table 1.
Proportion of shoots according to cultivar, tree manipulation and type, in 2004 and 2005
| Shoot type proportion | |||||||
|---|---|---|---|---|---|---|---|
| Cultivar | Tree manipulation | No. of replications | B | SBS | LBS | SVS | LVS |
| 2004 | |||||||
| ‘Scarletspur Delicious’ | C | 3 | 0·08 | 0·14 | 0·01 | 0·72 | 0·06 |
| L | 3 | 0·13 | 0·22 | 0·00 | 0·59 | 0·06 | |
| ‘Golden Delicious’ | C | 3 | 0·11 | 0·04 | 0·06 | 0·65 | 0·14 |
| L | 3 | 0·18 | 0·09 | 0·08 | 0·54 | 0·11 | |
| ‘Granny Smith’ | C | 3 | 0·33 | 0·26 | 0·05 | 0·28 | 0·08 |
| L | 3 | 0·19 | 0·12 | 0·06 | 0·49 | 0·14 | |
| Cultivar mean | |||||||
| ‘Scarletspur Delicious’ | 0·10b | 0·06 | 0·00b | 0·39 | 0·06b | ||
| ‘Golden Delicious’ | 0·14b | 0·18 | 0·05a | 0·59 | 0·11a | ||
| ‘Granny Smith’ | 0·26a | 0·19 | 0·07a | 0·66 | 0·13a | ||
| Cultivar effect | |||||||
| F | 4·95 | 1·75 | 16·68 | 3·20 | 5·37 | ||
| P | 0·02 | 0·21 | <0·01 | 0·07 | 0·02 | ||
| Tree manipulation mean | |||||||
| C | 0·17 | 0·15 | 0·04 | 0·55 | 0·10 | ||
| L | 0·16 | 0·14 | 0·05 | 0·54 | 0·10 | ||
| Tree manipulation effect | |||||||
| F | 0·04 | 0·00 | 2·23 | 0·07 | 0·14 | ||
| P | 0·85 | 0·94 | 0·16 | 0·92 | 0·71 | ||
| 2005 | |||||||
| ‘Scarletspur Delicious’ | C | 3 | 0·19 | 0·24 | 0·00 | 0·54 | 0·04 |
| L | 3 | 0·24 | 0·29 | 0·00 | 0·43 | 0·04 | |
| ‘Golden Delicious’ | C | 3 | 0·22 | 0·17 | 0·06 | 0·44 | 0·11 |
| L | 3 | 0·23 | 0·20 | 0·04 | 0·40 | 0·13 | |
| ‘Granny Smith’ | C | 3 | 0·33 | 0·27 | 0·10 | 0·22 | 0·08 |
| L | 3 | 0·33 | 0·26 | 0·07 | 0·22 | 0·12 | |
| Cultivar mean | |||||||
| ‘Scarletspur Delicious’ | 0·22 | 0·26 | 0·00c | 0·48 | 0·04b | ||
| ‘Golden Delicious’ | 0·23 | 0·19 | 0·05b | 0·42 | 0·12a | ||
| ‘Granny Smith’ | 0·33 | 0·26 | 0·09a | 0·22 | 0·10a | ||
| Cultivar effect | |||||||
| F | 2·34 | 1·24 | 21·25 | 3·83 | 6·48 | ||
| P | 0·13 | 0·32 | <0·01 | 0·05 | 0·01 | ||
| Tree manipulation mean | |||||||
| C | 0·25 | 0·23 | 0·05 | 0·40 | 0·08 | ||
| L | 0·27 | 0·25 | 0·04 | 0·35 | 0·09 | ||
| Tree manipulation effect | |||||||
| F | 0·18 | 0·27 | 2·13 | 0·36 | 0·96 | ||
| P | 0·67 | 0·61 | 0·17 | 0·56 | 0·34 | ||
Within columns, different superscript letters indicate significant differences according to Duncan's test, P < 0·05.
Table 2.
