To produce conditions that will encourage establishment of a wide range of seedlings through natural seeding, and avoid revegetation failing, an understanding of certain abiotic and biotic factors is required. The main factors that affect establishment through natural seeding are: species present, soil type, moisture, competition by grasses and herbs, available seed trees, and weather conditions (heat, dryness etc). It is important to know the timing and periodicity of seed production and dispersal. Basic knowledge about the period for the high rates of seed dispersal is necessary when practicing natural regeneration. In order to encourage natural seeding, ground preparation must be undertaken prior to seed dispersal. Specific characteristics of a species, such as number of seeds per tree, seed weight and frost resistance, greatly influence the establishment of seedlings. Seeds from some species are wind dispersed (e. g. birch and sallow (Salix caprea L.)) and others water dispersed (e. g. alder); a combination of methods may be used. Studies of wind-mediated seed dispersal for different species indicate the following order of decreasing dispersal:

birch>elm=maple>alder>hornbeam>beech>oak (Augspurger and Franson, 1987; Okubo and Levin, 1989; Willson, 1990; Karlsson, 2001). Table 1 contains data on birch and alder seed dispersal.

Distance from forest stand, m

<50———— 50-100——- 100-150——— >150 Country Reference

cut area have been found to spread at a rate of about 100 seeds m-2 up to 200 m from the tree (Fries, 1984). Most of these birch seeds were dispersed during September, although the process continued until December. In a study of sweet birch (Betula lenta L.), Matlack (1989) reported seed were dispersed 3.3 times further than the distance measured by Fries (1984). In a study of silver birch in Estonia, 21 % of the seeds were dispersed in July, 77 % in August and 2 % in September (Kohh, 1936). Heikinheimo (1932, 1937), who reported the same dispersal periods, commented that the weather during summer and autumn is the main factor affecting the period of seed dispersal. Graber and Leak (1992) presented a study on seed fall for broadleaved species in New Hampshire. The mean seed fall (million ha-1) in a study lasting 11 years was: 6.58 for yellow birch (Betula alleghaniensis Britton); 6.38 for paper birch (Betula papyrifera Marsh.); 4.11 for sugar maple (Acer saccharum Marsh.); and 0.17 for American beech (Fagus grandifolia Ehrh.). The seed viability was 30-50 %, depending on species.

Besides wind dispersal, there are some reports of secondary dispersal of seeds (Hesselman, 1934; Matlack, 1989; Greene and Johansson, 1997). The most common is by movement on snow, but for this to occur, seed fall must happen during winter months when snow is on the ground. The seeds can be damaged by friction on frozen snow, thus reducing viability. The level of seed production by alder depends on the number of hours of sunshine in the period April-September in the year before fruiting, the number of hours of sunshine in the seeding year and the level of seed production in the preceding year (MacVean, 1955). According to MacVean (1955), common alder (Alnus glutinosa (L.) Gaertner) seeds are generally dispersed within a radius of 30-60 m of the mother tree. Karlsson (2001) found that 50 % of the total number of alder seeds produced fell within 5 m and 90 % within 20 m of the stand. In a study by Johansson and Lundh (2006), 50 % of the common alder seeds were found to have fallen before December and 75 % before February. Alder seeds can also be transported by water in spring at the time of snow melt.

Seeds from European aspen (Populus tremula L.) are extremely small (low weight) with a limited growing capacity (Blumenthal, 1942, Latva-Karjanmaa et al., 2006). A large aspen growing close to Tartu city, Estonia, produced 49 kg or 54 million seeds (Reim, 1930). Only a small proportion of the aspen seeds produced will grow; success depends on site conditions, seed size and the level of competition. Aspen seeds can grow on poor sandy sites, burned areas and small patches without vegetation (Blumenthal, 1942). Seeds of sallow are also small and have a plume to aid dispersal (Grime et al., 1988). Seeds of both species can be dispersed over long distances.

The most favorable soil types for rapid establishment of seedlings are fine sand, silt and light clay, sandy-silty till and light clay till. Even peat soils can provide an ideal site, providing there is sufficient water. A mixture of mineral soil and humus is common on farmland, where the area has been cultivated for many years.

Birch seeds establish well on undisturbed sites with a high level of moisture (Mork, 1948; Fries, 1982). During the first part of the growing season in Nordic countries (April-May) soil moisture tends to be low. The lack of rain combined with the sunshine during this period results in a dry soil. Therefore any soil treatment (plowing, harrowing or screefing) should be undertaken in autumn or very early in spring. Studies to determine the best soil treatment to ensure limited cover of competitive vegetation indicate that removal of topsoil is preferable (Karlsson, 1996).

