Step 1: Understanding the Concept:
Heritability estimation methodologies: intra-sire regression of offspring on dam removes confounding paternal environmental effects and non-random assortative mating biases.
Key Formula or Approach:
\[ h^2 = 2 \cdot b_{\text{Offspring, Dam (Intra-Sire)}} \quad [\text{Free from Maternal Common Environment \& Assortative Mating Bias}] \]
Step 2: Detailed Explanation:
Evaluating heritability ($h^2$) estimation methods in quantitative genetics:
1. Full-Sib Correlation: Inflated by dominance variance ($1/4\; V_D$) and common maternal environmental variance ($V_{Ec}$).
2. Half-Sib Correlation: Requires large paternal family sizes and assumes random mating.
3. Intra-Sire Regression of Offspring on Dam (D): By computing regression within sire groups:
- It completely removes all confounding differences between sires (sire environmental and genetic effects).
- It is completely free from non-random assortative mating biases.
- It eliminates common environmental effects ($V_{Ec}$) that inflate full-sib methods.
- Hence, it is universally recognized as the most statistically robust and reliable procedure for estimating narrow-sense heritability ($h^2 = 2 b_{\text{OD}}$).
Step 3: Final Answer:
Hence, the most reliable procedure is Intra-sire regression of offspring on dam, corresponding to option (D).