Proportion of bourse with zero, one and more than one bourse-shoot(s) according to cultivar and tree manipulation in 2004 and 2005
| 2004 | 2005 | |||||||
|---|---|---|---|---|---|---|---|---|
| Cultivar | Tree manipulation | No. of replications | 0 | 1 | >1 | 0 | 1 | >1 |
| ‘Scarletspur Delicious’ | C | 3 | 0·09 | 0·26 | 0·73 | 0·08 | 0·59 | 0·33 |
| L | 3 | 0·07 | 0·35 | 0·64 | 0·11 | 0·62 | 0·27 | |
| ‘Golden Delicious’ | C | 3 | 0·16 | 0·76 | 0·08 | 0·07 | 0·84 | 0·08 |
| L | 3 | 0·15 | 0·76 | 0·09 | 0·09 | 0·84 | 0·07 | |
| ‘Granny Smith’ | C | 3 | 0·18 | 0·71 | 0·11 | 0·08 | 0·77 | 0·15 |
| L | 3 | 0·22 | 0·68 | 0·10 | 0·09 | 0·78 | 0·13 | |
| Cultivar mean | ||||||||
| ‘Scarletspur Delicious’ | 0·08b | 0·31b | 0·69a | 0·10 | 0·61c | 0·30a | ||
| ‘Golden Delicious’ | 0·16a | 0·76a | 0·09b | 0·08 | 0·84a | 0·08c | ||
| ‘Granny Smith’ | 0·19a | 0·70a | 0·10b | 0·08 | 0·77b | 0·14b | ||
| Cultivar effect | ||||||||
| F | 50·95 | 71·39 | 366·21 | 2·74 | 213·51 | 78·61 | ||
| P | 0·01 | 0·01 | <0·01 | 0·26 | <0·01 | 0·01 | ||
| Tree manipulation mean | ||||||||
| C | 0·12 | 0·58 | 0·30 | 0·08 | 0·73 | 0·19 | ||
| L | 0·12 | 0·59 | 0·28 | 0·10 | 0·74 | 0·15 | ||
| Tree manipulation effect | ||||||||
| F | 0·07 | 0·34 | 0·59 | 11·79 | 2·11 | 5·02 | ||
| P | 0·82 | 0·62 | 0·52 | 0·07 | 0·28 | 0·15 | ||
Within columns, different superscript letters indicate significant differences according to Duncan's test, P < 0·05.
Cultivar significantly influenced the proportion of long shoots, whether LBS or LVS, in the two years (Table 1). ‘Scarletspur Delicious’ had almost no LBS and a low proportion of LVS, while ‘Granny Smith’ reached 0·09 for LBS in 2005 and 0·13 for LVS in 2004. Bourse proportion was also affected by cultivar in 2004, but the difference was non-significant in 2005. SVS proportion differed, but not significantly, between the three cultivars.
In all cases, ‘Scarletspur Delicious’ had significantly fewer long shoots (LBS and LVS) than ‘Golden Delicious’ and ‘Granny Smith’ in both years. ‘Granny Smith’ had higher proportions of B (although non-significant in 2005) and LBS than the other two cultivars.
Sequence of shoot development over 2 years
There was a significant effect of tree manipulation on bourse-over-bourse (β), with higher values on L-trees compared with C-trees but without significant effect on the length of BS (Table 3). Trend toward flowering (ϕ) was not significantly changed by tree manipulation, but there was a significant increase of the length of VS supporting a terminal inflorescence on C-trees compared with L-trees (Table 3).
Table 3.
Bourse-over-bourse (β) and trend toward flowering (ϕ) according to cultivar and tree manipulation
| Cultivar | Tree manipulation | No. of replications | β | From LBS | ϕ | From LVS |
|---|---|---|---|---|---|---|
| ‘Scarletspur Delicious’ | C | 3 | 0·28 | 0·05 | 0·25 | 0·23 |
| L | 3 | 0·50 | 0·01 | 0·33 | 0·17 | |
| ‘Golden Delicious’ | C | 3 | 0·23 | 0·74 | 0·31 | 0·39 |
| L | 3 | 0·40 | 0·54 | 0·55 | 0·20 | |
| ‘Granny Smith’ | C | 3 | 0·56 | 0·31 | 0·65 | 0·69 |
| L | 3 | 0·76 | 0·42 | 0·65 | 0·45 | |
| Cultivar mean | ||||||
| ‘Scarletspur Delicious’ | 0·36b | 0·03 | 0·29 | 0·17b | ||
| ‘Golden Delicious’ | 0·31b | 0·64 | 0·43 | 0·30b | ||
| ‘Granny Smith’ | 0·66a | 0·36 | 0·65 | 0·57a | ||
| Cultivar effect | ||||||
| F | 447·10 | 15·03 | 8·35 | 38·57 | ||
| P | <0·01 | 0·06 | 0·10 | 0·03 | ||
| Tree manipulation mean | ||||||
| C | 0·35b | 0·37 | 0·40 | 0·44a | ||
| L | 0·54a | 0·32 | 0·51 | 0·25b | ||
| Tree manipulation effect | ||||||
| F | 310·38 | 0·25 | 2·26 | 23·94 | ||
| P | <0·01 | 0·67 | 0·27 | 0·04 |
For both variables, the proportion of 2005-inflorescences issued from long 2004-shoots, i.e. LBS for β, LVS for ϕ, was determined.
