Introduction
Snake melon (Cucumis melo L. var. flexuosus; 2n = 2x = 24) is an important cucurbit crop in India, commonly known as ‘Kakri’ or ‘Tar’. The fruits of this non-dessert melon are long and slender in shape and are available in three colours: yellow-green, light green and dark green, and they have furrowed or wrinkled exterior surfaces; young fruits are eaten fresh in salad. The open-pollinated cultivars ‘Punjab Long melon 1’ (PLM1) and ‘Arka Sheetal’ are grown in India. High-yielding F1 hybrid snake melon cultivars have not entered the market. This can be attributed to low genetic diversity available among snake melons in India (Sheshadri and More, Reference Sheshadri and More2002). Other non-dessert types of melon eaten in salad in India are snapmelon (var. momordica), ‘Vellari’ (var. acidulus), ‘Arya’ (var. chate) and ‘Wanga’ (unknown botanical variety). In Sudan, Tibish melon (var. tibish) is eaten raw in salad (Mohamed and Pitrat, Reference Mohamed and Pitrat1999). Our objective was to determine the potential for increasing fruit yield and quality parameters of snake melon through intraspecific crosses of snake melon with representatives of the genetically diverse melon groups (Pitrat et al., Reference Pitrat, Hanelt and Hammer2000).
Experimental
The highly inbred snake melon (var. flexuosus) ‘PLM1’ was crossed with five highly inbred landraces of melon: KP 7 (var. momordica), AM 72 (var. acidulus), ‘Arya 1’ (var. chate), 04-02 (var. tibish) and ‘Punjab Wanga’ (unknown botanical variety).
Evaluation of morphology and productivity of the breeding material was carried out in spring 2008 at the Indian Institute of Vegetable Research, Varanasi, Uttar Pradesh, India. The six parents and their five F1 hybrids were sown in compost, and were transplanted to the field at the three-leaf stage of growth. Each of the three replications containing ten plants was arranged in a randomized complete block design with a row spacing of 3 m and within a row spacing of 0.45 m. Plants were furrow irrigated and fertilized and protected against pathogens and pests according to standard horticultural practices. Five central plants of each entry were used for taking observations. The following traits were recorded: (1) days to marketable fruit maturity, (2) marketable fruit number per vine, (3) fruit length (cm), (4) fruit width (cm), (5) fruit skin colour, (6) fruit flesh colour, (7) fruit taste, (8) fruit texture and (9) consumer preference of fruit traits. The fruit skin and flesh colours were measured using the Royal Horticultural Society Colour Chart. Consumer preference was determined by a panel of eight trained judges who evaluated the following fruit qualities: shape, size, skin colour, flesh colour, taste and texture. The panel consisted of scientists (2), undergraduate students (2) and housewives (4); all were familiar with the crop. Taste (bitter/non-bitter) and texture (soft/crisp) were judged by eating marketable fruit. A nine-point hedonic scale test was used for measuring consumer preference (Amerine et al., Reference Amerine, Pangborn and Rossler1965).
Hybrids PLM1 × 04-02 and PLM1 × ‘Punjab Wanga’ exhibited superior horticultural performance at Varanasi, and were evaluated in 2009 at Ludhiana, Punjab, and at Pasighat, Arunachal Pradesh, for marketable fruits per vine, fruit length and fruit width. Five young (10 d post anthesis) fruits of each entry in each replication were harvested for biochemical assays. Ascorbic acid was bioassayed as described by Bajaj and Kaur (Reference Bajaj and Kaur1981). Total carotenoids were estimated by the method described by Thomas and Joshi (Reference Thomas and Joshi1977). Fe and Zn were measured in diacid digest of fruit samples with an atomic absorption spectrometer (Varian AA 20).
Results and discussion
The values of the various horticultural traits of the best performing F1 hybrids and their parents at Varanasi are given in Tables 1 and 2. Hybrids PLM1 × 04-02 and PLM1 × ‘Punjab Wanga’ exhibited high values of heterosis over PLM1 for the number of marketable fruits per plant (224 and 205%, respectively) and days to marketable fruit maturity ( − 16 and − 5.0%, respectively). High heterosis (71.4%) over PLM1 for fruit carotenoids was observed in the hybrid PLM1 × 04-02, while both hybrids exhibited less heterosis for the ascorbic acid content of fruit (37.8%). Fe and Zn contents of these two hybrids were 1.14 mg/100 g and 1.05 mg/100 g of marketable fruit weight and 0.22 mg/100 g and 0.18 mg/100 g of marketable fruit weight, respectively (Table 2). Heterosis for these traits was not observed. Both hybrids had high consumer preference ratings (Table 2). Both hybrids had elongated wrinkled fruits (19–21 cm long and 2.8–3.1 cm wide), which were crisp and non-bitter with green skin and yellow-green flesh.
Table 1 Fruit traits and ascorbic acid and carotenoid contents of the parents and their hybrids, Varanasi, Uttar Pradesh, India, 2008
LSD, least significant difference.
a Days taken to harvest fruits after anthesis.
b Heterosis values over the commercial PLM1 (P1).
Table 2 Fe and Zn contents and qualitative fruit traits of the parents and their hybrids, Varanasi, Uttar Pradesh, India, 2008
LSD, least significant difference.
a Royal Horticultural Society Colour Chart was used for the identification of fruit skin and flesh colours.
b Heterosis values over the commercial PLM1 (P1).
During field evaluation in the trial conducted in Punjab, PLM1 × 04-02 as well as 04-02 was observed to be resistant to leafminer [Liriomyza trifolii (Burgess)]. Other Tibish accessions have been found to possess resistance to this insect (El Tahir and Taha Yousif, Reference El Tahir and Taha Yousif2004). Dominant resistance to leafminer was reported in Nantais Oblong (var. cantalupensis; Dogimont et al., Reference Dogimont, Bordat, Pages, Boissot and Pitrat1999).
Snake melon is very popular among the travellers in India especially during the summer as its consumption helps to quench the thirst. The marketable fruits of the two hybrids possessed moisture (94%) similar to that available in PLM1 (Table 1). The two hybrids performed similarly at another warm, dry location in the Punjab state of northern India. At this location, PLM1 × 04-02 and PLM1 × ‘Punjab Wanga’ yielded 16.7 and 15.5 marketable fruits per vine, respectively, and these two hybrids had a high consumer preference rating. Fruit length (20.4 and 19.5 cm) and breadth (3.1 and 3.2 cm) were also similar to those observed in the trial conducted at Varanasi. The two hybrids did not have a yield advantage in a trial conducted in the humid tropics (high rainfall) of Arunachal Pradesh, an eastern state of India (data not presented), probably because of high incidence of powdery mildew [Podosphaera xanthii (Castagne) Braun and Shishkoff] and downy mildew [Pseudoperonospora cubensis (Berk et Curtis) Rostovzev]. More multi-location testing of these two hybrids is required.
Acknowledgements
We are grateful to Professor G. J. Jellis, Dr James D. McCreight and Dr Harry S. Paris for their useful comments.
