Modelling of colour development in the fruit of Actinidia chinensis ‘Hort16A’

Colour development in fruit of Actinidia chinensis var. chinensis ‘Hort16A’, the new yellowfleshed cultivar produced commercially in New Zealand, was monitored during three seasons in four 42 New Zealand Journal of Crop and Horticultural Science, 2003, Vol. 31 The pericarp of ‘Hort16A’ fruit changes from a light green colour to a light yellow colour as fruit develop on the vine. Virtually no colour change occurs after fruit are harvested and put in cold storage. For this reason, commercial clearance for picking ‘Hort16A’ has been linked to the flesh colour, which is measured with a chromometer or by comparison with colour charts. Fruit that are still green are unacceptable in the market, and it is desirable to produce fruit of uniform yellow flesh colour, which is quantified by the hue angle (see below). Therefore the shape of flesh hue angle distributions, and in particular the proportion of the fruit that are still green, is of importance in determining clearance to pick. Historically, limits have been set for the numbers of green fruit that are allowable in preharvest samples. However this approach does not always give a reliable measure of the population. McGhie & Ainge (2001) have found that the colour of ‘Hort16A’, and most other kiwifruit, is derived from two pigment groups, chlorophyll and the carotenoids. Also they reported that mature fruit of ‘Hort16A’ contain carotenoids dominated by bcarotene 0.17 mg/100 g fresh weight (FW), lutein 0.11 mg/100 g FW, and esterified carotenoids 0.24 mg/100 g FW, but contained negligible quantities of chlorophyll. Therefore it is likely that the colour change in ‘Hort16A’ is primarily because of chlorophyll degradation resulting in un-masking of carotenoid pigments. Loss of chlorophyll pigment is common among fruits undergoing maturation, usually as chloroplasts are converted to chromoplasts rich in carotenoid pigments (Marano et al. 1993). The degeneration of chlorophyll pigments in the chloroplasts is usually associated with enzymatic degradation by chlorophyllase, this can also be induced by environmental stress such as temperature, light, nutrition, and phytohormones (Gross 1987; Rudiger & Schoch 1998). The aim of this paper is to present a quantitative description of the seasonal development of the average hue angle, as well as the distribution of individual fruit’s hue angle at any particular time. MATERIALS AND METHODS Changes in flesh colour during maturation were monitored on five ‘Hort16A’ kiwifruit vines growing at The Horticulture & Food Research of New Zealand Ltd orchards in each of the four main regions of New Zealand where ‘Hort16A’ is grown commercially, between 1998 and 2000. The orchards were located near Kerikeri (35°10¢S), Te Puke (37°49¢S), Havelock North (39°40¢S), and Riwaka (41°06¢S). To minimise any influence of seedling rootstocks, scions of ‘Hort16A’ were grafted onto clonal A. deliciosa (A. Chev) C.F. Liang et A.R. Ferguson ‘Chieftan’ rootstock in 1995. Vines were trained on T-bar trellises and managed using standard commercial management practices with cane replacement in winter (Sale 1983). Fruit were monitored during the 1998, 1999, and 2000 seasons. To estimate the time of mid bloom, the number of open flowers at each site was estimated each 3 or 4 days throughout the flowering period. Estimates were based upon a total of 10 canes, taking two canes per vine. Individual canes were randomly selected during spring, and flowering monitored on three shoots spaced along each cane (Snelgar et al. 2002). Flowers were considered to be open when styles and filaments were exposed to honeybee access. The date of mid bloom for the vine was considered to be when 50% of the flowers were open and this was calculated by fitting a cumulative Normal distribution to the data (Snelgar et al. 2002). To follow the development of flesh colour, three average sized fruit were sampled from each of five replicate vines at each site within the curtain of the T-bar, avoiding fruit that were very exposed, near the tips of canes or very close to the main leader. A 2-mm-thick layer of skin and cortical tissue was removed from the fruit to expose a flat, even surface in the outer pericarp, to avoid measuring the flesh immediately beneath the skin where colour is variable because of uneven exposure to direct light. Colour of the exposed flesh was then measured immediately using a chromometer CR300 (Minolta Camera Co. Ltd) fitted with a glass light projection tube CR-A33f under CIE illuminant D65 lighting conditions. Calibration of the chromometer was conducted before each use with a standard white plate (Minolta Camera Co. Ltd). Colour was expressed in the L*C*h° notation (McGuire 1992). Changes in the within-vine distribution of fruit flesh colour were monitored on a single commercial orchard at Te Puke. The 10 mature vines used were stump grafted with ‘Hort16A’ scions in 1997 and trained on pergola trellises. Vines were arranged in a “stip-male” configuration with female vines in every second row and male vines in intervening rows pruned to a tight strip between each female row. This arrangement resulted in a zone on the edge of female canopies where fruit were relatively well exposed to light due to strip-male rows being pruned hard after 43 Minchin et al.—Modelling colour development in ‘Hort16A’ flowering and female canes terminating at variable positions. Vines were managed using a conventional 1-year-old cane replacement strategy but there were a number of fruit present on short spurs near to the leader. Because of the potential differences between fruit from near the leader, those in the main fruiting canopy and those on the edge of the canopy, the canopy was segregated into three zones. The leader zone was 1 m each side of the leader and mainly included fruit borne on spurs. The main fruiting zone was a 2-m strip on each side of the vine, and included fruit beyond the leader zone from the middle of 1year-old canes. The edge zone was a c.1-m-wide strip on each side of the vine beyond the fruiting zone. Fruit samples were collected at weekly intervals from early April (4 April, 11, 18, 26, and 8 May) until fruit reached the commercial clearance to pick standard. From each of the three canopy zones on each vine, 10 fruit were randomly selected from over the entire length of the vine and colour measurements were made on two surfaces of each fruit as already described. When fruit were harvested from vines the number of fruit in each of the canopy zones on each vine was recorded in order to allow weighting of colour data. Air temperatures, in a Stevenson screen mounted 3 m above ground level, were obtained from standard meteorological sites, located on each orchard. Daily maxima and minima were calculated from a 600 s running average of measurements made every minute between 0900 and 0900 h the next day. RESULTS AND DISCUSSION Modelling the change in average hue angle during fruit maturation The change in average hue angle during fruit growth and maturation is shown in Fig. 1. Until c. 120 DAMB hue angle remains fairly constant at c. 115° and begins to fall rapidly after c. 150 DAMB. The drop in hue angle continues for c. 50 days to reach a lower asymptotic value of c. 97°. The general shape of this curve is sigmoid with lower and upper bounds of c. 97° and 115° respectively. Consequently, hue Fig. 1 Comparison of fitted curves of the complementary log-log and Boltzman functions to hue angle data from 1998/99 season across four sites (Havelock North, Kerikeri, Riwaka, and Te Puke, New Zealand). 44 New Zealand Journal of Crop and Horticultural Science, 2003, Vol. 31 values were scaled (y) to lie in the range (0, 1) using the linear transformation: y = (hue – 97)/18 (1) Empirical equations were fitted to this scaled data.