Guava (Psidium guajava L.) is one of the most popular tropical and subtropical fruit crops grown in India owing to its several health promoting properties and value-addition avenues. It is well known fact that guava has two distinct botanical characteristics; one is the flowers are always borne on newly emerging vegetative shoots, irrespective of the time of year (Rathore & Singh, 1974; Singh, 1985). This feature makes guava unique, that it can be pruned as severely as temperate fruit tree (Lotter, 1990) for high density management. Several workers reported the beneficial effects of pruning on yield and fruit quality of guava (Jadhav et al., 1998; Singh & Singh, 2001; Dhaliwal & Kaur, 2003; Dhaliwal & Singh, 2004). Second is that guava has more than one bearing season (Singh & Kumar, 1993). These two features provide an opportunity to regulate guava crop through pruning along with high density management.

In tarai region of India, three flowering seasons are very common, viz. April-May (for rainy season crop), July-August (for winter season crop) and October-November (spring season crop) (Singh & Kumar, 1993). During winter season, the flowering and vegetative growth is almost negligible due to low temperature (Chadha & Pandey, 1986). As a result plants accumulate sufficient food reserve, which results in maximum new vegetative growth in the following spring due to optimum temperature. This vegetative flush produces floral buds which produces flower during summer season (40 days after floral initiation)for rainy season crop (Sehgal & Singh, 1967). The production is being maximum during the rainy season (Dwivedi et al., 1990). However, the fruits produced during rainy season are severely attacked by fruit fly (Stonehouse et al., 2002) which leads significant loss in fruit production and it also have poor nutritive value and keeping quality. On the other hand, winter season crop is superior in quality, free from the pest and diseases, having long storage life and fetches more prices in the market as compared to the rainy season crop (Rathore & Singh, 1976). By keeping the above mentioned points in mind, it is beneficial to take winter season crop mainly. Crop regulation in guava is also used in other parts of world like in Hawaii and Kauai, where it is known as cycling (Bittenbender & Kobayashi, 1990). Pruning can be used for crop regulation (Lal, 1992). Pruning has its physiological effects basically due to changes in the partitioning of the reserves. It changes sink preference for allocation of photosynthates. Depending upon the time of the year, the extent and frequency of pruning, some sites of accumulation will disappear and others will be created. As a result, changes in seasonal fluctuations of reserves can appear as well (Clair et al., 1999). In this way, pruning helps in both ways, firstly to regulate crop (Kindo, 2005) and secondly to manage high density (Kaur & Dhaliwal, 2001). Standard spacing for guava is 6 m × 6 m. Whereas, high density planting consists of planting at 3 m × 1.5 m, 3 m × 3 m and 6 m × 3 m. Meadow orcharding which is an ultra-high density planting accommodates 5000 plants/ha planted at 2 m × 1 m distance (Singh, 2008). These densities are for either square planting systems or rectangular planting system. But, guava can be planted in other planting systems also with higher densities as compare to square system like paired system, hedge row system, double hedge row system and cluster system. Out of these systems, double hedge row system is higher in density accommodating 20 trees in a plot of 24 m × 24 m (2.22 times more than square system of planting) (Lal et al., 2007). Hence, this work has been carried out to assess profitability of crop regulation methods in high density management.

The study was conducted at Horticulture Research Centre, Patharchatta, Govind Ballabh Pant University of Agriculture and Technology, Pantnagar, Uttarakhand. Pantnagar is situated at the foothills of the Himalayas (29° N 79.3° E). The altitude of the place is 243.84 msl. Pantnagar has a humid sub-tropical climate with hot humid summers and cold winters. The maximum and minimum temperature range 33° − 42°C and 4° − 8°C during summer and winter respectively. The soil texture of experimental field is sandy loam with pH 7.6. The experiment was planted under double-hedge row system of planting, accommodating 20 trees in a plot of 24 m × 24 m. In double hedge row system of planting, there was pair of rows. Each pair of row situated 8 m apart and within pair of rows, the distant between row to row and plant to plant was 4 m (Fig. S1 [supplementary]).

The study was conducted on 6-yr old guava trees, cv. ‘Pant Prabhat’. The treatments consisted of seven different forms of pruning: FBT, FBTT, RLFO, RLF, OLPS, FSP, OLPF and control (Table 1). All the trees were maintained under uniform cultural practices during entire course of investigation.