Within columns, different superscript letters indicate significant differences according to Duncan's test, P < 0·05.
Bourse-over-bourse (β) was significantly higher for ‘Granny Smith’ compared with ‘Scarletspur Delicious’ and ‘Golden Delicious’. It occurred more frequently from LBS on ‘Golden Delicious’ than on ‘Scarletspur’, with ‘Granny Smith’ in the intermediate position.
There was no significant difference in the trend toward flowering (ϕ̇) between cultivars. However, trend toward flowering occurred significantly more on LVS for ‘Granny Smith’ than for ‘Golden Delicious’ and ‘Scarletspur’.
When a 2004-latent bud developed in 2005, it mainly gave rise to a VS (ω̄) without a significant effect of cultivar or tree manipulation (Table 4). However the proportion of 2005-shoots developing from 2004-latent buds (λ) was significantly affected by tree manipulation although there was no significant effect of the cultivar.
Table 4.
Proportion of 2004-latent buds leading to 2005-vegetative shoots among all latent buds developing in 2005 (ϖ) and proportion of 2005-shoots developing from 2004-latent buds (λ) according to cultivar and tree manipulation
| Cultivar | Tree manipulation | No. of replications | ϖ | λ |
|---|---|---|---|---|
| ‘Scarletspur Delicious’ | C | 3 | 0·81 | 0·19 |
| L | 3 | 0·88 | 0·67 | |
| ‘Golden Delicious’ | C | 3 | 0·48 | 0·00 |
| L | 3 | 0·94 | 0·39 | |
| ‘Granny Smith’ | C | 3 | 0·73 | 0·00 |
| L | 3 | 0·79 | 0·22 | |
| Cultivar mean | ||||
| ‘Scarletspur Delicious’ | 0·85 | 0·43 | ||
| ‘Golden Delicious’ | 0·71 | 0·20 | ||
| ‘Granny Smith’ | 0·76 | 0·11 | ||
| Cultivar effect | ||||
| F | 0·39 | 6·21 | ||
| P | 0·72 | 0·14 | ||
| Tree manipulation mean | ||||
| C | 0·68 | 0·06b | ||
| L | 0·87 | 0·43a | ||
| Tree manipulation effect | ||||
| F | 2·14 | 22·88 | ||
| P | 0·28 | 0·04 |
Within columns, different superscript letters indicate significant differences according to Duncan's test, P < 0·05.
Vegetative shoot length
Due to the low number of long bourse-shoots for ‘Scarletspur Delicious’ (data not shown) and the absence of significant differences between ‘Golden Delicious’ and ‘Granny Smith’ for this trait, only vegetative long shoots were considered. For all cultivars, there was a significant effect of tree manipulation on shoot length, with longer shoots on L-trees compared with C-trees (Table 5). In the former case, the length of vegetative shoots was also dependent on the pruning status of the parent shoot: 2005-LVS branched on 2004-pruned shoots were significantly longer than 2005-LVS branched on 2004-non-pruned shoots (Table 6).
Table 5.
Length of long vegetative shoots (LVS; sample size indicated between parentheses) in 2005 borne on non-pruned parent shoots in 2004, as affected by tree manipulation, for the three cultivars
| Tree manipulation | |||
|---|---|---|---|
| Cultivar | C-Trees | L-Trees | P |
| ‘Scarletspur Delicious’ | 14·97 (n = 65) | 18·30 (n = 98) | * |
| ‘Golden Delicious’ | 15·70 (n = 358) | 18·25 (n = 307) | * |
| ‘Granny Smith’ | 16·69 (n = 317) | 21·11 (n = 94) | * |
Between columns, significant differences according to independent t-test at P < 0·05 are indicated by *.
