The Horticulture Journal 85 (3): 264 271. 2016. doi: 10.2503/hortj.MI-072 JSHS The Japanese Society for Horticultural Science http://www.jshs.jp/ Effects of Interrupted Lighting on the Spray Formation of Summer-toautumn-flowering Small-flowered Spray-type Chrysanthemum Cultivars Haruka and Subaru Yoshio Mori 1,3 *, Katsuhiko Sumitomo 2, Tamotsu Hisamatsu 2 and Tanjuro Goto 3 1 Okayama Prefectural Technology Center for Agriculture, Forestry and Fisheries, Akaiwa 709-0801, Japan 2 NARO Institute of Floricultural Science, Tsukuba 305-8519, Japan 3 Graduate School of Environmental and Life Science, Okayama University, Okayama 700-8530, Japan Summer-to-autumn-flowering small-flowered spray-type (SAFS) chrysanthemums (Chrysanthemum morifolium Ramat.) are commercially important in Japan, and there is an occasional increased demand for these flowers. However, under current cultivation methods, the quality of spray formations is compromised in order to maintain a sufficient production rate at times of peak demand. This study investigated the effect of interrupted lighting (IL), the intercalation of a period of long days (by night interruption) into the natural day (NDL) period of growing, in order to regulate spray formations in SAFS chrysanthemum cultivars Haruka and Subaru, in which lighting can markedly delay flowering. The effects of the NDL period before IL and those of the IL period were investigated. First, we investigated the effect of 2 12 days of NDL followed by 12 days of IL. Plants subjected to 2 6 days of NDL had longer flower clusters and a greater number of flower buds on the upper lateral flower stems and developed a broom-like spray. Next, we investigated the effect of 4 20 days of IL after 4 days of NDL. Plants subjected to 8 20 days of IL had longer flower clusters, more flower buds on the upper lateral flower stems, and broom-like sprays. These changes were more pronounced when the IL periods were prolonged. Our results showed that spray formation in SAFS chrysanthemum cultivars Haruka and Subaru can be regulated by controlling the timing and period of IL. Key Words: flower cluster, light culture, long day, number of flower buds, short day. Introduction In Japan, small-flowered spray-type chrysanthemums are used in many religious festivals, and hence, their demand increases at particular times of the year. About 40% of these flowers are marketed from July to September. During periods of high demand, summer-toautumn-flowering small-flowered spray-type (SAFS) chrysanthemums are used; these are usually grown under natural day (NDL). Since the flowering time of SAFS chrysanthemums is strongly influenced Received; April 22, 2015. Accepted; October 26, 2015. First Published Online in J-STAGE on December 16, 2015. Part of this study was presented at the 2014 and 2015 Spring Meeting of the Japanese Society for Horticultural Science. This work was supported by a grant-in-aid from the Ministry of Agriculture, Forestry and Fisheries of Japan (A scheme to revitalize agriculture and fisheries in disaster area through deploying highly advanced technology). * Corresponding author (E-mail: yoshio_mori@pref.okayama.lg.jp). by climatic conditions, maintaining their production to meet peak demand is difficult. Thus, we investigated a light culture using 9 small-flowered spray-type chrysanthemum cultivars that flower from June to July under NDL; we found that flowering could be controlled in 4 cultivars ( Haruka, Subaru, Seikomaki, and Seichigusa ) by modifying lighting regimens, allowing growers to meet the peak demand supply from July to September (Mori et al., 2014). However, in these 4 cultivars, the terminal capitulum tended to be set above the lateral capitula, and spray formation was different from that of the traditional SAFS chrysanthemum cultivars (the terminal capitulum is set under the lateral capitula, making the spray formation appear broom-like). Moreover, in Haruka, only few capitula were present on the upper lateral flower stem under both light and no-light culture conditions; Subaru and Seikomaki occasionally have only a few capitula on the upper lateral flower stem under light culture. Therefore, some customers want growers to 2016 The Japanese Society for Horticultural Science (JSHS), All rights reserved.
