Most of us are quite happy if not overwhelmed by the very wide choice of fruit varieties on offer to us commercially. And if we want even more to choose from there are rare, obsolete and heritage varieties to search out. The internet even offers the possibility of buying (certified for importation to be legal) foreign sorts as well though these may obviously be ill suited to our general growing conditions. So why should any of us ever want to develop our own varieties?
Well, it may be you want to grow something really really different not already sold such as a 'blood' pear similar to the Sops in Wine red fleshed apples and ruby oranges. Maybe you want a variety resistant to some pest or disease or resilient to local weather conditions. Perhaps you desire a bigger, sweeter or oddly shaped fruit. Or more likely, like me, you are just curious or vain enough to want to have your very own sorts just for their sake alone.
But will it be worthwhile? This depends on the effort taken, the value of the new variety, and the ease with which it can be propagated. Well the effort may be little as we all find chance seedlings growing here and there, and sometimes we come across one fruit that is different to all others on a plant. In general these are the two ways we may get a new variety; by seedling variation or by bud mutation. The latter is where an existing variety spontaneously produces a different fruit from a mutated bud. Say an all red apple on a tree normally producing all green ones. Radiation, some chemicals, frost and chance may so affect buds changing their genes with the resultant fruit being obviously different to others on the same bush. (Of course most mutations are unfavourable, affect other parts and not the fruit such as the leaf shape, or go unnoticed entirely).
Bud mutations can only be propagated vegetatively, usually by grafting the altered piece of plant behind the fruit onto a suitable rootstock. Only extremely rarely does a bud mutation cause any resultant seeds to grow into something resembling the mutated fruit. Seeds from mutated fruits usually produce results very similar to seeds from unaltered fruits on the same plant.
Seedling variation is more interesting as once the 'new' variety has been produced it may usually be propagated vegetatively and there is the possibility that any seed then produced may carry the new genes and give rise to a whole family of similarly endowed new fruits. Usually each will not be exactly the same and only with very few plants can you establish a 'true' line with identical offspring. However once a variety has started to produce seedling variation this process may usually be encouraged to further improve the fruits in a desired direction generation by generation which process cannot usually be undertaken with bud mutations.
In other words; if your purple plum tree throws a branch with yellow fruits then it may be fairly easy to graft buds of that sport onto new plum rootstocks and produce a batch of yellow fruited plum trees. However the seeds of those fruits are not likely to come true and produce yellow fruits but red ones. On the other hand if you grow a thousand Alpine strawberry plants from seed and get one superb plant with blue fruits then the seeds from that may produce some similar though not identical offspring some of which may be better still, -and some worse, and some resembling either parent!
Thus the intending innovator has two avenues. The first is carefully observing their crops hoping for a significant mutation to appear and then carefully multiplying it vegetatively. This may never happen as such bud sports occur fairly rarely and although it is possible to induce mutations this is tricky. So apart from being lucky there is little the amateur can do to improve their chances of finding a favourable bud sport other than growing and inspecting very large numbers. (But don't give up looking -many of our best varieties are due to bud mutations!)
The second avenue is seedling manipulation. Once again chance and luck play a hand as any seedling may be different in a favourable and distinct way. If the plant produced can be propagated vegetatively (as can most perennials but not most annuals) it can be preserved and multiplied indefinitely and as said above may be selected and bred to give other even more improved strains from successive generations of seed. And more importantly we can more easily induce the initial seedling variation than we can induce bud mutation.
Most plants produce a degree of randomness in their seedlings. This enables some of their offspring to succeed even though the conditions are not exactly the same as for the parents. The majority of variations will be barely discernible and of little interest to us, however occasionally a different quality in the fruit may result as the result of chance or our deliberate intervention. If we simply take any fruit and grow all the seeds into fruiting plants some will inevitably do better than others and a few may have better fruits in some way. BUT if we successfully cross pollinate the original fruit's flower with pollen from a distant relative the chances of something unique dramatically improve. The Loganberry and Boysenberry were chance crosses between raspberries and blackberries while the Tayberry was the result of a deliberately made such cross.
The prospective breeder may thus try to cross distant relatives. Done properly with net bags to exclude unwanted pollinators this can give rapid results, many complete failures and only a relatively few seedlings to grow on. However, generally, just growing several varieties of the same or related fruits in close proximity also improves the chances of seedling variation -and usually improves the fruit set and quality anyway. In such a garden situation the odd self sown seedlings may well be quite variable but many seedlings will need to be grown on to find a distinctly improved one.
