Modification and refinement of breeding techniques

During the course of a breeding programme extending over 50 years, experience has provided opportunities to refine both theoretical approaches to the work and practical techniques. Some of these developments, notably those applying to progeny-testing, have already been mentioned, but others which do not clearly fit into the way in which the programme has been described so far are dealt with in this chapter.

Intensity of plus-tree selection

When plus-tree selection was first carried out, it was to a very high selection intensity. Much manpower was devoted to this task which at times proceeded very slowly because of the stringency of the selection criteria used. Those adopted for Sitka spruce were laid down by Fletcher and Faulkner (1972), but the same protocol was broadly used for all species. Selections had to be fast growing (and exceed neighbouring dominants by 10%) , straight, have fine branches, show spiral grain of less than 6 degrees, and be disease free. It was estimated (Fletcher, 1992) that the selection intensity was about 1:75,000 (one tree per 25 hectares approximately). Trees were selected regardless of whether they were flowering. Those which were not flowering would be revisited annually in an attempt to obtain seed for progeny testing. Following selection, most plus trees were systematically archived in clone banks but this process was restricted by the capacity of the annual grafting programme. As clone banks matured, the alternative method of acquiring seed for progeny testing arose using artificial pollination with pollen mixtures.

As early progeny test results were collected, it became apparent that many of these intensively selected superior phenotypes did not emerge as superior genotypes based on the measurement of their progeny; some were even inferior to the QCI unimproved direct import used as a standard control. The mean performance of the open-pollinated families was around 7.5% better than the QCI control. Heritability estimates of early height growth from the 'Population Study' confirmed that the environment would be expected to influence the appearance of the phenotypes by around six times more than the genotype.

In 1973 the decision was made to reduce the former intense method of plus tree selection. The practice then became to visit good quality stands of known origin and to select a large number of above average individuals which were flowering so that cones could be immediately collected, seed extracted, and progeny tests established just two years later. The selection intensity now fell to between 1:40 and 1:1,000. As a further cost-saving measure, the trees had to be relatively young. In this way the genetic quality of a parent tree could be evaluated from progeny tests before it reached rotation age so that it would be necessary to graft and establish in clone banks only those plus trees which were subsequently found to be superior through testing. This new practice sometimes led to plus-tree selection taking place at a very early age eg 300 trees in 20-year old stands in Glentrool forest.

As a result of these changes, the superiority of open-pollinated families fell to around 5% better than the QCI control. The screening of more plus-trees in progeny tests was therefore required to obtain a given number of parents for the breeding population. A change of emphasis in funding from plus-tree selection, to progeny testing had taken place.

Nursery techniques in raising progeny

A standardised method of raising material for progeny testing was developed during the 1960s (Faulkner, 1967). This involved the use of raised beds, 1 metre wide in which seedlings were pricked out in rows of 20, one or more rows forming a plot containing a single family. Each bed was contained within vertical boarding and provided a uniform, easily accessible and manageable system. Detailed experimental designs were later developed so that blocks could be confined to individual beds. Plants were raised in two seasons under this system.

Progeny-raising over the period 1982-4
Progeny tests created 9
Total families to be tested 405
Lines in progeny beds 4 136
Plants pricked out 83 235

However, when the peak of progeny testing was passed, the opportunity arose to investigate the production of container-grown plants in a single growing season. This more intensive system also provided the facility to manipulate plant size to satisfy the greater range demanded by restocking sites. Progeny material is now routinely raised as one-year-old cell-grown plants.

Grafting work

The grafting techniques established at Grizedale (Lakes) in the 1950s and 1960s (Lightly and Faulkner, 1963) involved grafting onto 2- or 3- year old rootstocks raised under normal open nursery conditions; this became the established practice in the branch. However, by the 1970s, all grafting had been brought together on one site at Bush nursery, attached to the new Northern Research Station. This provided the opportunity to develop a range of facilities for grafting work, in particular polythene houses with heating and lighting facilities. By the 1980s a more rigorous approach to grafting had increased the success rate considerably. The main features of this were the use of fast grown 1 year old rootstocks grown under polythene with heat and light and the continued maintenance of grafts under the same conditions in the early stages. The use of freshly collected scion material also contributed to this success.

At its peak, the grafting programme reached 10 000 per year across all species.

Average grafting programme for 1982-5 for the major species
  Sitka spruce Scots pine Corsican pine Larch
Number of grafts made 6013 982 821 1474
Number of grafts successful 3477 827 646 1134
% success 58 84 79 77

It is clear that grafting success varies with species; some improvement on these figures has been made in recent years.