Hort. J. 85 (3): 264 271. 2016. 265 regulate spray formation produced using light culture. However, the demand for different spray formations differs across customers; at present, growers change the cultivar type to meet the demands of customers. If growers can regulate spray formation, the demands for various spray formations can be met by using the same cultivar. Osuga et al. (1978) showed that standard-type chrysanthemums developed small upper leaves under light culture, and that these could be restored to their normal by using interrupted lighting (IL). IL is the intercalation of a period of long days (LD) by night interruption (NI, a short exposure to light during the middle of the night) into the short day (SD) phase of growing. This method has since been used widely in Japan to improve the quality of standard-type chrysanthemum cut flowers. In spray-type chrysanthemums, IL has been shown to regulate the cut flower characteristics (Naka and Maeda, 2006; Takahashi and Nakano, 1999; Tjia and Glasser, 1976; Yamagata et al., 2008). In particular, IL has been investigated in winter- and autumn-flowering small-flowered spray-type chrysanthemums (Sato, 1981; Yamagata et al., 2008); however, few studies have determined its effectiveness in summer-toautumn-flowering types. In the autumn-flowering small-flowered spray-type chrysanthemum cultivar Tsubasa cultivated using IL, the spray formation of plants became broom-like because the of lateral flower stems increased and axillary buds developed on lateral flower stems (Yamagata et al., 2008). This indicates that IL could also regulate spray formation in SAFS chrysanthemums. However, a previous study showed that the effect of IL on spray formation varied markedly according to its timing (Yamagata et al., 2008). This study aimed to investigate the effects of the timing and period of IL on spray formation in two SAFS cultivars Haruka and Subaru. Materials and Methods Plant material and growth conditions Haruka and Subaru were grown in an open field at the Okayama Prefectural Technology Center for Agriculture, Forestry and Fisheries (Akaiwa, Okayama) in 2013 and 2014. In both years, 6 rooted cuttings were transplanted on May 14 into a plastic container (width, 23 cm, 64 cm height, 18.5 cm) filled with a mixture of silty loam, perlite, and peat moss (1:1:1 volume ratio). To all the plants, 900 mg N, 705 mg P 2 O 5, and 835 mg K 2 O of slow-release fertilizer (Ecolong 413-100; Jcam Agri. Co., Ltd., Tokyo, Japan) was applied. After 10 days, plants were pinched at the uppermost leaf, and each plant was allowed to grow three shoots. Plants were grown under NI conditions, i.e., NDL plus a daily 4 h (22:00 2:00, 2013) or 6 h (22:00 4:00, 2014) of NI with incandescent bulbs (K- RD110V75W; Panasonic Corp., Osaka, Japan), to maintain the plants in a vegetative state until the treatments were started. Incandescent bulbs were set at every 4 m 2, at a height of 1.5 m from the soil surface of the container, since horizontal illuminance at the soil surface became higher than 50 lx. Experiment 1: Effect of timing of IL on flowering and cut flower characteristics All plants were transferred to NDL treatment on June 20, 2013. SAFS cultivars have longer critical day for flowering than autumn-flowering cultivars (Oda et al., 2010; Sumitomo et al., 2013, 2014); therefore, NDL induced flowering in the 2 cultivars. Plants were transferred to NI (22:00 2:00) at intervals of 2, 4, 6, 8, or 12 days, and then subjected to NI for 12 days after NDL treatment (IL treatments, Fig. 1). Subsequently, the plants were transferred back to NDL until flowering. Control plants were not subjected to IL. Six plants were used for each treatment. At the beginning of each IL treatment, 1 shoot apex of the second lateral shoot was collected from each plant and dissected under a stereoscopic microscope to observe the stage of flower development. For the remaining plants, the date of flowering was recorded, and this was defined as the appearance of a visible disk floret on the terminal capitulum. At flowering, each shoot was collected, and the following characteristics of the cut flowers were recorded (Fig. 2): cut flower, stem, flower cluster, neck of the terminal capitulum, number of primary lateral flower stems, total number of capitula per cut flower, flower cluster width, number of capitula on each lateral flower stem (1st to 10th lateral flower stems down from the terminal capitulum), and of each lateral flower stem (3rd, 6th, and 9th lateral flower stems). Chrysanthemums have a condensed indeterminate inflorescence (capitulum) structure composed of many individual florets, which bears a single flower-like architecture. Thus, the capitulum of chrysanthemums apparently mimics a large single flower. Therefore, in this report, we refer to the mass of capitula as a flower cluster (Fig. 2). Experiment 2: Effect of IL periods after NDL on flowering and cut flower characteristics From June 20, 2014, plants were grown for 4 days under NDL. Plants were then transferred to NI (22:00 2:00) on June 24, and grown for 4, 8, 12, or 20 days (IL treatments, Fig. 1). Control plants were not subjected to IL. After IL, plants were returned to NDL. Twelve plants were used in each treatment. The stage of flower development, flowering date, and cut flower characteristics (including leaf number besides those described above) were measured.
