Hybrid dahlia (Dahlia hybrida L.) seed production : a thesis presented in partial fulfilment of the requirements for the degree of Master of Applied Science in Plant Science (Seed Technology) at Massey University, New Zealand

Abstract

Three experiments related to dahlia seed production were conducted at the Seed Technology Centre, Massey University, New Zealand in the 1995/1996 season, with the objectives being to: evaluate herbicide toxicity to dahlia; determine the effect of establishment method and plant density on dahlia seed yield and yield components; and determine the effect of sowing date on dahlia seed yield. Thirteen herbicides applied pre-emergence or post-emergence were evaluated in the herbicide experiment. EPTC (4.32 kg a.i/ha), oxyfluorfen (0.72 kg a.i/ha), oryzalin (4.5 kg a.i/ha), oxadiazon (1.52 kg a.i/ha), simazine (1.0 kg a.i/ha) and terbacil (0.96 kg a.i/ha) all caused some injury to direct sown and transplanted dahlias, and can not therefore be used safely on the dahlia crop. Of the pre-emergence herbicides, alachlor (1.92 kg a.i/ha), chlorpropham (3.2 kg a.i/ha), chlorthal-dimethyl (7.5 kg a.i/ha), pendimethalin (1.32 kg a.i/ha) and trifluralin (1.2 kg a.i/ha) did not injure either direct- sown or transplanted plants. Methabenzthiazuron (1.05 kg a.i/ha) did inhibit the early growth of direct sown dahlia, but plants recovered very quickly. All five herbicides could be used as pre-emergence herbicides for dahlia, but on a cost and weed control spectrum basis, alachlor or trifluralin are recommended. Terbacil (0.96 kg a.i/ha), while not affecting transplanted seedlings did damage direct sown dahlia, and should only be used as a pre-emergence herbicide in transplanted dahlia. From the post-emergence herbicides, chlorpropham (3.2 kg a.i/ha) did not injure either direct-sown or transplanted dahlia. Chlorthal-dimethyl (7.5 kg a.i/ha), haloxyfop (0.3 kg a.i/ha), and methabenzthiazuron (1.05 kg a.i/ha) caused some plant injury to early growth of direct-sown seedlings, but injured plants recovered quickly. Therefore, these herbicides can be recommended for both direct-sown and transplanted dahlia when applied post-emergence. Of these, the cheapest is methabenzthiazuron, and on this basis it is recommended for use in dahlia post-emergence. Method of establishment (transplanted seedlings or tubers) did not affect seed yield. Of five plant densities (0.8, 0.6, 0.4, 0.3 and 0.2 m square spacings), seed yield per square meter from a harvest when 80% of the seedheads had turned brown was greatest from the 0.4 m square spacing (12.3 g/m²). However, when seedheads were harvested as they ripened (i.e. over several weeks), the highest yield was at the 0.3 m square spacing (15.97 g/m²). Individual plant yield was highest (4.49 g/plant) at the lowest density (0.8 m square spacing), and the lowest (0.105 g/plant) at the highest density (0.2 m square spacing). Seedheads per plant contributed most to the differences, as more branches per plant at lower densities produced more seedheads per plant. Seed weight was slightly bigger at lower densities, while seed number per seedhead was greater at higher densities than at lower densities. The feasibility of direct seed sowing for dahlia seed production was confirmed in Palmerston North, New Zealand. Sowing dates from 7 November to 5 December did not produce any difference in seed yield and yield components. However, a 19 December sowing produced a significantly lower seed yield per plant, seedheads per plant and thousand seed weight. Later sowing delayed flowering time, shortened flowering duration and made seedheads ripen later. All sowing were harvested after frosts, and the seedheads in the latest sowing were very immature when harvested. In this environment, seed should not be sown later than 7 November.