Classification of Inbreds for Their Value in Improving Elite Maize Hybrids

Zanoni, Urs

Permalink

https://hdl.handle.net/2142/71650 Copy

Description

Title

Classification of Inbreds for Their Value in Improving Elite Maize Hybrids

Author(s)

Zanoni, Urs

Issue Date

1988

Doctoral Committee Chair(s)

Dudley, John W.

Department of Study

Agronomy

Discipline

Agronomy

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

Agriculture, Agronomy

Abstract

• The identification of inbred lines (I$\sb{\rm w}$) as sources of favorable alleles not already present in an elite hybrid (I$\sb1$ x I$\sb2$) is important. A non-reciprocal diallel of 4 maize (Zea mays L.) inbreds of Stiff Stalk Synthetic, 5 of Lancaster, and 5 of unrelated origin were evaluated for grain yield, grain dry matter, earliness of flowering, total lodging, and plant and ear height at two locations in 1984 and 1985. Inbreds B73 and Mo17 crossed to the remaining 12 inbreds, B73 x Mo17, their F$\sb2$'s, and the inbreds per se were evaluated at two locations in 1986. For the hybrids B73 x Mo17, B73 x B77, and B75 x N152 donor lines (I$\sb{\rm w}$) with low and high estimates of the relative number of favorable dominant alleles ($\mu$G) not present in the elite hybrid (I$\sb1$ x I$\sb2)$ were selected and random testcrosses from their F$\sb2$ generation were evaluated at two locations in 1986.
• Using inbred and F$\sb1$ data, the modified estimator of favorable dominant alleles $\mu$G* reliably distinguished donor lines with low and high estimates of favorable alleles. Ranking of lines with similar estimates was less reliable. The $\mu$G* estimator was highly correlated with predicted threeway cross (PTC) and minimum upper bound (UBND) for most traits and with general combining ability (GCA) for more additive traits. Correlations of $\mu$G* with $\mu$G and inbred performance per se were generally low. For traits with little heterosis, such as dry matter, $\mu$G* could often not be estimated. Different combinations of inbred and F$\sb1$, inbred and F$\sb2$, or F$\sb1$ and F$\sb2$ data could be used to estimate $\mu$G*, but may result in different precision. Estimates of relative genetic distance of I$\sb{\rm w}$ to I$\sb1$ or I$\sb2$ agreed well with pedigree information. F$\sb2$ testcross populations developed from inbreds with large $\mu$G* estimates had on the average more testcrosses genetically superior to the check mean than inbreds with low estimates. Changes of the F$\sb2$ population mean compared to the check mean and the expected variability among testcrosses could be predicted. These predictions were reliable for dominant and recessive traits.

Graduation Semester

1988

Type of Resource

text

Permalink

http://hdl.handle.net/2142/71650

Owning Collections