Tree improvement and forest genetics

In short

Tree improvement is the planned use of genetics - selection, breeding, testing and mass multiplication - to raise the volume, form, wood quality, adaptability and pest resistance of planted trees; forest genetics is the underlying science of how variation in tree populations is created, partitioned and inherited. It matters because India's wood demand is met largely from plantations and farm forestry on a short rotation, where a 20-30% genetic gain in mean annual increment is cheaper than any silvicultural input. The work is done by ICFRE institutes (IFGTB Coimbatore, FRI Dehradun, TFRI Jabalpur, AFRI Jodhpur, IFP Ranchi), state forest departments and industry (ITC-Bhadrachalam, JK Paper, WIMCO poplar, Mysore Paper Mills). It is examined in B.Sc/M.Sc Forestry papers, ICAR AIEEA-PG and ARS/NET Forestry, and the Indian Forest Service, and is asked in interviews for plantation R&D, seed-centre and nursery/clonal-production roles.

Explained

Genetic variation and its sources

• Genotype (G), environment (E), phenotype (P): P = G + E + GE. Only the heritable part of G responds to selection, so the central task is separating G from E by replicated testing.
• Trees are mostly outcrossing, long-lived, highly heterozygous and hold huge within-population variation (typically 80-95% of total variation is within a stand, 5-20% among stands - so provenance choice matters, but individual selection matters more).
• Sources of variation: mutation, recombination, migration (gene flow by pollen/seed), genetic drift and selection. Wind-pollinated species (pines, *Cedrus*, *Casuarina*) have long-distance pollen flow; insect-pollinated species (teak and eucalypts, pollinated by thrips, bees and birds) and heavy-seeded ones (sal, Dipterocarps) are more structured.
• Variation patterns: clinal (gradual, with latitude/altitude - as in *Pinus roxburghii*, *Cedrus deodara*), ecotypic (discontinuous, site-linked), and random. Provenance = geographic origin of a seed lot; land race = an exotic naturalised over generations (Mysore gum, *E. tereticornis* in Karnataka).
• Seed zones / seed source classification: India groups provenances into seed zones so that seed is moved within matched climatic zones; the usual quality ladder is seed-collection area → Seed Production Area (SPA) (a good natural stand thinned and rogued) → Seed Orchard → clonal plantation.
• Gene conservation: *in situ* - Preservation Plots, Gene Pool Conservation Areas, protected areas; *ex situ* - provenance plantings, clone banks, seed banks (orthodox seed), field gene banks for recalcitrant seed (sal, *Hopea*), cryopreservation of pollen/embryos.

Population genetics used in forestry

• Hardy-Weinberg equilibrium: with random mating, no selection/mutation/migration/drift, allele frequencies stay constant and genotypes are p² : 2pq : q².
• Gene diversity (Nei's He) and observed heterozygosity (Ho) measure variation; Fis = 1 − Ho/He shows inbreeding within a population; Fst / Gst measures differentiation among populations.
• Genetic drift is strongest in small populations and is the reason fragmented stands and narrow-based seed orchards lose alleles.
• Inbreeding depression is severe in conifers and many broadleaves - selfed progeny show empty seed, reduced germination and 10-40% height loss; hence isolation of orchards, avoidance of relatives, and minimum clone numbers.
• Effective population size (Ne) - the number of parents that actually contribute genes; unequal sex ratio, unequal flowering and clustered relatives all reduce it below the census number.

Quantitative genetics: variance, heritability, gain

• Variance partition: V_P = V_A + V_D + V_I + V_E (additive, dominance, epistatic, environmental).
• Narrow-sense heritability h² = V_A/V_P predicts response to selection of parents; broad-sense H² = V_G/V_P is what clonal propagation can capture, so H² ≥ h² and clonal forestry can exploit non-additive variance too.
• Typical forest values: height/volume h² ≈ 0.10-0.30 (low), stem straightness 0.2-0.3, wood specific gravity 0.4-0.6, branch angle and leaf traits moderate-high. Low h² for growth is why intensive testing, not eye judgement, is needed.
• GCA (general combining ability) = average performance of a parent's offspring, additive; SCA (specific combining ability) = extra performance of a particular cross, non-additive. Seed orchards exploit GCA; clones and controlled crosses exploit GCA + SCA.
• Genotype × environment interaction decides whether one breeding population serves many sites or separate zone-wise programmes are needed - tested by multi-site trials and type-B genetic correlations.
• Age-age correlation lets early selection (e.g. at 1/4 to 1/3 of rotation) substitute for full-rotation assessment, trading accuracy for a shorter generation interval and higher gain per year.
• Correlated response - selecting for growth can lower wood density in eucalypts and poplars; a selection index (economic weights on several traits) or independent culling levels handles multiple traits.

