Clonal selection using the methods of mass and individual selection is directed toward isolating already existing, economically most valuable clones from a variety population. However, these methods do not allow for the deliberate capture of new somatic changes affecting individual shoots or branches. The source for obtaining fundamentally new hereditary forms in vegetatively propagated perennial plants is bud selection, which is based on the selection and fixation of spontaneous bud mutations.

The Nature of Bud Mutations in Grapes

Summaries of scientific data, initiated as early as by Ch. Darwin, confirm the wide distribution of somatic mutations among cultivated plants. In grapes, bud mutations are not a rarity. In the ampelographic literature, there is a fairly large number of indications regarding deviations in alternative, clearly visible traits in individual shoots of a grape vine. Most classical historical varieties that differ exclusively in berry skin coloration arose precisely as a result of fixing such mutations. Science knows the following examples:

  • Changes in berry coloration: The finding of white berries on Chasselas rose in Crimea (Korzhinsky, 1910); the occurrence of black berries in Pinot gris and pink berries in Aligote (Merzhanian, 1928); frequent mutations in fruit coloration in the Burgunder group (Mittmann, 1924); and the observation of a bud variation of Pinot gris with white clusters by Negrul in Odessa. In the south of France, black mutations are frequently recorded in Terre blanc. There are forms of Terre bourret which, upon cutting, stably yield segregation into vines with white and black berries, as well as panaché (variegated) forms whose berries are divided into gray, white, and black sectors. A similar phenomenon of variation in fruit coloration takes place in the varieties Carignan noir, Pinot gris, Aspiran, and Gamay, in which multiple shades arose as a result of such mutation — black, gray, white, pink, green, matte, and red.
  • Changes in flower morphology: The discovery of a unique bud mutation with a true female flower type on a vine of the hermaphroditic variety Mourvèdre (Tupikov, Baranov). Although in this case the origin of the variation from a bud mutation has not been strictly proven, there is no other way to explain the appearance of this form. This mutation is of exceptional interest because it yielded flower properties that are classified by taxonomists as belonging to an entirely different genus (Baranov).
  • Changes in berry size and seedlessness: Bud variations in berry size are rare, but large-berried mutations are known in the varieties Jacquez, Othello, and Petit Bouschet. Seedless forms (Korinkas, Kishmishes) are also viewed as bud mutations. Tetraploid forms such as Muscat gigas and Sultanina gigas (2n = 76) were, in all probability, obtained from a bud mutation.

The numerous cases of identical mutation and the presence of panaché forms lead to the assumption that all the listed varieties represent chimeras, and repeated mutations are the manifestation of one or another cell layer component.

Theoretical Principles and Technique of Selection

Until relatively recently, bud selection was applied by practical gardeners accidentally, without special scientific methods. Selection was performed only when sharp, easily noticeable deviations occurred that interested the gardener. It was only recently, following the works of de Vries, Korzhinsky, Bateson, Chittenden, Shamel, and others, that bud mutations drew the attention of researchers and bud selection began to take shape as a distinct methodology.

The essence of this method lies in the continuous study of individual plant shoots, identifying their deviations from the normal type in morphological, biological, and economic traits, and establishing in the vegetative offspring that the discovered deviation is indeed a mutation. It is well known that different parts of a plant develop unevenly. This difference is especially pronounced in perennial woody plants due to different developmental stages (ontogenetic variance) of plant parts, unequal development conditions for individual elements, and the long-term differentiated effect of the environment on different sections of old wood.

All these and many other causes lead to significant variation among parts of a single plant — unequal shoots, leaves, and fruits in size, shape, color, pubescence, etc. Such a difference in the development of individual organs of perennial woody plants forces us to consider a tree or bush as a complex individual and to study the development of its branches and shoots separately. This circumstance must be taken into account during bud selection.

The technique of bud selection currently applied is as follows. In plantations or collection vineyards of those varieties that we want to improve in a specific direction, observations are conducted throughout the summer over the shoots of as many plants as possible. Alternative traits are usually taken into account: the shape, size, and color of fruit, leaves, buds, etc. Due to technical difficulties, traits such as yield and resistance are rarely recorded directly; they are usually judged by correlative morphological markers.

If it is discovered that a shoot deviates in a particular trait from the general type of the entire plant or variety, an attempt is first made on the spot to establish the causes of this deviation. In the event that after examining the plant as a whole and its surroundings, no causes for the found deviation are detected, a small numbered tag is hung on the shoot, and the observation data are entered into a notebook. Following this, sporadic observations are conducted on the marked shoot, and measurements regarding other traits are recorded each time.

At the end of the year, based on all data accounting for developmental conditions and any confounding circumstances, as well as on the correlative dependence of the aggregate of traits, the marked shoot or branch is definitively recognized as a bud mutation and is designated for verification in the vegetative offspring or is excluded from the experiment. Verification in the vegetative offspring is performed either by grafting cuttings from the marked shoot onto adult vines of another variety (for rapid evaluation) or by simple routing of cuttings and subsequent testing on a cloning trial plot. To accelerate the work, those bud mutations that possess great economic value must be put into mass propagation for final verification of their authenticity.

General Evaluation of the Method

Giving a general evaluation of this method of clonal selection, we must note that despite its major role in obtaining new grape varieties in the past, currently, when breeding is required to ensure the rapid production of necessary varieties, much cannot be expected from it. The situation will change only when bud selection can be built not on the registration and selection of accidental mutations noticed by us, but on the selection of more frequent bud mutations obtained experimentally (such as induced physical or chemical mutagenesis).

As can be seen from all that has been stated before, clonal selection does not make it possible to obtain deep genetic changes in plants, but merely leads to the improvement of the varieties cultivated by us. With mass selection, the cleaning of the variety from poor clones and the propagation of the healthiest plants are achieved. With individual selection from a mixture of clones, we obtain a pure sorted clone variety, thereby cleaning the variety from low-value admixtures. In both cases, a large practical effect is often obtained due to a significant increase in yield, etc.; however, we do not obtain a fundamentally new variety, but merely improve the old one.

With bud selection, new hereditary types (genotypes) are obtained. However, as is well known, bud mutations usually affect a small number of traits. Thanks to this, through bud selection, we obtain the same variety but with one or several altered characteristics. This circumstance contributed to the fact that varieties obtained as a result of clonal selection usually do not change their primary names, but merely add "clone No..." or "white mutation," "red mutation," etc.