Spatial distribution of leaf area
The spatial distribution of shoots was affected by both cultivar and tree manipulation, with more leaning and bent shoots in C-trees than in L-trees, leading to higher canopy volume, especially for ‘Golden Delicious’ and ‘Granny Smith’ (Fig. 1). These effects could be quantified by the following variables: total canopy volume, and relative volume occupied by, and distribution of, FS and/or VS.
There was a significant effect of the cultivar on canopy volume, with values for ‘Golden Delicious’ and ‘Granny Smith’ around 3-fold greater than for ‘Scarletspur Delicious’, without significant effect of tree manipulation in both years (Table 7). The relative volume occupied by leaves of FS ranged between 30 and 75 %, and was not significantly influenced either by cultivar or by tree manipulation (Table 7). In contrast, the relative volume occupied by leaves of VS was significantly influenced by the cultivar in both years, with higher values for ‘Scarletspur Delicious’ and ‘Golden Delicious’ compared with ‘Granny Smith’. From 20 to 46 % of canopy volume was simultaneously occupied by FS and VS, with no significant influence of the cultivar or tree manipulation (Table 7). However, the distribution of leaf area density of FS and VS was almost spatially uncorrelated, which indicated some spatial separation in both types of shoots, and without significant effect of cultivar and tree manipulation on these correlations (data not shown).
Table 7.
Total canopy volume per tree, and relative volume occupied by leaves of flowering (FS) and vegetative (VS) shoots separated, and by leaves of both fruiting and vegetative shoots (FS + VS)
| Relative volume | ||||||
|---|---|---|---|---|---|---|
| Cultivar | Tree manipulation | No. of replications | Total tree volume (m3) | FS | VS | FS + VS |
| 2004 | ||||||
| ‘Scarletspur Delicious’ | C | 3 | 1·38 | 0·30 | 0·92 | 0·22 |
| L | 3 | 1·34 | 0·41 | 0·94 | 0·35 | |
| ‘Golden Delicious’ | C | 3 | 4·51 | 0·49 | 0·92 | 0·41 |
| L | 3 | 3·69 | 0·57 | 0·89 | 0·46 | |
| ‘Granny Smith’ | C | 3 | 4·65 | 0·66 | 0·76 | 0·43 |
| L | 3 | 3·19 | 0·51 | 0·83 | 0·34 | |
| Cultivar mean | ||||||
| ‘Scarletspur Delicious’ | 6 | 1·36b | 0·36 | 0·93a | 0·29 | |
| ‘Golden Delicious’ | 6 | 3·92a | 0·53 | 0·90a | 0·39 | |
| ‘Granny Smith’ | 6 | 4·10a | 0·59 | 0·80b | 0·43 | |
| Cultivar effect | ||||||
| F | 10·48 | 2·11 | 4·64 | 1·93 | ||
| P | <0·01 | 0·16 | 0·03 | 0·19 | ||
| Tree manipulation mean | ||||||
| C | 9 | 3·52 | 0·49 | 0·87 | 0·35 | |
| L | 9 | 2·74 | 0·50 | 0·89 | 0·38 | |
| Tree manipulation effect | ||||||
| F | 2·03 | 0·01 | 0·30 | 0·27 | ||
| P | 0·18 | 0·91 | 0·59 | 0·62 | ||
| 2005 | ||||||
| ‘Scarletspur Delicious’ | C | 3 | 1·70 | 0·59 | 0·85 | 0·44 |
| L | 3 | 1·71 | 0·56 | 0·87 | 0·43 | |
| ‘Golden Delicious’ | C | 3 | 4·94 | 0·51 | 0·90 | 0·41 |
| L | 3 | 4·77 | 0·51 | 0·92 | 0·43 | |
| ‘Granny Smith’ | C | 3 | 6·72 | 0·75 | 0·57 | 0·32 |
| L | 3 | 3·73 | 0·64 | 0·68 | 0·32 | |
| Cultivar mean | ||||||
| ‘Scarletspur Delicious’ | 6 | 1·70b | 0·58 | 0·91a | 0·44a | |
| ‘Golden Delicious’ | 5 | 4·84a | 0·51 | 0·86a | 0·42a | |
| ‘Granny Smith’ | 6 | 5·22a | 0·70 | 0·62b | 0·32b | |
| Cultivar effect | ||||||
| F | 8·25 | 1·62 | 4·04 | 4·17 | ||
| P | <0·01 | 0·24 | 0·04 | 0·04 | ||
| Tree manipulation mean | ||||||
| C | 8 | 4·39 | 0·61 | 0·76 | 0·34 | |
| L | 9 | 3·40 | 0·59 | 0·82 | 0·99 | |
| Tree manipulation effect | ||||||
| F | 1·77 | 0·09 | 0·27 | 0·01 | ||
| P | 0·21 | 0·49 | 0·61 | 0·86 | ||
Within columns, different superscript letters indicate significant differences according to Duncan's test, P < 0·05.
