Seed from these species remains dormant due to the presence of hard seed coats which contain cutinised layers which prevent the passage of water to the embryo.  Under natural conditions this physical or true hard seededness is removed gradually as a result of biological decay.  This may take years and result in sporadic germination.  To obtain a more uniform Rootrainer crop synchronous germination is needed.

The dormancy can be overcome by breaching the integrity of the relatively thin impermeable brittle seed coat and so permit imbition of the embryo.  This can be achieved chemically or physically in a number of ways, but the simplest method, especially where relatively small quantities or seed are involved, is mechanical scarification.  If the seed is large and too valuable to risk bulk treatment it may be chipped by hand.  More commonly a single layer of seed is put in a tray lined with sandpaper and gently abraded by rubbing with a sandpaper block.  There is usually no obvious difference in the seed after treatment.

The proportion of hardseeds before treatment varies depending on factors such as length of storage, maturity on collection, humidity etc., thus the optimum period of scarification varies for different samples.  To determine the optimum period a sample of seed is abraded, at intervals of 1-5 minutes samples of one hundred or so scarified seeds are removed.  These are then soaked in water overnight.  The seed with coats that have been breached will be imbibed, i.e. they will have taken up the water and will have swollen.  The proportion of imbibed swollen seed compared to hard seed can then be calculated for each sample and will indicate the ideal period of scarification which can then be used on the bulk of the seed.

In any sample of stored dried seed a small number wil normally have lost their hardseededness even bfore treatment.  These seeds may have low viability.  They could be removed by soaking them and sieving off any swollen seed.  The remainder would then have to be redried before scarfication.  Unless the seed is known to have stored badly the  number of seeds affected is usually small and so this procedure is not normally necessary.

The Rootrainer cells should be filled to within 1/4" of the top.  The seed can be precitted in seed trays and then pricked out into the Rootrainer cells, however disturbance at this delicate stage can retard the subsequent development of plants.  Best results are achieved by direct seeding of the cells.  In the large seeded species such as Robinia and Laburnum, losses should be low and one seed per cell should be sufficient.  For the smaller seeded types -Cytisus two seeds per cell is advisable and any excess plants are thinned out.  In all cases the seed should be covered by  sprinkling a thin layer of sand equal in depth to the thickness of the seed to prevent drying out.

The smallest Rootrainer suitable for these species is the Sherwood.  There is a great variety in the rate of growth between the different species.  As a guide dor a fast growing example such as Robinia, from a seed sown in March a plant of at least 90-120cms can be produced by September. Laburnum species should reach 45-60cms by August/September, but larger plants can easily be achieved by sowing early in March and applying a little heat to ward off sub-zero temperatures.  

Small Ulex grown for liner production can be sown in Rannoch, but for early sowings and for Cytisus in the range 30-45cms Sherwood can be used.