266 Y. Mori, K. Sumitomo, T. Hisamatsu and T. Goto Fig. 1. Interrupted lighting (IL) treatments in Experiments 1 and 2. White and grey horizontal bars represent days of natural day (NDL) with or without night interruption, respectively. Fig. 2. Measurements of cut flower characteristics. Results Experiment 1: Effect of the timing of IL on flowering and cut flower characteristics Average temperature from the date of first transfer to NDL (June 20) to the final end date of IL (July 14) was 25.2 C at the Okayama Prefectural Agriculture, Forestry and Fisheries Research Center. There were no morphological changes (such as flower initiation) at the shoot apex until 4 days after transfer to NDL (Fig. 3). At 6, 8, and 12 days after transfer to NDL, the shoot apex reached the dome-formation stage, and the first and latter stages of involucre primordia differentiation, respectively. Flower development at 12 days after transfer to NDL was slightly earlier in Haruka than in Subaru, but not significantly (P = 0.061). IL after 2 12 days of NDL significantly delayed flowering in both the cultivars compared with that in the controls (Table 1). Flowering occurred latest when subjected to IL after 2 and 4 days of NDL in the Haruka and Subaru plants, respectively. With a decrease in NDL period, cut flower and stem increased. Compared with those of controls, cut flower and stem were significantly greater when subjected to IL after 2 6 and 2 days of NDL, respectively, in Haruka, and after 2 8 and 2 4 days of NDL, respectively, in Subaru. Flower clusters were
Hort. J. 85 (3): 264 271. 2016. 267 Fig. 3. The effect of natural day (NDL) period on flower development of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Flower development stages are (0) vegetative growth, (1) dome formation, (2) first stage of involucre primordia differentiation, and (3) latter stage of involucre primordia differentiation. One dot indicates one shoot. There was no significant difference between the 2 Table 1. cultivars Effects at of P natural < 0.05 by day using Fisher s (NDL) exact period test. before 12 days of interrupted lighting (IL) on flowering and cut flower characteristics of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Table 1. Effects of natural day (NDL) period before 12 days of interrupted lighting (IL) on flowering and cut flower characteristics of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Cultivar Haruka Subaru NDL period (days) Date of flowering (m/d) Cut flower Stem Flower cluster Neck of terminal capitulum Number of primary lateral flower stems Total number of capitula per cut flower Flower cluster width 2 8/28 a z 96 a 89 a 7.3 b 2.9 abc 14.7 a 29 ab 8.8 ab 4 8/25 b 96 a 84 ab 11.5 a 4.4 a 9.5 a 36 a 11.8 a 6 8/21 c 89 ab 83 ab 6.3 c 4.1 abc 9.4 a 11 b 8.0 b 8 8/21 c 86 bc 80 ab 5.6 c 3.6 abc 9.8 a 11 b 7.2 b 12 8/20 c 85 bc 80 b 5.3 c 2.7 bc 14.4 a 18 ab 9.0 ab No IL 8/14 d 80 c 76 b 4.6 c 2.4 c 15.5 a 19 ab 8.6 ab 2 8/22 b 97 a 93 a 3.6 c 1.5 b 28.1 a 76 ab 10.8 a 4 8/25 a 94 a 82 b 11.8 ab 2.5 b 15.0 b 85 ab 12.7 a 6 8/20 bc 93 ab 79 bc 13.3 a 4.3 a 19.2 ab 116 a 15.0 a 8 8/20 c 90 ab 81 bc 9.7 ab 4.1 a 17.1 ab 66 ab 14.2 a 12 8/17 d 86 bc 78 bc 8.2 b 4.8 a 16.1 ab 38 b 13.3 a No IL 8/9 e 80 c 76 c 4.3 c 2.4 b 21.3 ab 57 ab 12.8 a z Different letters indicate significant difference at P < 0.05 by using the Tukey-Kramer multiple range test. significantly longer when subjected to IL after 2 4 days of NDL in Haruka and 4 12 days of NDL in Subaru, with the longest clusters formed in plants subjected to IL after 4 and 6 days of NDL treatment, respectively. Thus, earlier application of IL resulted in a strong plant growth stimulus. However, the effect varied among the 2 cultivars. The neck of the terminal capitulum was significantly greater with IL