And of course there is the problem of time and space, and viability. Most tree fruits take such a long time to bear, take so much space and are often available in such wide numbers of varieties there seems little point looking for yet another. In particular apples exist in thousands of excellent, and less good, varieties. Any new apple you come up with has to contend with so many others it will have to be really distinct; say melon sized or strawberry flavoured to ever be commercially viable. The same goes for most of the more popular fruits, but some of the less commonly grown still have great scope for improvement.
Pears, plums and cherries already exist in great numbers of probably quite sufficient variety but quinces, medlars, mulberries and the many edible but undeveloped 'wild' fruits (elderberries, rowans, sloes etc.) all offer the amateur a chance to grow something really exciting, and potentially valuable, if you have the space and time. I grew a batch of apricot seedlings; one looked very different in leaf so I grew it on keeping it pot grown to bring on earlier fruiting. After a few years it was most definitely a pear, I don't know where that seed came from but after a decade or so it cropped- every bit like Conference and a fortnight later ripening.
Certainly it is quicker and takes less space to grow soft fruits. Again the same proviso's apply and it is pointless, if still good fun, to try and grow the most popular as these are available in so many varieties already. It is better therefore to have a go at the less well loved- not only is there more chance of a distinct improvement as these are not so highly bred already but also they may be less well loved simply because they need improving. Several 'wild' berries could be radically improved such as the Berberis or Barberry as it was known and the delicious once jammed but aptly named Chokeberry, Aronia melanocarpa.
Redcurrants are really quick and easy to raise from seed. I have grown dozens of chance seedlings on but so far have not yet discovered one significantly different to those already available. Indeed the various commercial varieties are all acid and barely distinct anyway so leaving room for a sweeter sort, and a huge berried very sweet red or white currant would be a real asset. We could certainly do with a larger berried sort and crosses made with blackcurrants or gooseberries ought to be very interesting.
Blackcurrants from chance seedlings usually prove of little value except as vigorous bearers and as they take longer to crop than cuttings with the added risk of being potentially poorer there seems little point. But again a big berried sweet sort would be useful, I've got one that can throw berries the size of a 20p coin if well pruned and fed. I certainly think hybrids could be valuable like the Josta which is a much improved fruit said to be a cross of a blackcurrant with a gooseberry so showing the potential.
Gooseberries are extremely easy and have been bred for size for competitions and are already quite varied with fruits in almost every colour. Sadly most seedlings are prone to mildew which is a great disadvantage and many revert to a thorny lax form resembling a Worcesterberry. The thorn free and disease resistant sorts are poor eating so they deserve crossing with older tastier forms. I've been lucky and have a seedling which produces fruits with a creamy almost Muscat overtone quite distinct from the usual gooseberry flavour.
Strawberries are already available in vast numbers and the commercial breeders have the advantage especially as most seedlings I've grown seem more inclined to runner than to fruit! None the less they are selecting for shelf life and yield not flavour. Better try blueberries which might just be crossed with cranberries, blaeberries or whortleberries -and we could really do with a big berried sort. And if you were very lucky and got a lime tolerant variety it would certainly sell.
Raspberries offer a really good opportunity- I've grown many seedlings on with superb results; i.e. they have flavour, something which has been lost in most modern varieties which were bred purely for shelf life! And remember these cross easily with the blackberry family. For decades I've been trying to cross the Japanese Wineberry with a raspberry and in particular with the Tayberry hoping for a really exciting development but with little luck so far. Though I have bred a bramble berry which resembles a loganberry but when jammed tastes just like the finest blackberry and apple crumble.
Grapes exist in such number and variety that at first glance it seems pointless pursuing them -yet we could really do with a disease resistant sort for outdoors combining the flavour of such as Muscat Hamburg or Siegerrebe with the reliability of Boskoop Glory. But it might be quicker to develop a self fertile dwarf bush Kiwi which would be a grand garden plant to breed.
However my personal favourite is to get a fuchsia with really large sweet fruits. Few people realise that most commonly grown fuchsias produce edible fruits which have long been used to make conserves and jellies. All fuchsias are currently bred for their flowers thus leaving the field wide open to the amateur grower looking for a new fruit. The berries are not at all unpleasant with tiny seeds and a soft skin, they are bland and slightly sweet. The largest fruited I have found to be the California Dreamer strain which gives fruits of nearly thumb size though not in big numbers and sadly it's not hardy. I predict great fortune awaits whoever breeds a hardy, sweet sort with big berries as almost every gardener in the world could not refuse such a plant- beautifully flowered and tastily fruited; what a combination!