Cultivars were clearly discriminated by vertical and radial profiles of leaf area, with half of total leaf area below 1 m and within 1 m from the trunk in ‘Scarletspur Delicious’, whereas it was between 1 and 1·8 m from the ground and between 1 and 1·5 m from the trunk for ‘Golden Delicious’ and ‘Granny Smith’ (P < 0·01 for both profiles, data not shown; see Figs 2 and 3). Since for both leaf area distributions the respective effects of year, treatment and shoot type differed between cultivars, analyses were carried out on each cultivar independently.
Fig. 2.
Vertical profile of leaf area of flowering (FS) and vegetative (VS) shoots according to tree manipulation (C, Centrifugal training; L, Central Leader), and year (2004, 2005), and for three cultivars, ‘Scarletspur Delicious’ (A), ‘Golden Delicious’ (B) and ‘Granny Smith’ (C). Vertical bars (mean per tree ± s.d. when different from 0; n = 3) represent the height reached by 50% of the cumulated leaf area from the orchard floor. Results of the ANOVA are given for each graph.
Fig. 3.
Radial profile of the leaf area of flowering (FS) and vegetative (VS) shoots according to tree manipulation (C, Centrifugal training; L, Central Leader), and year (2004, 2005), and for three cultivars, ‘Scarletspur Delicious’ (A), ‘Golden Delicious’ (B) and ‘Granny Smith’ (C). Horizontal bars (mean per tree ± s.d. when different from 0; n = 3) represent the distance from the trunk at crown base reached by 50% of the cumulated leaf area. Results of the ANOVA are given for each graph.
Concerning the vertical profile of leaf area, the only significant effect in ‘Scarletspur Delicious’ was shoot type with VS above FS (Fig. 2A), whereas in ‘Granny Smith’ it was year, with a higher canopy in 2005 compared with 2004 (Fig. 2C). In ‘Golden Delicious’, all variables had significant effects, with higher trees in 2005 compared with 2004, higher C-trees compared with L-trees, and higher VS compared with FS (Fig. 2B).
Concerning the radial profile of leaf area, there was no significant effect of year (Fig. 3A–C). ‘Scarletspur Delicious’ was characterized by a significant effect of shoot type, with an outward development of VS compared with FS (Fig. 3A). ‘Granny Smith’ was characterized by a significant effect of tree manipulation, with a larger canopy for C-trees compared with L-trees (Fig. 3C). There was no effect of either variable on the radial distribution of leaf area for ‘Golden Delicious’ (Fig. 3B).
DISCUSSION
The respective effects of cultivar and tree manipulation were analysed on several architectural traits, especially flowering, of apple trees over 2 years using a digitizing method which permitted the creation of a database of shoot types and position in space.
There was a significant effect of the cultivar on shoot demography and 2-year shoot sequence. These results confirmed previous findings describing apple ideotype characteristics (e.g. higher bourse-over-bourse in type IV compared with type II cultivars; Looney and Lane, 1984; Lespinasse, 1992; Lauri and Lespinasse 1993; Lespinasse and Delort, 1993; Lauri et al., 1997a). The present study further showed that the number of bourse-shoots per bourse could be suggested as another discriminating factor among apple cultivars.
A significant effect of tree manipulation was shown on the behaviour of latent buds and shoot length. L-trees were characterized by a significantly higher proportion of 2005-shoots developed from 2004-latent buds. This new shoot population, typically composed of long vegetative shoots, originated mainly on the side of shoots that were subjected to heading cuts in the previous winter. From an architectural point of view, regular heading cuts would therefore induce the formation of reiterated complexes (Hallé et al., 1978; Fisher and Hibbs, 1982), which renewed the structural vegetative framework of the tree. Tree manipulation could also affect the flowering pattern, and Centrifugal training has been shown to increase return-bloom at whole-tree level and especially bourse-over-bourse (Lauri et al., 2004). This was not observed in the present study, and could be attributed to the fact that in the first year of the experiment tree manipulation on C-trees removed older flowering shoots in the tree centre that had a greater capacity for bourse-over-bourse compared with the younger reproductive shoots on the external part of the canopy (Lauri et al., 1997a).