Selection and testing: the improvement cycle

1. Define objectives and base population - species, site, product (pulp, plywood, timber, fodder, biofuel).
2. Species and provenance trials - replicated, common-garden; eliminate unsuitable origins first.
3. Plus-tree (CPT) selection - *Candidate Plus Trees* scored for volume, straightness, self-pruning, crown/branching, health. Methods: comparison-tree method (regression/check trees: compare the candidate with 4-6 best neighbours within ~30 m), baseline/absolute method in uniform plantations, subjective grading for rough work. A selected CPT must be confirmed by progeny testing; selection on phenotype alone is "mass selection".
4. Multiplication and mating - scions grafted into a clone bank and Vegetative Multiplication Garden (VMG)/clonal multiplication area; controlled pollination for crosses.
5. Progeny testing - open-pollinated (half-sib) or controlled-cross (full-sib) families in replicated designs (RCBD, incomplete blocks, single-tree plots) to estimate GCA, breeding value and heritability.
6. Seed orchards - produce improved seed in quantity:
   • SSO (seedling seed orchard): raised from half-sib/full-sib families; cheap, doubles as a progeny test; rogued after assessment.
   • CSO (clonal seed orchard): grafts/rooted cuttings of tested clones in a non-adjacency/permuted-neighbourhood layout; higher and more predictable gain, risk of clonal incompatibility and flowering asynchrony.
   • Practice: isolation from unimproved stands (commonly 300-500 m or an isolation belt), ≥25-40 unrelated clones to keep Ne high, roguing of poor clones/families, flower stimulation (girdling, fertiliser, paclobutrazol in some species), supplemental mass pollination.
7. Advanced generations - recurrent selection from the best within the best families, breeding-population management (e.g. sublining to control relatedness).
8. Deployment - improved seed, or clonal forestry with tested clones; monitor and feed data back.

Mating designs and breeding methods

• Open pollination/polycross - cheap, gives GCA only; tester (top-cross); full diallel, half diallel, partial diallel - GCA + SCA, costly; NC Design I, II, III; single-pair mating for large advanced-generation populations.
• Intraspecific hybridisation for heterosis; interspecific hybrids widely used - *Eucalyptus camaldulensis × E. tereticornis* and *E. grandis × E. urophylla* clones, *Casuarina* hybrids, *Populus deltoides* clones (G-48, S7C15, Udai, Bahar in WIMCO/Tarai agroforestry), teak clones, *Melia dubia* and *Leucaena* selections.
• Heterosis (hybrid vigour) measured over the mid-parent or better-parent; captured permanently only through vegetative propagation, since F2 segregates.
• Polyploidy and mutation breeding - triploid *Populus*, colchicine-induced polyploids in mulberry; limited field use in forestry.
• Introduction and acclimatisation of exotics (eucalypts, *Casuarina*, *Acacia auriculiformis*, *Gmelina*) followed by local selection to build land races.

