GERMINANT SOWING IN SOUTH AFRICA David B. South and Chris Young

Germinant sowing is operational for tree nurseries in South Africa. The technique reduces seed costs for eucalypts and pines. Filled cell percentage is usually near 98%. Seed efficiency at many North American container nurseries can be improved by adopting either germinant sowing or single sowing technology. Tree Planters' Notes xx(x):xx; 199x. INTRODUCTION Managers at most container nurseries attempt to produce one tree per container cell. This makes efficient use of containers, bench space, and potting media. Three different approaches are used to minimize the number of empty cells. A traditional approach in North America is to sow multiple seeds per cell and to thin cells that have more than one seedling (Schwartz 1993; Wenny 1993). A second method was developed in Sweden and involves removing dead and unfilled seeds prior to sowing (Simak 1984; Donald 1986). This method involves three steps (IncubationDrying-Separation) which is abbreviated IDS. Since the germination percentage can be increased to over 93%, the IDS treatment promotes the sowing of one non-germinated seed per cell. A third method involves germinating seeds prior to sowing and sowing only germinated seed. Although the concept of fluid drilling (sowing germinants in a viscous gel) was introduced into the Southern United States in early 1980's (Barnett 1983; 1985), mechanical sowing of germinants is not common in tree nurseries in North America. However in South Africa, mechanical sowing of germinants has been operational since 1986. The South African method suspends the germinated seed in water instead of mixing germinants in with a viscous gel. This paper reviews some of the advantages of germinant sowing and suggests that managers of container nurseries in North America consider adopting this technology. SEED EFFICIENCY Seed efficiency is defined as the percentage of plantable seedlings produced per pure live seed (South 1990). Achieving high seed efficiency is important when using valuable genetically improved seed or when seed cost is high. For example, in the Southern United States, seed efficiencies were often low (e.g. 33%) when nursery stock was not genetically improved. However, today, most pines are genetically improved and seed efficiencies in bareroot nurseries often exceed 80%. When seed are valued at 0.3 cent or more, there is a strong economic incentive for improving seed efficiency (South 1990). In North America, seed efficiency in container nurseries can be low if multiple seed are sown in each cell. For example, in British Columbia, seed efficiency from container nurseries is expected to range from 28 to 40% for woods collected seed (Table 1). At some operational container nurseries, seed efficiency can be less than 35% (Eremko et al. 1989). In some cases, seed efficiency can be higher at bareroot nurseries (Table 2). In general, container nurseries will have high seed efficiency when single-sowing (i.e. one seed per cell) is used. Many managers will single-sow when germination percentage is more than 90%. However, some recommend sowing two or more seeds when the germination percentage is less than 95% (Wenny 1993). Four or more seeds are sometimes recommended if germination is less than 70%. When seed costs and thinning costs are considered, the logic for multiple sowing is less attractive (Space and Balmer 1977). Table 3 compares the cost of production when using seed that costs 0.3 cent per pure live seed. In this example, seeding plus thinning costs were 34% greater for double sowing than for single sowing. Although seed and thinning costs make single-sowing more attractive, many nurseries in North America continue to multiple sow and thin. In North America, a typical nursery worker can thin about 40 trees per minute. IDS SYSTEM In Sweden, tree seed are routinely sorted to remove dead and unfilled seeds. Fully imbibed seed are incubated (kept in warm, moist conditions) for a few days and then are placed under controlled drying condtions. During the drying process, dead seed loose water at a greater rate than live seed. After drying, seed can be separated by density separation (dead seed float and live seed sink). The IDS procedure is used on Pinus sylvestris and Picea abies to produce nongerminated seed with a high germination rate (98%). Researchers in North America have not developed the technique to an operational level. However, this method has promise for several North American