The distribution of leaves in a fruit tree canopy has usually been studied through the use of horizontal and vertical transects across the canopy (e.g. Sansavini and Corelli, 1992) or a laser-assisted canopy scanning device (e.g. Wünsche and Lakso, 2000), which permit only a 2D representation of the tree canopy. In the present study, whole-tree digitizing, although involving higher labour costs than the above-mentioned studies, proved to be efficient to obtain detailed information on the 3D distribution of shoot and leaf area. It was shown here that the architectural disjunction between vegetative and flowering shoots characterizing type II cultivars to which ‘Scarletspur Delicious’ belongs (i.e. flowering occurs on poor vegetative growth; Lauri and Laurens, 2005; Costes et al., 2006) corresponded to a spatial disjunction between shoot types, with vegetative shoots in the outer part of canopy and flowering shoots in the inner part with no influence of tree manipulation (Figs 2A and 3A). A similar trend was noticed in the type III cultivar, ‘Golden Delicious’. On the other hand, the distribution of vegetative and flowering shoots was affected by tree manipulation in type IV cultivar ‘Granny Smith’: vegetative shoots were stimulated in the periphery of Central Leader trees, whereas flowering shoots were stimulated in the periphery of Centrifugal-trained trees (Figs 2C and 3C), with a significant radial development of C-trees compared with L-trees. Both traits were related to the flowering behaviour of ‘Granny Smith’ (‘tip-bearing’ cultivar with a high proportion of bent shoots; Lespinasse, 1977; Forshey et al., 1992). These contrasted reactions to tree manipulation bring more support to the idea that to optimize fruiting of the apple tree, training systems should be adapted to the cultivar intrinsic growth and flowering habit (Lauri and Laurens, 2005).
In conclusion, this study clearly showed that although each cultivar has its own architectural characteristics, tree manipulation may significantly affect shoot type population and shoot leaf area distribution in space. The study therefore suggested that research on canopy architecture and eventually light climate (Wünsche and Lakso, 2000; Willaume et al., 2004), especially in relation to flowering (Sansavini and Corelli, 1992; Palmer and Warrington, 2000), should integrate tree cultivar-related architectural reactions to pruning and training procedures.
ACKNOWLEDGEMENTS
We thank the American University of Beirut, and especially Dr Mohamed Farran for orchard monitoring and availability, and the students Dahlia Mansour, Gebran Nassif, Marica Abi Nader and Majed Feghali for tree digitizing.
LITERATURE CITED
- Adam B. POL95–software to drive a Polhemus Fastrak 3 SPACE 3D digitizer. Version 1·0. Clermont-Ferrand, France: UMR PIAF INRA-UBP; 1999. [Google Scholar]
- Adam B, Donès N, Sinoquet H. VegeSTAR–software to compute light interception and canopy photosynthesis from images of 3D digitised plants. Version 3·0. Clermont-Ferrand, France: UMR PIAF INRA-UBP; 2002. [Google Scholar]
- Barden JA, Marini RP. Growth and cropping of young apple trees as affected by heading and support of the central leader. Hortscience. 1998;33:39. [Google Scholar]
- Barritt B. Intensive orchard management. Yakima, WA: Good Fruit Grower; 1992. [Google Scholar]
- Bernhard R. Congrès pomologique INRA. Paris: 1961. Octobre. Mise à fleur et alternance chez les arbres fruitiers; pp. 91–116. 1960. [Google Scholar]
- Costes E, Lauri PÉ, Lespinasse JM. Modélisation de la croissance et de la ramification chez quelques cultivars de pommier. Editions INRA, Les Colloques. 1995;74:27–39. [Google Scholar]
- Costes E, Sinoquet H, Kelner JJ, Godin C. Exploring within-tree architectural development of two apple tree cultivars over 6 years. Annals of Botany. 2003;91:91–104. doi: 10.1093/aob/mcg010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Costes E, Lauri PÉ, Regnard JL. Analysing fruit tree architecture. Implication for tree management and fruit production. Horticultural Reviews. 2006;32:1–61. [Google Scholar]