Artificial pollination techniques

The first artificial pollination work was carried out at Thetford (East Anglia) in the 1950s. The techniques involve the isolation of female flowers from extraneous pollen before they are receptive, their pollination with specifically chosen pollen and their subsequent release when the chance of extraneous pollination is gone. Flowers were initially isolated in paper bags, sealed at the mouth with cotton wool and tied with raffia. Ripe male flowers can be removed from trees and pollen carefully extracted and containerised; at first milk bottles were used.

The appearance over the last 50 years of a range of materials now commonplace has contributed to the effectiveness of these techniques. One of the major innovations of the branch was the use of the more robust Terylene for isolation bags, and this was used for many years before the advent of stronger paper bags specially designed for the work. Subsequently, sheet celluloid was cut and made into tubes which were then sealed at each end with a foam rubber bug, split at the bottom to introduce the flowering stem. Only more recently have these and the fearsome solvent used to seal them been superceded by a ready-made plastic form of tubing. Raffia has been displaced first by paper-covered and then by plastic-covered wire twists.

The following table summarises the achievements of pollination programmes over the peak years of activity (1982-5):

Achievements of pollination programmes over the peak years of activity (1982-5)
  Sitka spruce Scots pine Lodgepole pine Larch Total
Number of crosses made 473 102 155 205 875
Number of isolations made 6 333 2959 3788 3 207 14 685
Number of flowers isolated 21 435 7240 9023 12 213 46 025
Number of pollen lots collected 108 55 136 52 341

One of the main problems encountered in artificial pollination work is that female flowers will have their development accelerated inside isolation tubes and have passed receptivity before pollen is shed. There was thus the need to find the conditions under which pollen will remain viable under storage over long periods. For most species (larch is normally the exception), pollen will remain viable if it is dried to below 10% moisture content and kept deep frozen at -20oC. Under these conditions, viability can be maintained for up to five years and the use of these techniques has enabled pollination programmes to proceed in a more predictable and successful way, given the seasonal variation in flowering under British conditions.

Flowering studies

From the start of forest genetics work, it was recognised that the study of flowering with subsequent seed production should be a major area of research. Knowledge of the underlying factors controlling the processes was limited and it was decided to collect phenological data over a period of years to investigate any relationship with meteorological data. These studies, started in 1951, were a major investment in time and data collection from eleven broadleaved and seventeen conifer species continued for many years.

A number of factors began to emerge, but before large-scale processing by computer was possible it was difficult to correlate the vast amount of data collected. However, a few possibilities were suggested by the results and these had a strong influence on later work. It appeared that high summer temperatures produced good flowering in the next year, whereas the analysis of data for light and nutrition was inconclusive.

It was during this early period that physical treatments on trees in the forest and in clone banks were made. Partial girdling (removing separated, overlapping but incomplete bands of bark) to restrict nutrient movement into the crown was investigated, but the uncontrolled conditions in the forest made the results of a wide range of experiments unreliable. It was decided in the late 1950s to adopt a more physiological approach and to limit the number of variables and species under test.

The physiological studies were transferred to the universities whereas field experiments continued within Genetics Section. In-house work therefore continued with the collection of phenological data and the establishment of a series of more controlled field experiments testing the effect of several treatments which included root and shoot pruning, stem and branch girdling and fertiliser and herbicide application. This research continued through the 1960s but no definite conclusions could be drawn.

The work in the universities, particularly at Manchester and Aberystwyth was beginning to change the fundamental thinking behind the flowering studies. The idea of maturation in trees became accepted and this supported the idea that flowering could only occur if the tree has passed through all the epigenetic stages of phase-change in moving from the juvenile to the mature phase. The researchers felt that simple phenological observations and subsequent environmental treatments were not enough and that there was an underlying mechanism that controlled and could be used to manipulate the flowering mechanism. Results in the early 1970s from a number of research organisations around the world indicated that the treatment of plants with one group of plant growth regulators, the gibberellins, had a profound effect on subsequent flowering.

Some preliminary work with gibberellin treatment of species such as Western Red Cedar had been carried out in the early 1970s and it was subsequently decided to undertake a major study on Sitka spruce which would combine the new concepts of growth regulator control with the older practice of physical treatments of the trees. This research, in collaboration with Long Ashton Research Station, quickly demonstrated that a significant increase in the level of flowering in grafted plants could be achieved by treatment with a mixture of gibberellins. It also indicated that the production of seed from plants managed under polythene was a possibility.

