Plant Cultivation and UV-blocking- Seeds of red amaranth (cv. BARI Lal shak-1) were sown in raised beds of 100 × 100 × 50 cm (L×B×H) during October to December 2013 at the experimental field of Sher-e-Bangla Agricultural University, Dhaka, Bangladesh. Application of fertilizers, irrigation, and other operations were followed by FRG. Fifteen plastic tunnels of 140 × 140 × 60 cm (L×B×H) were prepared after the germination (6 DAS) of the seedlings, and three tunnels each were covered with different polyolefin (PO) films (0.13-mm thickness) which can block UV radiation shorter than 400 (<400), 360 (<360), 350 (<350), and 340 nm (<340), respectively (Mitsubishi Plastics Agri Dream, Tokyo, Japan). The remaining three tunnels were also covered with a UV-transmitting PO film. The first 15 cm (20%) from the soil level remained open to control heat in the tunnels and allow the invasion by insects. At 30 DAS, the seedlings were harvested and plant growth was measured. Measurement of Environmental Conditions- Temperature, humidity, and visible and UV-light irradiations were measured daily during the experiment. Both the temperature and humidity were recorded at 8.00h, 12:00h, and 24:00h, while visible and UV-light irradiations were measured at 12.00h. Chlorophyll Reading and Total Soluble Solid Content (TSS)- Five plants were randomly chosen from each tunnel and were used for the measurements of chlorophyll, TSS, and leaf color. Each measurement was repeated three times. Leaf chlorophyll reading at the mid-portion of the plants was taken just before harvesting of the crops with a chlorophyll meter (SPAD 502 plus, Konika Minolta, Tokyo, Japan) and is shown as a SPAD value. TSS of the leaves was also measured at the mid portion of the plants with a refractometer (ERMA, Tokyo, Japan), and is shown as °brix. Leaf Color and Total Anthocyanin- The plants were then divided into the stem (5 cm above the ground level), upper leaves, and lower leaves, and a color scale of each plant part was measured using a color meter (NF333, Nippon Densyoku, Tokyo, Japan). Ten leaves per tunnel were then harvested from the distal end, and three discs per leaf were collected using a 1.5 cm cork borer. The combined disks were mashed in a mortar with 15 mL of MeOH: HCl (99:1 v/v; pH 4.5) at an ambient temperature, filtration was conducted and then the homogenate was centrifuged with a mini centrifuge. The supernatant was filtered with filter paper, and the absorbance of the filtrate was measured using a UV-VIS spectrophotometer (PD-303S, Apel, Saitama, Japan) at 500 nm and 900 nm. The anthocyanin concentration is shown as the cyanidin-3-glucoside equivalent. Ten stems (5 cm) per tunnel were also harvested from the distal end and the anthocyanin concentration was measured as described above. Cell Length and Number- Ten stems (5 cm) per tunnel were collected from the mid-portion of the plants, and the epidermal tissues were smeared with a transparent varnish for 5 min. Strips were then peeled off and the size and number of cells in 0.0186 mm2 for each replica were recorded under a photomicroscope system with a digital camera. Measurement of Invading Insects and Feeding Damage- On 21 November, blue and yellow sticky trap films (18 × 5 cm) (BSTs and YSTs, respectively) were suspended in middle positions of the tunnels to capture invading insects. The adhesives were replaced every week, and the kinds and number of captured insects were recorded daily until 19 December. During the 30 days of cultivation under different UV-blocking conditions, leaves of some seedlings showed feeding damage showed on leaves by herbivores. Therefore, the feeding damage of seedlings was evaluated in two different ways, as follows:
Rate of damaged seedlings (%) = Number of damaged seedlings per plot/Total number of seedlings per plot × 100
Rate of damaged leaf area (%) = Damaged area/Total leaf area × 100
Rate of damaged leaf area calculated using open access software (Image J, NIH, MD, USA).
Statistical Analysis- Data were subjected to analysis of variance, and significant differences of the means among treatments were analyzed using MSTAT-C software (East Lansing, MI, USA) and the differences among treatment means were evaluated by Duncan’s multiple range test (DMRT).