after 4 days of NDL in Haruka and with IL after 6 12 days of NDL in Subaru. In both cultivars, the number of primary lateral flower stems, total number of capitula per cut flower, and flower cluster width did not differ from those of the controls. The effect of IL on the number of capitula on each lateral flower stem varied among the cultivars. The number of capitula on the 1st to 7th lateral flower stems increased with IL after 2 4 days of NDL in Haruka (Fig. 4). In Subaru, IL after 2 days of NDL had no effect on the number of capitula, whereas IL after 4 12 days of NDL increased the number of capitula. The lateral flower stems of which the capitula number increased with IL appeared at a lower position with the delay in the timing of IL. IL after 4, 6, 8, and 12 days of Fig. 4. The effect of natural day period before 12 days of interrupted lighting (IL) on the number of capitula on each lateral flower stem of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Vertical bars represent SE (n = 10 12).
268 Y. Mori, K. Sumitomo, T. Hisamatsu and T. Goto NDL increased the number of capitula on the 1st to 10th stems, 3rd to 10th stems, 6th to 10th, and 10th stems, respectively. The number of capitula on the upper 5 flower stems did not increase in either cultivar with IL after 8 12 days of NDL. Late IL had no effect on the upper lateral flower stems of the plants. This trend was prominent in Haruka. The number of capitula on the 1st to 7th lateral flower stems increased with IL after 2 4 days of NDL, but not after 6 12 days of NDL. Thus, the IL effect on the number of capitula on each lateral flower stem varied markedly depending on the IL timing and cultivar type. Haruka showed longer lateral stems than those of the controls with IL after 4 days of NDL for the 3rd, 6th, and 9th stems, after 6 days for the 3rd and 6th stems, and after 8 days for the 6th stem (Fig. 5). In Subaru, the 3rd and 6th lateral flower stems were longer with IL after 4 and 12 days of NDL, and the 3rd, 6th, and 9th stems were longer after 6 and 8 days of NDL. The spray formation in the 2- and 4-day NDL treatments in Haruka and the 4- to 6-day treatments in Subaru differed substantially from that in the control, i.e., those in the treated plants became broom-like with many flower buds over the terminal capitulum (Fig. 6). Experiment 2: Effect of IL period after NDL on flowering and cut flower characteristics Average temperature from the date of first transfer to NDL (June 20) to the final end date of IL (July 14) was 24.3 C at the Okayama Prefectural Agriculture, Forestry and Fisheries Research Center. There were no morphological changes (such as flower initiation) at the shoot apex at 4 days after transfer to NDL. Eight to 20 Fig. 5. The effect of natural day period before 12 days of interrupted lighting (IL) on the of each lateral flower stem of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Vertical bars represent SE (n = 10 12). days of IL showed quantitative effects on cut flower characteristics, although 4 days of IL had little effect (Table 2). IL increased the days to flowering, cut flower, stem, flower cluster, neck of terminal capitulum, total number of capitula per cut flower, and flower cluster width. Further, the number of capitula on each lateral flower stem (Fig. 7) and the of each lateral flower stem (Fig. 8) increased. Plants showed a simultaneous increase in these parameters with IL period. The number of leaves and primary lateral flower stems showed no consistent trend in their response to IL. Discussion In this study, we showed that IL could substantially regulate the appearance of the cultivated SAFS chrysanthemum cultivars Haruka and Subaru. The findings of our first experiment suggest that the NDL period before IL is important for flower development