Recently I was one of several gardeners putting together an excellent book on gardening techniques (-that I'm too humble to mention other than it was associated with a well known BBC radio four gardening programme.......). However the point is that we all had to agree on who was to write on what and on which topics to include or exclude. Most subjects are simple enough to choose but what about grafting & budding. It is done by the millions for commercial purposes for roses, fruit trees and shrubs. Yet it is something hardly any amateurs at all ever seem to do, or even attempt.

I must admit of having been a bit damning myself on aforesaid radio programme, mostly about my own fumbling attempts. You see it is just like with any other skill; it seems obviously simple when the professional demonstrates it but as soon as we try it at home things seem to go awry. There are so many factors; having to lay aside grafting material beforehand, choosing the right time (crucial so it seems), choosing the right spot on the right side in the right manner of the right sized bits and using a blade at just the right angle and with just the right pressure... it's hard to get that bark to lift, then it splits... and the grafting stick gets whittled down to a bud unintentionally..... And when eventually all is assembled nearly successfully then it needs binding, just right.....

So in the past I made quite a few various and awkward attempts at grafts and buds and most rather predictably did not work. Maybe in this skill I am particularly inept but certainly over the years I have met comparatively few gardeners who skilfully practice this apparently simple task -excepting those who have done so commercially. Yet I can't help wondering why we don't all have more of a go as after all the potential could be rather interesting and greater than we first imagine.

And sometimes I have been amazed at my own unwitting success. Once when a new apple tree was snapped almost completely through by a visiting child it was carefully refitted together, bound and strongly splinted to a steel pole for temporary support. Although only a shred of bark had remained connected the whole healed and the joint is now imperceptible. Recently I had to cut the top off a huge old pear and decided to try and work it over to better varieties so half a dozen different sorts were tried yet only one took. Another time a plague of rabbits, debarked many of my newly planted orchard trees. I was distraught, some were only partially barked and might recover but several were totally devoid and certainly doomed. Consulting various friends and books bridge grafting was deemed worth a try. Supple pieces of young strong stems were grafted in parallel to the trunk to replace it. These had to be curved so they could be grafted in top and bottom and so accuracy was crucial. Was I surprised when most took!

I got to thinking about it and the grafts that took were where they had all been repaired with bits of their own young stems, most of the failures were on one or two trees which being short on new wood had been given stems from neighbours. An important point emerged- grafting is rather like spare part surgery or blood donation, some donors are easy and others near impossible. It seems easily successful with (fresh) bits of the same plant repairing itself but more difficult the remoter the relationship....

In theory with grafting, or budding, you place two pieces of (compatible) plant together so that the living tissues (usually of the 'top' scion and 'bottom' root stock bits) are in sufficient contact to unite and form a continuous bond. This occurs more or less and the two tissues do merge -but not all mixtures are equal and some bonds are stronger and easier to get to take well than others. (a bit like blackcurrant cuttings root easily but plum cuttings do not) Frequently although the living bark merges the wood does not so there remains a weak place in the trunk. Grafted fruit trees on dwarf stocks are sensibly staked to prevent these snapping at that point. (I lost two pear trees this way when their posts decayed.)

Some bonds are so difficult or remain so weak that an inter-stock piece may be grafted in between to bridge the two nearly incompatible pieces. This may otherwise be done with the aim of restricting growth in such as 'double worked' cherries where a suitable inter-scion is chosen to act as a constriction. It may sound difficult or complicated to get some of these various grafts to take. But then other plants, allegedly, stick back together just as easily as blackcurrants root -apparently many cacti can be grafted rather imaginatively! (Not much use to us Kitchen Gardeners though unless climate warming accelerates even more.) However this knowledge of what may be grafted easily is lacking.

Our books, being based on commercial practice and Victorian gentlemen's imaginations, resulted in two veins. The first is the purely practical budding and grafting of mass market numbers on a near industrial scale. We should examine these methods for general success. However the second vein contains the most amazing assemblage of the most curious grafts of all sorts, sizes and descriptions. These latter works make confusing and seemingly pointless discriminations between many near identical processes. It is possible they found such diversity necessary to succeed with the less common situations, or maybe it was all a fashion... What both veins however miss is sensible analysis and listing of what can actually be got to graft on what and with which method.

We are usually informed that widely different species and genus cannot be grafted -so for example you cannot graft a lemon stem onto a pear tree and have it live. Yet that is exactly what L.V. Michurin describes in his Selected Works reporting on Russian fruit tree breeding experiments. Indeed in the past, many odd, to us, combinations were used such as pears grafted on hawthorn instead of on quince roots as now. You see so often what is said to be so may be true not in a factual sense but an economic. one...

If you asked a commercial nursery whether pear or lemon cuttings could be either rooted or grafted on each other they'd probably say of course not -yet they can. When someone (else) on that well known radio panel stated that Gingko could not be rooted from cuttings we were soon dismayed by photos arriving showing success! You see if you commercially set one hundred cuttings or one hundred grafts or buds then you need something like ninety percent success or you go out of business, and you need the results fast. So often we are told such and such cannot be rooted/ grafted/ budded or even grown when in fact a small percentage may just get away with it. It is not economically feasible but may still happen sporadically. After all we only saw photos of those singular gingko cuttings that did root and not of the guestimated 103,674 that failed!