- De Wit I, Cook NC, Keulemans J. Characterization of tree architecture in two-year-old apple seedling populations of different progenies with a common columnar gene parent. Acta Horticulturae. 2004;663:363–368. [Google Scholar]
- Elfving D. Growth and productivity of ‘Empire’ apple trees following a single heading-back pruning treatment. Hortscience. 1990;25:908–910. [Google Scholar]
- Elfving D, Forshey CG. Growth and fruiting response of vigorous apple branches to pruning and branch orientation treatments. Journal of the American Society for Horticultural Science. 1976;101:290–293. [Google Scholar]
- Ferree DC, Palmer JW. Effect of spur defoliation and ringing during bloom on fruiting, fruit mineral level, and net photosynthesis of ‘Golden Delicious’ apple. Journal of the American Society for Horticultural Science. 1982;107:1182–1186. [Google Scholar]
- Fisher JB, Hibbs DE. Plasticity of tree architecture: specific and ecological variations found in Aubreville's model. Canadian Journal of Botany. 1982;69:690–702. [Google Scholar]
- Forshey CG, Elfving DC, Stebbins RL. Training and pruning of apple and pear trees. Alexandria, VA: American Society for Horticultural Science; 1992. [Google Scholar]
- Hallé F, Oldeman RAA, Tomlinson PB. Tropical trees and forest. An architectural analysis. New York: Springer-Verlag; 1978. [Google Scholar]
- Heinicke DR. High-density apple orchards–planting, training and pruning. US Department of Agriculture Handbook 458; 1975. [Google Scholar]
- Hoying SA, Robinson TL. The apple orchard planting system puzzle. Acta Horticulturae. 2000;513:257–260. [Google Scholar]
- Johnson RS, Lakso AN. A carbon balance model of a growing apple shoot. II–Simulated effects of light and temperature on long and short shoots. Journal of the American Society for Horticultural Science. 1986;111:164–169. [Google Scholar]
- Lauri PÉ. From tree architecture to tree training–an overview of recent concepts developed in apple in France. Journal of the Korean Society for Horticultural Science. 2002;43:782–788. [Google Scholar]
- Lauri PÉ, Kelner JJ. Shoot type demography and dry matter partitioning: a morphometric approach in apple (Malus × domestica) Canadian Journal of Botany. 2001;79:1270–1273. [Google Scholar]
- Lauri PÉ, Laurens F. Architectural types in apple (Malus × domestica Borkh.) In: Ramdane D, editor. Crops: growth, quality and biotechnology. Helsinki: World Food Limited; 2005. pp. 1300–1314. [Google Scholar]
- Lauri PÉ, Lespinasse JM. The relationship between cultivar fruiting type and fruiting branch characteristics in apple trees. Acta Horticulturae. 1993;349:259–263. [Google Scholar]
- Lauri PÉ, Térouanne É. Analyse de la croissance primaire de rameaux de pommier (Malus domestica Borkh.) au cours d'une saison de végétation. Canadian Journal of Botany. 1995;73:1471–1489. [Google Scholar]
- Lauri PÉ, Térouanne É. The influence of shoot growth on the pattern of axillary development on the long shoots of young apple trees (Malus domestica Borkh.) International Journal of Plant Sciences. 1998;159:283–296. [Google Scholar]
- Lauri PÉ, Térouanne É. Effects of inflorescence removal on the fruit set of the remaining inflorescences and development of the laterals on one year old apple (Malus domestica Borkh.) branches. Journal of Horticulture Science and Biotechnology. 1999;74:110–117. [Google Scholar]
- Lauri PÉ, Trottier C. Patterns of size and fate relationships of contiguous organs in the apple (Malus domestica) crown. New Phytologist. 2004;163:533–546. doi: 10.1111/j.1469-8137.2004.01136.x. [DOI] [PubMed] [Google Scholar]
- Lauri PÉ, Térouanne É, Lespinasse JM, Regnard JL, Kelner JJ. Genotypic differences in the axillary bud growth and fruiting pattern of apple fruiting branches over several years–an approach to regulation of fruit bearing. Scientia Horticulturae. 1995;64:265–281. [Google Scholar]
- Lauri PÉ, Térouanne É, Lespinasse JM. Quantitative analysis of relationships between inflorescence size, bearing-axis, size and fruit-set–an apple tree case study. Annals of Botany. 1996;77:277–286. [Google Scholar]