Vegetative propagation, biotechnology and markers

• Macropropagation: rooted stem cuttings in mist chambers/root trainers with IBA/NAA, juvenile coppice shoots from VMG hedges, grafting (teak, mango-type cleft/veneer), budding, air-layering, root suckers. Ortet (the original tree) → ramets (its propagules) → clone.
• Micropropagation: axillary bud multiplication (eucalypts, bamboo), somatic embryogenesis (sandal, teak, some conifers), synthetic seed; hardening and field performance testing are mandatory. India's hardened tissue-culture output (bamboo, eucalypts) is certified through DBT's National Certification System for Tissue Culture Raised Plants (NCS-TCP), which also recognises testing laboratories.
• Risks of clonal forestry: narrow genetic base → disease/pest epidemics (e.g. gall insect *Leptocybe invasa* on eucalypt clones, pink disease), site-specific failure, C-effects (carry-over of mother-plant condition) inflating apparent clonal superiority; manage by deploying a clonal mixture/mosaic of many tested clones and continuous screening.
• Molecular tools: isozymes → RAPD/AFLP → SSR (microsatellites) and SNPs for clone identification/fingerprinting, mating-system and gene-flow studies, germplasm structure, seed-orchard pedigree checks; QTL mapping, marker-assisted and genomic selection (promising for long-generation trees); transgenics in *Populus*/*Eucalyptus* studied abroad, with field release in India governed by GEAC rules under the Environment (Protection) Act framework.
• Seed technology links: seed quality testing by ISTA procedures (purity, moisture, germination, 1000-seed weight); most forest species are not notified under the Seeds Act, 1966, so tree-seed certification in India is largely voluntary and run through forest-department seed centres and the National Seed Project rather than as a statutory crop certification.

Indian legal and institutional framework

• Biological Diversity Act, 2002 (as amended in 2023) with NBA/SBB/BMC - access to Indian genetic resources, benefit sharing and export of germplasm; collection and exchange of tree germplasm needs compliance.
• Protection of Plant Varieties and Farmers' Rights Act, 2001 - registration of new/extant/farmers' varieties for notified genera; few forest tree genera are notified, so most tree clones are protected commercially rather than by registration (check the current notified-species list before relying on it).
• Van (Sanrakshan Evam Samvardhan) Adhiniyam, 1980 (the former Forest (Conservation) Act, renamed by the 2023 amendment) governs use of forest land for plantations; National Forest Policy, 1988 set the productivity and quality-planting-material agenda, and agroforestry schemes (Sub-Mission on Agroforestry and successor programmes) fund quality planting material nurseries.
• ICFRE coordinates tree improvement nationally; IFGTB Coimbatore is the lead institute for tree breeding (casuarina, eucalypts, teak, *Melia*), with ICAR's CAFRI Jhansi and NBPGR supporting agroforestry germplasm.

Key points

• Tree improvement - repeated cycles of selection, testing and multiplication to raise plantation value; genetics is permanent and cumulative, silviculture is not.
• P = G + E + GE - replicated, randomised trials are the only way to separate G from E.
• h² = V_A/V_P predicts seed-orchard gain; H² = V_G/V_P is what clones capture.
• ΔG = h² × S = i × h² × σ_P - gain rises with selection intensity, heritability and variability.
• Gain per year = ΔG / generation interval - shortening the cycle (early selection) can beat chasing precision.
• CPT - candidate plus tree, phenotypically superior; becomes a "plus tree" only after progeny testing proves it.
• Comparison-tree method - score a candidate against its best 4-6 neighbours to cancel site effects.
• GCA vs SCA - additive (transmitted through seed) vs non-additive (captured only by cloning/repeat crosses).
• SSO vs CSO - seedling seed orchard is cheap and doubles as a test; clonal seed orchard gives higher, surer gain.
• Roguing - removing inferior clones/families from an orchard; the main post-establishment source of extra gain.
• Ne = 4NmNf/(Nm+Nf) - unequal parental contribution shrinks the effective population and raises inbreeding.
• Inbreeding depression - selfing cuts seed set and growth; isolation, clone numbers and non-adjacency layouts prevent it.
• Provenance - geographic origin; test before import, and match seed zones when moving seed.
• Land race - locally adapted exotic population (Mysore gum) often better than fresh imports.
• Clonal forestry - highest gain and uniformity, lowest genetic buffer; deploy many tested clones as a mosaic.
• C-effects - mother-plant carry-over that makes a clone look better than its genotype; control by common hedging.
• Markers (SSR/SNP) - fingerprinting, gene flow, pedigree checks; QTL and genomic selection for long-generation trees.
• Gene conservation - in situ preservation plots and gene pool areas plus ex situ clone banks and seed banks.