and spray formation in both cultivars. The spray formation is composed of the of lateral flower stems and number of capitula on the upper lateral flower stem. IL affected both these parameters, and thus changed the spray formation. With IL after 2 or 4 days of NDL, Haruka plants had longer flower clusters and more flower buds on the upper 5 lateral flower stems, leading to a broom-like spray. In Subaru plants, these changes occurred after longer NDL periods of 4 6 days. This suggests that spray formation in SAFS chrysanthemums can be regulated by IL; however, the NDL period before IL markedly affected spray formation, as has been previously shown in autumn-flowering small-flowered spray-type chrysanthemums (Yamagata et al., 2008), and the appropriate NDL period to obtain the broomlike spray formation is different across cultivars. In pompom chrysanthemums, Post (1943) showed that long days (LD) for a period of 5 20 days after short days (SD) for 7 35 days increased the peduncle of the upper lateral flower stems in most combinations of day, and that 7 days of SD had the greatest effect. Consistent with our study results, their findings showed that SD period before LD could substantially influence flower shape. The change in spray formation is likely to occur due to the effects of IL on the timing of different developmental stages. First, the timing of flower bud initiation was affected, with IL inhibiting the development of axillary flower buds near the apical flower buds. Subsequently, secondary lateral flower stems developed with the restoration of vegetative growth of the primary lateral flower stems. Next, with an increase in the of lateral flower stems and flower clusters, more flower buds developed on the secondary lateral flower stems. Thus, spray formation changed with the increase in the number of flower buds on the upper lateral flower stems. This effect was less marked when IL was started after very little or excessive NDL. Presumably, in the
269 Hort. J. 85 (3): 264 271. 2016. Fig. 6. The effect of natural day (NDL) period before 12 days of interrupted lighting (IL) on spray formation of summer-to-autumnflowering small-flowered spray-type chrysanthemums. Table 2. Effects of interrupted lighting (IL) periods after 4 days of natural day (NDL) on flowering and cut flower characteristics of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Table 2. z Effects of interrupted lighting (IL) periods after 4 days of natural day (NDL) on flowering and cut flower characteristics of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Cultivar IL period after NDL (days) Date of flowering (m/d) Haruka 4 8 12 20 No IL 8/17 dz 8/22 c 8/25 b 8/31 a 8/14 e Subaru 4 8 12 20 No IL 8/12 d 8/18 c 8/22 b 9/1 a 8/8 e Flower cluster Neck of terminal capitulum Leaf number 77 c 82 b 86 a 87 a 77 c 4.7 c 7.6 b 8.0 b 11.4 a 4.2 c 2.6 c 3.5 b 3.5 b 4.4 a 1.8 d 32 a 32 a 33 a 32 a 33 a 12.0 a 9.8 a 10.3 a 11.7 a 13.7 a 21 c 29 bc 33 ab 40 a 17 c 8.2 c 9.2 bc 9.7 ab 10.7 a 6.9 d 77 b 82 a 83 a 84 a 73 b 4.7 d 8.3 c 10.8 b 16.9 a 4.0 d 2.2 c 2.5 bc 2.8 b 3.6 a 1.9 c 40 ab 41 a 40 ab 40 ab 37 b 19.1 a 16.3 a 16.3 a 16.1 a 18.0 a 51 bc 67 abc 80 ab 94 a 39 c 9.6 b 10.7 ab 10.4 ab 12.1 a 9.5 b Cut flower Stem 82 c 90 b 94 ab 99 a 81 c 81 c 90 b 93 b 100 a 77 c Number Total number Flower of primary of capitula cluster width lateral flower per cut stems flower Different letters indicate significant difference at P < 0.05 by using the Tukey-Kramer multiple range test. former case, flower bud initiation had not yet started, whereas in the latter, flower bud development had markedly progressed for axillary bud development to be sufficiently suppressed. Thus, the timing of the beginning of IL may be very important. In our experiment, the appropriate NDL period to ob-