species (Donald 1986; Edwards 1989; Malek 1992; McRae et al. 1994) and could be used to eliminate the need for multiple sowing. GERMINANT SOWING In South Africa, filled cell percentages of 98 to 99% are consistently achievable with the use of germinant sowing. Originally developed in the United Kingdom, the concept was refined and simplified in South Africa. Much of the initial work was conducted in Natal by Bryan's Machinery in cooperation with Sappi Forests. The equipment is now available in North America from a distributor in Ontario. With the old system, seed were germinated in a tray, "pricked out" by hand, and transplanted into containers. With this method, about 150,000 Eucalyptus grandis seedlings could be produced from a kilogram of seed. With germinant sowing, the number of seedlings increased to 600,000/kg. With a value of $2,000/kg for genetically improved seed, the germinant sowing system reduced seed costs by $10/thousand seedlings. In addition to improving seed efficiency, labor costs were reduced at the Sappi Nursery. With the old system (manual transplanting into cells), the labor for one million plants was 175 person days. With germinant sowing, labor was reduced to 51 days. These are savings in the sowing operation. There are subsequent large savings in not having to thin the crop after emergence. At the Sappi Nursery, one machine can produce ten million plants/year. An added benefit is that seedling crops are very uniform because all the seed is sown at the same stage of germination. The key to success with this technique is sorting dead from live seed. The seed sample must be clean and well graded (of the same size and mass). This factor is imperative in order to successfully separate germinated from non germinated seed. For both pine and eucalypts, the germination fluid is water. If seeds are well graded, germinants will imbibe water and will change in size but not mass. Therefore, germinants have a lower specific gravity than nongerminants. The imbibed (swollen) seeds are separated by using a sugar solution (Taylor and Kenny 1985). Seed are placed in a small amount of water and a concentrated sugar solution is slowly added until imbibed seeds float to the top. These are then removed from the solution with a tea strainer. If seeds are germinated for too long, the radicals become elongated and tangling can result in multiple sowing. Ideally, the seed coat should be broken with the radical about to emerge. After separation, germinated seeds are placed in a water trough just below the vacuum head. Special needles on the vacuum head are dipped in the water and when removed, several germinants may adhere to each needle. A water rinse is used to remove excess germinants while one remains attached due to the vacuum. Needle size (hole size) varies from 0.1 mm to 0.9 mm. Correct needle selection is important (too small = misses; too large = doubles). Vacuum setting is also important (too low=misses; too high = doubles). Cycle time will vary with seed size. Large pine seeds require that the nozzles have a longer period in the fluid trough in order to become properly attached. For pine, almost no doubles occur but with the smaller eucalypts seeds, about 10% of the cavities will have doubles. Bryans' Miniseeder will sow 60 to 225 trays per hour (128 cavities/tray). The system can sow one row at a time or up to 1/4 tray at a time. At the Sappi Nursery, four machines are used for sowing. Germinant sowing is used for all eucalypts and for pine when germination percentage is less than 90%. Dry, single sowing of pine is still practiced when germination is greater than 89%. Two models of precision drilling machines are available in North America. Both are currently sold by INNO-TEC I.T.U. Inc. Thunder Bay, Ontario. The Miniseeder has a cost of $25,000 while a full size Precision Fluid Drilling System can cost about $48,000. The full size machine can sow full trays and production is about 66% faster than the Miniseeder. Currently, two full size machines are being used in Canada and one is in Mexico. If the purchase of a germinant sowing machine (@ $48,000) would eliminate double sowing, the potential savings in reduced seed costs and thinning costs could pay for the machine after only 10 million seedlings. For example, the estimated difference in cost between double sowing and germinant sowing could amount to $5,100 per million seedlings (Table 3). This savings results when each pure live seed is worth 0.3 cent and thinning costs amount to $4 per thousand thinned plants. In regions where seed is provided to nurseries free of charge (e.g. Canada), savings in thinning costs could pay for the machine after sowing 20 million seedlings. However, in situations where both seed and labor are free or inexpensive, it may be difficult to justify investing in germinant sowing. LITERATURE CITED Barnett, J.P. 1983. Optimizing nursery germination by fluid drilling and other techniques. In: Proceedings, Southern Nursery Conferences. 