Formulas and examples

Example 1 - heritability and breeding value from an open-pollinated progeny test. A 60-family half-sib trial of *Casuarina equisetifolia* at age 4 gives family variance σ²_f = 0.45 m² and within-family (error) variance σ²_e = 8.55 m² for height.

• V_P = 0.45 + 8.55 = 9.00 m², σ_P = 3.0 m.
• V_A = 4 × 0.45 = 1.80 m².
• h² = 1.80/9.00 = 0.20 (low, typical for height).
• Family A mean = 14.0 m against trial mean 12.0 m → GCA = +2.0 m, BV = 2 × 2.0 = +4.0 m (ignoring shrinkage for small family size).

Example 2 - expected genetic gain and gain per year. 20,000 eucalypt candidates, σ_P = 4.0 m, h² = 0.25; the best 1% are selected (i = 2.67 for p = 0.01).

• Both parents selected (clonal seed orchard of tested clones): ΔG = 2.67 × 0.25 × 4.0 = +2.67 m, i.e. 2.67/16.0 ≈ 16.7% over a 16 m mean.
• Seed collected from selected mothers but pollen unselected: gain = ½ × 2.67 = +1.34 m.
• With a generation interval of 8 years, gain per year = 2.67/8 ≈ 0.33 m/yr; cutting the interval to 6 years by early selection raises it to 0.45 m/yr even if h² falls slightly.

Example 3 - effective population size and inbreeding in a small orchard. A clonal orchard flowers unevenly: effectively 8 clones shed most pollen and 40 clones bear seed.

• Ne = 4 × 8 × 40/(8 + 40) = 1,280/48 = 26.7.
• ΔF = 1/(2 × 26.7) = 0.0187 (1.9% per generation).
• After 3 generations: F = 1 − (1 − 0.0187)³ = 1 − 0.9448 = 0.055 (5.5%) - enough to show measurable inbreeding depression, so flowering synchrony and more pollen clones are needed.

Example 4 - gene diversity and population differentiation. At one SSR locus in a teak seed source, allele frequencies are 0.50, 0.30, 0.20.

• He = 1 − (0.50² + 0.30² + 0.20²) = 1 − (0.25 + 0.09 + 0.04) = 0.62.
• If observed heterozygosity Ho = 0.56, Fis = 1 − 0.56/0.62 = 0.097 - slight inbreeding.
• Across provenances, Ht = 0.62 and mean Hs = 0.56 → Gst = (0.62 − 0.56)/0.62 = 0.097, i.e. about 9.7% of diversity lies among provenances and ~90% within them - the usual outcrossing-tree pattern.

Example 5 - economics of clonal deployment (pulpwood farm forestry). Seedling route MAI = 10 m³/ha/yr; tested clones give a 30% gain → 13 m³/ha/yr. Rotation 7 years, 1,250 plants/ha.

• Yield: seedlings 10 × 7 = 70 m³/ha; clones 13 × 7 = 91 m³/ha; extra = 21 m³/ha.
• At a standing price of ₹2,500/m³ (prices vary by state and year): extra revenue = 21 × 2,500 = ₹52,500/ha.
• Extra planting-stock cost at ₹5 more per clonal plant = 1,250 × 5 = ₹6,250/ha → net gain ≈ ₹46,250/ha per rotation, before discounting.

Where it is used

• Pulp and paper and plywood industry R&D: ITC-Bhadrachalam, JK Paper, West Coast Paper and Mysore Paper Mills run clone-testing networks and supply eucalypt/subabul/casuarina clones to farmers under buy-back arrangements.
• Government research and seed supply: ICFRE institutes (IFGTB, FRI, TFRI, AFRI, IFP) and state Forest Development Corporations maintain seed production areas, CSOs, clone banks and Forest Seed Centres; roles include breeder, tree improvement officer, seed-testing officer.
• Agroforestry and farm forestry: poplar and eucalypt clones in the Tarai, Punjab and Haryana; *Melia dubia* and *Casuarina* in the south; subabul in Andhra Pradesh - clone choice decides farmer returns.
• Commercial tissue-culture and nursery firms: mass production of bamboo, eucalypts, teak and sandal under NCS-TCP recognition; jobs in lab production, hardening and quality control.
• Plantation and TOF (trees outside forests) programmes and CAMPA planting: seed-zone-matched, quality planting material requirements in working plans and plantation guidelines.
• Biotechnology and bioinformatics roles: SSR/SNP fingerprinting of clones for legal and quality-assurance purposes, association genetics and genomic-selection projects at IFGTB and university labs.