- Lauri PÉ, Térouanne É, Lespinasse JM. Relationship between the early development of apple fruiting branches and the regularity of bearing–an approach to the strategies of various cultivars. Journal of Horticultural Science. 1997a;72:519–530. [Google Scholar]
- Lauri PÉ, Térouanne É, Lespinasse JM. Vegetative growth and reproductive strategies in apple fruiting branches–an investigation into various cultivars. Acta Horticulturae. 1997b;451:717–724. [Google Scholar]
- Lauri PÉ, Willaume M, Larrive G, Lespinasse JM. The concept of centrifugal training in apple aimed at optimizing relationship between growth and fruiting. Acta Horticulturae. 2004;636:35–42. [Google Scholar]
- Lespinasse JM. La conduite du pommier. I–Types de fructification–incidence sur la conduite de l'arbre. Paris: INVUFLEC; 1977. [Google Scholar]
- Lespinasse JM. La conduite du pommier. II–L'axe vertical, la rénovation des vergers (2ème partie) Paris: Ctifl; 1980. [Google Scholar]
- Lespinasse Y. Breeding apple tree: aims and methods. In: Rousselle-Bourgeois F, Rousselle P, editors. Proceedings of the joint conference of the E.A.PR. breeding and varietal assessment section and the E.UC.A.R.P.I.A potato section; January 1992; Landerneau, France. Ploudaniel, France: INRA: 1992. pp. 103–110. [Google Scholar]
- Lespinasse JM, Delort F. Apple tree management in vertical axis: appraisal after ten years of experiments. Acta Horticulturae. 1986;160:120–155. [Google Scholar]
- Lespinasse JM, Delort F. Regulation of fruiting in apple–role of the bourse and crowned brindles. Acta Horticulturae. 1993;349:239–246. [Google Scholar]
- Looney NE, Lane WD. Spur type growth mutants of McIntosh apple: a review of their genetics, physiology and field performance. Acta Horticulturae. 1984;146:31–46. [Google Scholar]
- Palmer JW, Warrington IJ. Underlying principles for successful apple planting systems. Acta Horticulturae. 2000;513:357–363. [Google Scholar]
- Phattaralerphong J, Sinoquet H. A method for 3D reconstruction of tree crown volume from photographs: assessment with 3D-digitized plants. Tree Physiology. 2005;25:1229–1242. doi: 10.1093/treephys/25.10.1229. [DOI] [PubMed] [Google Scholar]
- Proctor JTA, Palmer JW. The role of spur and bourse leaves of three apple cultivars on fruit set and growth and calcium content. Journal of Horticultural Science. 1991;66:275–282. [Google Scholar]
- Reche A, Martin I, Drettakis G. Volumetric reconstruction and interactive rendering of trees from photographs. ACM Transactions on Graphics (SIGGRAPH Conference Proceedings) 2004;23:1–10. [Google Scholar]
- Rom CR, Barritt B. Spur development of ‘Delicious’ apple as influenced by position, wood age, strain and pruning. Hortscience. 1990;25:1578–1581. [Google Scholar]
- Rom CR, Ferree DC. Spur leaf characteristics of nine apple cultivars. Fruit Varieties Journal. 1984;38:2–5. [Google Scholar]
- Sansavini S, Corelli L. Canopy efficiency of apple as affected by microclimate factors and tree structure. Acta Horticulturae. 1992;322:69–77. [Google Scholar]
- Sinoquet H, Rivet P. Measurement and visualization of the architecture of an adult tree based on a three-dimensional digitizing device. Trees. 1997;11:265–270. [Google Scholar]
- Sonohat G, Sinoquet H, Kulandaivelu V, Combes D, Lescourret F. Three-dimensional reconstruction of partially 3D-digitized peach tree canopies. Tree Physiology. 2006;26:337–351. doi: 10.1093/treephys/26.3.337. [DOI] [PubMed] [Google Scholar]
- Tustin DS, Hirst PM, Warrington IJ. Influence of orientation and position of fruiting laterals on canopy light penetration, yield and fruit quality of ‘Granny Smith’ apple. Journal of the American Society for Horticulture Science. 1988;113:693–699. [Google Scholar]
- Willaume M, Lauri PÉ, Sinoquet H. Light interception in apple trees influenced by canopy architecture manipulation. Trees. 2004;18:705–713. [Google Scholar]
- Wünsche JN, Lakso AN. The relationship between leaf area and light interception by spur and extension shoot leaves and apple orchard productivity. HortScience. 2000;35:1202–1206. [Google Scholar]