NASC

You need a binocular microscope or magnifying glass and 2 pairs of fine forceps (tweezers) - 5s is a suitable grade of forcep.

Method using genetic markers:

If you have a good clear dominant genetic marker on the pollen then to a certain extent you can just brush the tips of young opening buds with your pollen source - you will then select for the correct progeny using the marker (e.g. a glabrous maternal line vs 'hairy' wildtype pollen). This is very inefficient but can work if the stigma on young buds are not saturated.

If you are working with a cross whereby you have no genetic markers to track self-fertilised seed (the normal case) then you will have to be more careful:
• You need to look at the flower/bud cluster under magnification and remove any siliques, open flowers, or open buds (where you can see the tip of the stigma poking out through the top of the bud). All of these will be self fertilised and useless. You can use the forceps as crushing and tearing 'scissors' to do this.
• It can be useful to first immobilise the inflorescence by making a restraint from a thin strip of paper held with sticky tape (electrical tape works well). This restraint works like a sticking plaster and will hold the stem down without damage to the plant.
• You will need to carefully remove (or displace) the sepals and petals and expose the anthers. They should be immature (squat and pale; shaped like the spades on a pack of cards) so that pollen has not been released. Remove the anthers carefully (emasculation) without touching the stigma or style (it is easily damaged).
• You can then rub (or dab) a suitable anther from a mature flower onto the stigma of the emasculated plant. Actually sticking an anther head (without filament) onto the stigma and leaving it there can make you feel more confident that there is good contact and retention of pollen. Remember that if you are crossing several different lines - you MUST clean your forceps of pollen between lines.
• Once you have made as many crosses as possible on each inflorescence, remember to 'pinch out' the meristem and all smaller buds to prevent further development and confusion with the crossed flower.
• Remember to mark the inflorescence in some way to identify the pollen donor (annotated rings of electrical tape work well and don't hamper the elongation of the inflorescence)
• After a day or two, check the cross - if they have taken successfully, then there will be obvious elongation to generate a silique.

  You will have to take great care to harvest the seed before pod shatter causes you to lose the seed. Some people bag the sliques to avoid this, but we find this method to be somewhat clumsy.