Interview questions

• Q: Difference between narrow-sense and broad-sense heritability, and why it matters? A: Narrow-sense is V_A/V_P and predicts gain from sexual propagation (seed orchards); broad-sense is V_G/V_P and includes dominance and epistasis, so it applies to clonal propagation, which can capture the whole genotype.
• Q: What is a candidate plus tree and when does it become a plus tree? A: A phenotypically superior tree selected in the field; it qualifies as a plus tree only after its progeny or clones out-perform controls in a replicated test.
• Q: Why is the comparison-tree method preferred for plus-tree selection in natural stands? A: Comparing a candidate with its best nearby neighbours removes most of the site and age effect, so the comparison reflects genotype rather than micro-environment.
• Q: Give the genetic gain equation and the lever a manager can actually move. A: ΔG = i × h² × σ_P; heritability and variance are properties of the population, so the practical levers are selection intensity (screen more candidates) and a shorter generation interval, which raises gain per year.
• Q: Why are at least 25-40 unrelated clones kept in a clonal seed orchard? A: To keep the effective population size high and inbreeding low, to buffer against pest and disease risk, and to avoid losing alleles needed in later generations.
• Q: What is roguing, and when is it done? A: Removing genetically inferior clones or families from a seed orchard after progeny-test results are available; it raises the genetic quality of seed without any new establishment cost.
• Q: A clone performs brilliantly in the nursery but poorly in the field. Possible reasons? A: C-effects from a vigorous mother plant, genotype × environment interaction, poor rooting-system architecture, or ageing/plagiotropic growth from mature scion material.
• Q: Difference between GCA and SCA, and which one a seed orchard exploits? A: GCA is the average additive contribution of a parent and SCA the specific non-additive interaction of a pair; a seed orchard with open pollination exploits GCA, while controlled crosses or clones exploit both.
• Q: Why does interspecific hybrid vigour in eucalypts have to be deployed clonally? A: Heterosis is largely non-additive and segregates in the next generation, so only vegetative propagation of the F1 keeps it intact.
• Q: What is a seed zone and why move seed within it? A: A region of similar climate and site within which seed can be transferred with little risk of maladaptation; moving seed across zones risks poor survival, frost or drought damage and lost growth.
• Q: Which Indian law governs access to and export of forest tree germplasm? A: The Biological Diversity Act, 2002 as amended in 2023, through the NBA, State Biodiversity Boards and BMCs, with benefit-sharing conditions.
• Q: How are molecular markers actually used in an operational tree-improvement programme? A: For clone identification and fingerprinting (checking mix-ups and commercial identity), estimating outcrossing rate and pollen contamination in orchards, and describing germplasm structure; QTL and genomic selection are still mostly research-stage for Indian trees.

Common mistakes

• Writing ΔG = h²S but using the selected mean instead of the selection *differential* (selected mean minus population mean).
• Forgetting the halving when only mothers are selected (seed from plus trees pollinated by unimproved stands gives half the expected gain).
• Using V_A = σ²_f instead of 4σ²_f for half-sib families, or applying the half-sib coefficient to full-sib data.
• Treating a phenotypically superior tree as genetically superior, and reporting "gain" from un-tested CPTs.
• Claiming clonal forestry gives permanent gain in later generations - the gain is not transmitted through its seed, which segregates.
• Ignoring G × E and recommending one clone or provenance nationwide; and ignoring C-effects when ranking clones from differently-aged mother plants.
• Confusing broad-sense heritability with repeatability of a clone, or quoting growth heritabilities (0.1-0.3) as though they were wood-density values (0.4-0.6).

Related topics

Forest nursery and plantation forestry · Silviculture: principles and practices · Silviculture of important Indian trees · Plant taxonomy · Ecology and ecosystems · Forest seed technology and biotechnology.