So we may be able to get an unfruitful fruit tree to prosper by grafting a suitable pollinator onto it. Obviously it helps to know which variety would work but it may suffice to graft half a dozen or so different ones. Of course the best one may not graft easily but if we do several grafts of each variety we are more likely to succeed. And this may all be much less effort than planting one new tree as a pollinating partner.

Now most of us are familiar with family trees; these are several different fruit varieties all grafted onto one trunk and roots. This has advantages and drawbacks but suffice to say it is commonly done -and not a new idea; the Romans did the same! Likewise if you so wished you could create family currant bushes. The Romans also created family grapes and family shrubs; several different coloured or even different sorts of grape or shrub all proceeding, apparently, out of one stem. But of course, then as now, there were cheats. It was expensive, slow and difficult to successfully graft a family vine or shrub, it was quicker and much cheaper to fake one.

A fake family shrub was made by inserting a bundle of three or four suitably sized and matched cuttings into hollowed out small trunks which were then rooted, grown on and potted up. Once the heads formed the fraud was complete. Vast numbers of these were produced, some with pottery or metal trunks. Before many years the stems swelled, and constricted by the trunk, one or more branches died off and eventually all. However in one or two cases apparently a botanical miracle occurred. Buds being forced to grow through exceedingly restricted places very very occasionally grew through another and gave rise to fantastical plants called chimeras.

These still occasionally occur and are known as graft hybrid chimeras. Laburnocytisus adamii is one of the best known examples. A select variety of cytisus was being budded onto laburnum used as a cheap rootstock in a nineteenth century Paris nursery. A mishap enabled an existing bud to fuse with the new creating a stem that was unique in a sort of raspberry ripple way. This strange fusion was itself graftable onto new roots and you can buy an example today The stems and leaves and most especially the flowers resemble one parent or the other but most are an odd mix. You get yellow blooms, and purple blooms and mixed flowers, all on different parts coming from one trunk on one set of roots.

There are others of these weird graft chimeras; Crataegomespilus exists in two forms; one where the M. germanica Medlar coats a core of C. monogyna thorn which breaks through, giving branches of both but mostly Medlar like fruits, and another form the opposite. Pyrocydonia danielli is a chimera found at the graft junction of a William's pear on quince roots, it is a quince like tree with downy leaves and large apple like brown fruits. Another hybrid; the lilac Syringa correlate has an outer sheath of white flowering S. vulgaris with a core of lavender S. x chinensis breaking through.

Where all this leads is that it substantiates the Roman claim they were able to produce grapevines bearing family bunches of grapes- three different sorts of berry in the same bunch! The 'cheated' three cuttings in a tube giving three sorts of bunches on separate stems was claimed to occasionally give rise to exactly the mixed bunch result if the three all fused together and were then headed back. I tried this and failed. (However their even more implausible way to get seedless bunches has, rather annoyingly, worked. To get seedless grapes- take a good cutting, slice it through lengthways and remove all the pith then graft it carefully back together again and root it. This seemingly should not work- but it has done; the grapes produced are more seedless than the parent!)

So it seems although 'unusual' grafting combinations and even weird results may be difficult to get to take they may not actually be impossible; as with lemons on pear trees, it's just they have never been tried repeatedly enough, or were not tried by the right technique. It may be inarching is the answer; this is to grafting the equivalent of layering to cuttings. Also known as approach grafting this method keeps the grafting pieces continually attached to their parents until a union has occurred -which then happens to some extent in a surprising number of cases.

Too often grafts or buds fail because the union does not form before the new bit has lost all it's juices and shrivelled. If the new bit is still attached to it's parent then success is obviously much more likely. A new bit still on it's mother plant in a pot can be elevated to a suitable attachment point close to it's new home. In most cases both the stock and scion stems are scraped or half cut through and the exposed surfaces mated together then bound. (often the top of the receiving shoot may also be removed to direct more sap into the new shoot.) If union takes a long time it is little matter as all parts remain sustained. By this crafty method the Romans and other curious people have successfully grafted all sorts of odd combinations.

Inarching is also well suited for grafting softer, smaller and even herbaceous shoots. A commonly done graft is putting more popular or tasty tomato variety tops on KNVF multipli- resistant rootstocks. This is a commercial solution to the problem of a build up of tomato diseases and pests in the soil. Tricky to do by most grafting methods when the stems are soft and match stick sized this becomes easily achievable at home by approach grafting the two seedlings in pots.

This is where potential lies I wonder whether it would be practical to graft cabbages and cauliflowers onto the more club root resistant root-stocks such as the kales? Anyone keen to give it a go? The Victorians also claimed to be able to cut out and rejoin the eyes and small shoots of potatoes to produce graft hybrid chimeras with mixed tubers; say red, white and swirled or spotted all on the same plant. Could this mean the possibility of grafting potato blight resistant tops such as Axonia' or Mira's onto tastier producing 'root systems' such as Arran Victory??? (actually it is the base of the stem not the roots which produce tubers so the graft has to be on the 'tastier' stem and under the soil surface to be safe from blight Why not give it a try!

And it is a very very old trick to graft a tomato top onto a potato bottom to produce a (poor) crop of both. But could this be improved upon....