270 Y. Mori, K. Sumitomo, T. Hisamatsu and T. Goto Fig. 7. The effects of interrupted lighting (IL) periods after 4 days of natural day on the number of capitula on each lateral flower stem of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Vertical bars represent SE (n = 23 24). Fig. 8. The effects of interrupted lighting (IL) periods after 4 days of natural day on the of each lateral flower stem of summer-to-autumn-flowering small-flowered spray-type chrysanthemums. Vertical bars represent SE (n = 23 24). tain the broom-like spray formation was different between the 2 cultivars, i.e., 2 4 days in Haruka and 4 6 days in Subaru (Fig. 6). This difference may be attributed to the timing of flower initiation: flower development at 12 days after transfer to NDL in Haruka was slightly earlier than that in Subaru ; however, the difference was not significant (P = 0.061; Fig. 3). The most effective time to start IL was between the vegetative stage and dome formation on terminal buds. Thus, during commercial production, flower development needs to be uniformly controlled across many plants to maximize the profit obtained from regulating spray formation by using IL. However, under NI conditions, undesirable premature flower initiation is often observed in SAFS chrysanthemums (Mori et al., 2014). Thus, selected SAFS cultivars, which stay vegetative under non-inductive photoperiodic conditions, and show synchronous flower initiation in response to transfer from non-inductive to inductive photoperiodic conditions such as Haruka and Subaru, should be used for stable regulation of spray formation using IL. Mori et al. (2014) suggested that Seikomaki and Seichigusa could be used for light culture as well as Haruka and Subaru. Therefore, the regulation of spray formation in Seikomaki and Seichigusa may also be possible. However, the rate of flower development is known to be affected by environmental factors even in these SAFS chrysanthemums (Koyama and Wada, 2004; Mori et al., 2014). Therefore, monitoring the status of flower development is required to determine the effective IL timing, although microscopic observation of flower bud development may not be sufficient to decide when to start IL, because such observation may not indicate the time when flower bud initiation began. Therefore, an index needs to be developed that can be used to determine the exact time to start IL. Our second experiment investigated the effects of IL period. Both the cultivars developed longer flower clusters and more flower buds on the upper 5 lateral flower stems after 8 20 days of IL, transforming the spray to a broom-like one, with prominent changes noted after longer IL periods. Thus, quantitatively controlling spray formation in SAFS chrysanthemum cultivars Haruka and Subaru by using IL and producing various spray formations by changing the IL period may be possible. Indeed, previous studies have shown that 8 14 days of IL after 20 days of SD could increase cut flower and the number of flower buds, with longer IL periods causing greater changes in winterflowering small-flowered spray-type chrysanthemums (Sato, 1981). This is consistent with our findings suggesting that IL period affects the number of flower buds. Our study results suggest that regulating spray formation in SAFS chrysanthemum cultivars Haruka and Subaru is possible by using IL. We found that 2 6 days of NDL followed by 8 20 days of IL was suitable; however, these periods varied according to environmental factors and responses of cultivars. Therefore, determining the suitable NDL and IL periods for each cultivar and each cropping type are necessary. Furthermore, we found that quantitatively controlling spray formation in Haruka and Subaru by changing the IL period may be possible. Because different customers prefer different types of spray formations, determining