growing individual stems. In studies of dormant buds on birch, most have been found close to the ground: 0-10 cm above or 0-5 cm below ground level (Kauppi, 1989; Kauppi et al., 1987; 1988 Johansson, 1992a). The number of sprouts per living birch stump has been found to vary between 1 and 52, mean 10±8, decreasing to 3-8 sprouts per stump after five years (Johansson, 1992 b, c). Rydberg (2000) found the number of birch sprouts had decreased by >40 % of the initial number two years after stump creation nine years after cutting, Johansson (2008) found that the initial number of sprouting birch stumps had decreased to 61 and 55 % respectively for downy and silver birch stumps. In a study of downy birch growing in central Finland, the number of sprouts decreased from an average of 9.5 one year after cutting to 5 after three years and 3 after seven years. The sprouting abilities of red oak (Quercus rubra L.), white oak (Quercus alba L.), black cherry (Prunus serotina Ehrh.), sugar maple and yellow poplar (Liriodendron tulipifera L.) growing in West Virginia were studied by Wendel (1974). After ten years the number of sprouts per living stump was 15-20 % of the initial number produced. In another study of yellow poplar, the average number of sprouts recorded six years after cutting was 7.0 per stump (Beck, 1977). Sprouting capacity is highest when a tree is young (Johansson, 1992c). Kauppi et al. (1988) reported the poorest sprouting results from old (40 year) downy birch stumps. Older trees have thicker stem bark, so the buds cannot penetrate the bark and develop into sprouts (Mikola, 1942). Sprouting capacity may depend on carbohydrates in the roots. However, Johansson (1993) found no pronounced peaks in the carbohydrate content in birch roots during the year. Sprouting capacity may also depend on the cutting date. Johansson (1992b) found the highest number of living birch stumps producing sprouts cut in all months but June-October. Etholen (1974) found no effect of cutting time on the sprouting ability of young downy birch stumps.

In southeastern New York, Kays and Canham (1991) studied the sprouting ability of four hardwood species: red maple (Acer rubrum L.), gray birch (Betula populifolia Marsh.), white ash (Fraxinus Americana L.) and black cherry (Prunus serotina Ehrh.). They reported that gray birch had the highest mortality (87 %) of stumps after cutting in May but the other species only had mortalities of 10-20 % depending on cutting date. In a study of the suckering capacity of parent trees of American beech, a mean of 41,365 (3,924-89,765) suckers ha-1 was found (Jones and Raynal, 1986).

European aspen and trembling aspen (Populus tremuloides Michx.) are two Populus species with a high capacity for sucker production. The number of suckers after cutting the mother tree differs depending on the cutting date (Johansson, 1993) and on site, stand and management factors (Frey et al., 2003). The age of the mother tree also influences the suckering ability (Brinkman and Roe, 1975). A trembling aspen stand was found to produce 8000 suckers ha-1 after cutting (Tew, 1970). In a study by Alban et al. (1994) of trembling aspen growing in Minnesota, the number of suckers the first year after disturbance was >250,000 per hectare. The number had decreased to 40,000 after five years (Stone and Elioff,

1998) . Trembling aspen stands growing on similar soils in Minnesota and British Columbia produced 50,000 suckers ha-1 after five years and the mean sucker height was 2.1 m (Stone and Kabzems, 2002). The root system of an individual aspen is widely spread, with root lengths up to 20 m (Reim, 1930). In a Swedish study, about 70 % of the suckers occurred within 10 m of the parent aspen tree (Barring, 1988). In a study by Johansson (1993) the content of starch in roots of European aspens fluctuated during the year with the lowest levels in May-July. The same pattern has been reported for trembling aspen by Baker (1925), Zehngraff (1946), Tew (1970) and Brinkman and Roe (1975).The lowest content has been recorded in late May and early June. When aspen is cut in the winter the highest numbers of suckers are produced (Stoeckler and Macon, 1956; Steneker, 1976; Peterson and Peterson,

1992) . In other studies (Shier and Zasada, 1973; Fraser et al., 2002) on trembling aspen, no relationships have been identified between carbohydrate content in roots and the number of suckers initiated.

Alder regenerate vegetatively by sprouts or suckers depending on species. In a study of red alder (Alnus rubra Bong.), the number of sprouts per living stump ranged between 5 and 9 (Harrington, 1989). In another study of the same species, the number of sprouts was in the range 9-13 (DeBell and Turpin, 1989). According to Rytter (1996), young grey alders (Alnus incana (L.) Moench) produce sprouts after cutting, but the old trees produce suckers. In a Finnish study, grey alder stumps sprouted within three weeks of cutting (Paukkkonen and Kauppi, 1992). Sucker production by grey alder is the main means of vegetative regeneration when the trees are more than 25-30 years old (Schrotter, 1983). In a study of seasonal variation of carbohydrates in the roots of common and grey alders, levels were found to be highest during September-November (Johansson, 1998). In a study of the influence of felling time on sprout and sucker production by common and grey alder, the carbohydrate content in the roots was found to influence biomass production (Johansson, 2009). The highest number of sprouts from common alder stumps was produced after cutting in August-October (23-24 sprouts stump-1). Ten years later, the number of sprouts had decreased to 1.3-2.3 sprouts stump-1. The average number of sprouts on living grey alder stumps was highest after cutting in March (3.0), August (3.4) and September (3.4), with a reduction to an average of 2.0 after five years. The number of grey alder suckers per m2 was highest, 21.0, after cutting in September with a reduction to 1.5 after five years. The recommendation, therefore, is to cut grey alder in August and September ad common alder in August-October when the largest number of sprouts and suckers will result. In a study on the initial sprouting of 4-year-old red alders, the percentage of sprouting stumps was highest when the alders were cut in January (Harrington, 1984).

In a study of the spouting ability of Eucalyptus in plantations, the number of sprouts per living stump varied, but the highest number was 5-6 sprouts stump-1 (Sims, 1999). The stumps have the capacity to resprout several times, depending on their vigor.