Observations were recorded for new shoot emergence, tree growth (height, crown spread, trunk diameter and tree volume), physical fruit variables and yield per tree. New shoot emergence was counted on selected four branches before and after application of the treatments. The average data recorded from four branches is presented as new shoot emergence per branch for rainy as well as for winter season crop (fruiting season July-September and December-February, respectively). The observations on tree growth variables were recorded before application of treatments (i.e. during April) and in winter season (i.e. during January) and annual increase in the tree height, spread, volume and trunk diameter was calculated as per standard procedures. Physical fruit variables, viz. fruit weight, diameter, length and volume were measured as per the standard procedures. Fruit yield per tree was computed by multiplying the total number of fruits on each tree with the mean fruit weight for that tree for rainy and winter season crop separately. A representative sample of ten fruits per treatment per replication was taken randomly from all directions of the tree to take observations on physical variables of fruits for both seasons.

C/B ratio for all the treatments were calculated by considering all inputs and fruit yield during both the years (i.e. 2009-10 and 2010-11).

The minimum new shoot emergence per branch for winter season crop recorded in control (C) might be due to non-disturbance in the apical dominance of the growing shoots. In guava, flowers are always borne on newly emerging vegetative shoots irrespective of the time of year (Rathore & Singh, 1974), due to which it is suitable for pruning for various purposes. Decapitation usually results in the growth of one or more of the lateral buds due to removal of apical dominance. The various extent of new shoot emergence depends on the fact that how many lateral buds were present after pruning, which depends upon the severity of pruning (Tiwari, 1985; Lal, 1992). Apical dominance is associated with more vertical growth. When apical dominance disturbed by any means, then plant corrects the change (Acquaah, 2002) and results in change in tree canopy size due to new shoot emergence. The treatments OLPS and OLPF produced new shoots from leaf axils (if fruits were not there) which resulted in more lateral growth. This is the reason behind minimum annual increase in tree height and volume in treatments OLPS and OLPF. The treatment FSP recorded maximum annual increase in tree height because full shoot pruning forced the axillary buds, present on old shoots, to sprout and grow, as it was reported previously by Chandra & Govind (1995), Jadhav et al. (1998) and Kaur & Dhaliwal (2001). The annual increase in tree volume was half in case of treatments OLPS and OLPF as compared to other treatments.

Application of treatments reduced rainy season yield. It is due to fact that the treatments were applied in the last week of April and first fortnight of May, coinciding with the flowering season for rainy season crop in tarai regions (Singh & Kumar, 1993). When the harvest of rainy and winter seasons is compared, it was clear that the control (untreated) produced heavy fruiting during rainy season, with meagre crop in winter season. While, the remaining treatments comprising severe pruning (except RLF and FSP), produced higher yield in winter season. This confirms the fact that it is the way and extent of pruning which decides the yield in the forthcoming season by creation and destruction (reduction of crown volume, foliage removal and new sinks) of food reserves (Clair et al., 1999). The treatments RLF and FSP recorded intermediate yield. Lal (1992) and Kindo (2005) also reported similar results.

The effect of treatments on physical fruit variables is due to the fact that apart from affecting the apical dominance, pruning is known to produce changes in the partitioning of the photo-assimilates through modification in source-sink relationship. Pruning time, extent and frequency decides whether new sites are going to be created or existing one will disappear (Clair et al., 1999). Sink strength is the ability to attract metabolite from different sources and decides the direction of flow of photo-assimilates. Sink strength determined by the growth rate (sink activity) and the size of the sinks. Actively growing plant parts are strong sinks (Wolstenholme, 1990). Fruit, as the final growth stage of a reproductive organ, is commonly a strong sink for assimilates, at the expense of vegetative growth (Bollard, 1970). Once fruits start to develop, both the direction and pathway of assimilate transport change in favour of fruit growth (Ho & Hewett, 1986). In control trees, where more number of strong sinks (fruit) was present for rainy season crop, they produced less vegetative growth (new shoots) for winter season bearing. When the treatments were applied, the potential sinks (fruits) were removed, and shifted the flow of photosynthates towards new growing shoots, as potential sinks. In treatments where fruits were present, by adopting partial removal of fruits, the existing leaf area was supporting less number of developing fruits, making more photosynthates available for each fruit unit, leading to increased fruit size (Fischer et al., 2012). Economic analysis indicates that OLPF recorded maximum cost:benefit ratio followed by the treatment OLPS. Tiwari & Lal (2007) also reported that maximum return can be obtained by one leaf pair shoot pruning.

It can be concluded that the high density planting in guava can be maintained profitably by adopting one-leaf pair pruning of fruited shoots only, which results in reduction in the annual increase in tree volume by half as compared to un-pruned trees and recorded the highest cost: benefit ratio (1:2.96) due to higher production of quality fruits during winter season.