A project was established in the early 1980s, initially in Physiology Branch and subsequently transferred to Tree Improvement Branch, to build on the Long Ashton work. The studies concentrated on Sitka spruce and hybrid larch with the aim of identifying the optimal concentration and times of application of gibberellin for floral induction. The work, carried out on potted grafts under polythene, was combined with other treatments such as heat, drought and root pruning. Parallel studies with older field-grown grafts were combined with girdling. The research proved to be very successful and showed that flowering can be consistently induced in potted grafts of both species maintained under polythene by stem injections of a mixture of gibberellic acids four and seven (GA4/7) in combination with drought and high temperature (Philipson, 1987). Large polythene-covered flowering halls were constructed at Bush in the late 1980s, and this is now an established technique in breeding work and in the production of seed for mass vegetative propagation.

The induction of flowering in field grown plants does not give such predictable results due to the less controllable conditions. However, GA4/7 injection is routinely used in clone banks and has been found to boost flower numbers in moderate or good flowering years.

Research into flowering finished in 1996 with the publication of a manual summarising use of the methods developed (Philipson, 1996).

Clone bank establishment and management

The planting of clonal archives began as early as 1951, since when more than 30 individual clone banks have been established. However, some of these were small and short-lived. As the number of selected plus-trees rose in the 1960s, a policy of establishing larger reserves was favoured. This provided the opportunity to bring together as many clones as possible on a single site, making artificial pollination work easier and more efficient.

The initial policy of systematically archiving all selections was pursued until the late 1970s. The most common plan for clone banks was to plant 6 grafted ramets of each clone in numerical order in rows spaced six to eight metres apart with two metre spacing within rows. Banks were located in areas in which the local climate was likely to promote flowering; southern England, the Morayshire coast and Perthshire were the most common. At least two major former FC Conservancy nurseries were utilised and effort was also made to place banks near to the research stations or former Genetics Branch outstations. Clone bank sites are normally fairly flat and a grass sward is routinely maintained between rows to provide easy access for powered machinery

During the 1980s the grafts in clone banks began to reach heights which were inaccessible for pollination work and a series of trials was set up to investigate the top pruning of grafts to a height of two to three metres. This proved to have little effect on overall flower production and was thus included in routine clone bank management. Top-pruning is now recommended from around 6 years after planting on a two to three year cycle with reduction in side growth as it becomes necessary.

Because only 20 to 30% of the base population of plus trees originally selected were subsequently identified as members of the breeding population, large areas of original clone banks became redundant. When re-grafting became necessary because of tree size, only members of the breeding population were repropagated and planted in new areas. By 1998, most clone banks contained large numbers of over-mature grafts, many of which were now redundant in the breeding programme, together with younger material of breeding clones. Following a review of policy, it was decided that clonal archiving will in future be carried out on the following basis:

  • Breeding clones will be planted in two geographically separated banks in which two ramets per clone will be established
  • A proportion of the remaining base population clones will be regrafted on the same basis in the interests of conservation of genetic resources, in particular concentrating on clones of special interest such as indigenous Scots pine clones
  • Tese new archives will be phased in to replace existing banks over the next 5 to 10 years, and regrafting on a 25 to 30 year basis is anticipated.

In future, it is suggested that the chief purpose of clone banks should be to provide a safe archive of genetic material. When the current crosses being made for second generation work in Sitka spruce have been completed, the main artificial pollination work will be in the interests of providing seed from production populations to be used in mass vegetative propagation. Following the work on flower initiation in potted grafts under polythene, it is felt that this will be achieved most efficiently by concentration on these techniques to produce this seed for both Sitka spruce and hybrid larch.

The following table summarises active clonal archives for all species and compares this with the level of work under the new proposals:

Summary of active clonal archives for all species and comparison with the level of work under the new proposals
  Sitka spruce Scots pine Lodge-pole pine Corsican pine Euro. larch Jap. larch Douglas fir Beech Oak Misc
Stock (Jan 1998) Sites 5 4 1 4 6 5 3 1 1 5
Area (ha.) 10.4 3.5 2.4 0.8 2.5 2.0 1.6 0.2 0.2 0.7
Clones 1455 967 499 243 450 259 248 84 39 173
Future proposal Sites 2 2 2 2 2 2 2      
Area (ha.) 4.8 2.4 1.0 1.2


Clones 1000 500 200 250



Notes: Areas calculated on a standard spacing of 6 m x. 2 m (833 plants per ha).
Some clones will appear in more than one bank

What's of interest

These pages review the work performed by the Forestry Commission and Forest Research on tree improvement following the 50th aniversary of its establishment which passed in 1998.
The genetic background describes the scientific procedures of tree breeding and the technical terms used in the remaining pages.
All species are referred to by their common name in English.