Hort. J. 85 (3): 264 271. 2016. 271 suitable IL periods for the different types of spray formations to meet the demands of customers is important. Moreover, because IL delayed flowering, NI should be stopped earlier in consideration of the delayed flowering period. Further studies are warranted to investigate a range of cultivation methods combined with IL. These techniques could facilitate the production of various spray formations by using the same cultivar, thereby allowing growers to meet various demands by using only a few cultivars. This would stabilize the management of cut flower production. Introduction of these techniques mean that growing areas will need an electricity supply and lighting equipment; nonetheless, this will not be a challenge for growers already using light culture. Literature Cited Koyama, Y. and O. Wada. 2004. Flowering control of smallflowered chrysanthemums (Dendranthema grandiflorum (Ramat.) Kitamura) with July flowering type by night-break lighting treatment systematic production for high demand time. Hort. Res. (Japan) 3: 63 66 (In Japanese with English abstract). Mori, Y., T. Nakajima, T. Fujimoto, T. Tsunemi, K. Sumitomo, T. Hisamatsu and T. Goto. 2014. Selection of small-flowered spray-type chrysanthemum varieties suitable to meet peak demand from July to September using night-interruption treatment. Hort. Res. (Japan) 13: 349 356 (In Japanese with English abstract). Naka, T. and S. Maeda. 2006. Effects of interrupted lighting on the flowering form of light culture niringiku (double flowered chrysanthemum). Bull. Nara Agr. Exp. Sta. 37: 31 34 (In Japanese). Oda, A., K. Sumitomo, T. Tsunemi, M. Douzono, T. Motozu and T. Hisamatsu. 2010. Variations of photoperiodic response of flower buds initiation and development in July- and Augustflowering small-flowered spray chrysanthemum cultivars. Hort. Res. (Japan) 9: 93 98 (In Japanese with English abstract). Osuga, M., Y. Sakurai and M. Murakami. 1978. Studies on re lighting of flowering chrysanthemums (Chrysanthemum morifolium Ramat.) in light culture. Res. Bull. Aichi Agric. Res. Ctr. B10: 21 29 (In Japanese with English abstract). Post, K. 1943. The effect of an interval of long days in the short day treatment on the flowering of chrysanthemums. Proc. Amer. Soc. Hort. Sci. 43: 311 315. Sato, Y. 1981. The effect of the Saidenshou [interrupted lighting] treatment in the natural short day on the retardation of flowering and the quality of cut flower of small flowered winter flowering chrysanthemum in light culture. Bull. Kagawa Agric. Exp. Stn. 33: 29 35 (In Japanese with English abstract). Sumitomo, K., Y. Higuchi, A. Yamagata and T. Hisamatsu. 2013. Memory of prolonged winter cold inhibits flowering and increases long-day leaf number in the chrysanthemum cultivar Nagano Queen. J. Hort. Sci. Biotech. 88: 361 367. Sumitomo, K., A. Yamagata, A. Oda and T. Hisamatsu. 2014. Identification of high long-day leaf number cultivars and prevention of premature budding by cold pre-treatment for fine control of flowering in summer-to-autumn-flowering chrysanthemum (Chrysanthemum morifolium Ramat.). J. Hort. Sci. Biotech. 89: 647 654. Takahashi, T. and T. Nakano. 1999. Improvement of quality by interrupted lighting on autumn flowering spray type chrysanthemums. Tohoku Agric. Res. 52: 239 240 (In Japanese). Tjia, B. and M. Glasser. 1976. Effect of interrupted lighting on peduncle and spray formation of Chrysanthemum morifolium Ramat. Proc. Fla. State Hort. Soc. 89: 313 315. Yamagata, A., K. Sumitomo and T. Hisamatsu. 2008. Effects of interrupted lighting on spray formation and organ formation in terminal capitulum of single flowered spray type Chrysanthemum morifolium. Hort. Res. (Japan) 7 (Suppl. 2): 603 (In Japanese).