1982 July 12-15 Savannah, GA: USDA Forest Service, Southern Region, Technical Publication R8-TP4: 88-96. Barnett, J.P. 1985. Fluid drilling a technique for improving nursery seedling establishment. In: Proceedings, Third Biennial Southern Silvicultural Research Conference. 1984 November 7-8 Atlanta, GA: USDA Forest Service, Southern Forest Experiment Station, General Technical Report SO-54: 38-41. Donald, D.G.M. 1986. The separation of full dead seed from live seed in Pinus elliottii. p. 83-85. In D.South (ed.). Proc. of the International Symposium on Nursery Management for the Southern Pines. Auburn University, Alabama. Edwards, D.G.W. 1989. Prospects for IDS improvement of seed quality. FRDA Research Memo No 115. Forestry Canada, Pacific Forestry Centre. 2 pp. Eremko, R.D.; Edwards, D.G.W.; Wallinger, D. 1989. A guide to collecting cones of British Columbia conifers. FRDA report 055. Joint Pub., Forestry Canada and B.C. Ministry of Forests 114 p. Malek, L. 1992. Priming black spruce seeds accelerates container stocking in techniculture single-seed sowing system. Tree Planters' Notes: 43:11-13. McRae, J., Bergsten, U.; Lycksell, S. 1994. Use of the IDS treatment on southern pine seeds and its effect on seed cost and efficiency in the seedbed. In Proc. of the Southern Forest Nursery Association (IN PRESS). Schwartz, M. 1993. Germination math: calculation the number of seeds necessary per cavity for a given number of live seedlings. Tree Planters' Notes: 44(1):19-20. Simak, M. 1984. A method for the removal of filled dead seeds from a sample of Pinus contorta. Seed Science and Technology 12:767-775. South, D.B. 1990. Nursery seed efficiency can affect gains from tree improvement. In: Proceedings, Southern Forest Nursery Association; 1990 July 23-26; Biloxi, MS: 46-53. Space, J.C.; Balmer, W.E. 1977. Minimum cost calculation for container planting. USDA Forest Service, Southeastern Area State and Private Forestry. 18. p. Taylor, A.G.; Kenny, T.J. 1985. Improvement of germinated seed quality by density separation. Journal of the American Society of Horticultural Science 110:347-349.. Wenny, D.L. 1993. Calculating filled and empty cells based on number of seeds sown per cell: a microcomputer application. Tree Planters' Notes 44(2):49-52. Table 1. Recommended seeding rates, oversowing factor (i.e. extra cells sown to ensure meeting production targets) and expected seed efficiencyfrom container nurseries using 1994 B.C. Ministry of Forests sowing rules. Woods collected seed Seed orchard seed Germination seed oversow seed seed oversow seed percentage /cell factor efficiency /cell factor efficiency ------------%-----------------------%------------100 2 25 40 1 40 71 95 3 30 40 1 45 72 85 3 35 30 2 30 45 75 3 40 32 3 40 32 65 4 50 26 4 50 26 55 4 60 28 4 60 28 Table 2. Seed efficiency from container and bareroot nurseries in British Columbia during the 1980's (data from Eremko et al. 1989) and seed costs from a seed dealer. Species # plantable seedlings/pure live seed Pure live seed cost Container Bareroot ------------%------------------cent ------Coastal Douglas-fir 23 32 0.21 Western Hemlock 27 25 0.09 Western Larch 28 28 0.22 Lodgepole pine 30 41 0.08 Ponderosa pine 22 38 0.80 Western white pine 53 66 0.54 Sitka spruce 35 43 0.20 Grand fir 22 24 0.42 Pacific silver fir 59 23 0.59 Table 3. Estimated sowing and thinning costs associated when producing ten million seedlings. Two seed/cell One seed/cell One germinant/cell Cells needed 10,666,666 13,333,333 10,204,082 Germination % 75 75 75 Seeds required 21,333,332 13,333,333 13,333,333 Blanks expected 666,666 3,333,333 204,082 Excess trees 6,000,000 0 0 Seed cost $ 64,000 40,000 40,000 Sowing cost $ 5,333 6,666 5,102 Thinning cost $ 24,000 0 0 Cost of carrying empty cells $ 5,333 26,666 1,633 Total costs $ 98,666 73,332 46,735 This paper can be cited as follows: South, D.B. and C. Young. 1994. Germinant Sowing in South Africa. Combined Proceedings International Plant Propagators' Society 44:266-270