RU1796100C - Method of evaluating farm crop world collection samples by production collection per unit of area - Google Patents

Abstract

Usage: agriculture, breeding to increase the collection of products per unit area. SUMMARY OF THE INVENTION: Samples of a world collection of agricultural crops are sown as a collection nursery, and a number of morphological characters are measured. For each of these characteristics, the ecological (VCe,%) and additive (VCg,%) coefficient of variation are determined. Then, a search is made for marker features having low environmental and high additive coefficients with close correlation with the collection of products per unit area. In this case, the samples with the maximum severity of the marker feature provide a high collection of products per unit area. 12 tab.

Description

The invention relates to agriculture and can be used in breeding crops to increase yield per unit area, for example, grain yield.

There is a method for assessing the seed productivity of agricultural plants in a collection nursery in repetitive sowing. To do this, the mass of seeds is divided by the number of harvested plants, the productivity of the plants of the test sample is compared with the productivity of the nearest as well as the average standard. Excess is considered reliable only when the sample is 10-15% higher than the nearest and average standard.

The disadvantage of this method is the impossibility of obtaining, when using it, significant differences between the studied samples of the world cultural collection

plants by grain yield or the collection of other products, such as protein.

When checking this method by processing two-year data on the study of the world soybean collection, there were no significant differences between the samples by grain yield - Pf8i. Rteor. (see table 1).

The purpose of the invention is to increase the efficiency of selection of agricultural crops by isolating in the early stages of selection, i.e. in a collection nursery (KP), donors and sources with genetically determined high yields per unit area.

The goal is achieved in that when sowing samples of the world collection of agricultural crops by the type of collection nursery, the computer searches for market signs or indices of high yield per unit

Xi

YU

CS

Oh Oh

areas with low environmental (VCe-%) and high additive (VCg,%) coefficient of variation; close correlation with the collection: products of the area unit; reliable estimates of the magnitude of the marker trait or index in the studied samples in the process of analysis of variance

(Rfact. Rteor.).

The minimum values of the ecological variation (VCe) of the selected trait or index make it possible to use it in a wide range of weather and soil-climatic conditions of various breeding centers.

The maximum values of additive variation (VCa) indicate that a large proportion of the phenotypic variation of a trait or index is determined by the genotype and is significantly less susceptible to modification variation.

EXAMPLE 1. Field experiments with soybeans were carried out in 1987 - 1988 in Ussuriysk at the Primorsky Research Institute of Agriculture, and the obtained data were processed on an Electronics 100/25 computer in AIS (Leningrad).

Soybean samples (110 pcs.) From the world collection were grown as a collection nursery, i.e. 80 plants with an area of 45 x 5 cm in repeated crops were not divided. 20 plants from each crop were taken for analysis. In separately harvested plants, quantitative traits were measured (20 in total) and all simple indices were calculated (the ratio of one trait to another).

The ecological coefficient of variation (VCQ,%) of traits and indices was determined by the variation of their values among plants within the samples, and the additive (VCg,%) was determined by varying the average between samples of the soybean collection.

The genetic correlation coefficients (rg) between traits and indices on the one hand and soybean yield on the other hand were calculated using 18 samples with contrasting indices for the productivity of divisions and with a minimal variation over the years.

Ecological stability was judged by the values of VCe and VCg, the lower the inter-variety and the higher the inter-variety variability, the more stable the sign or index.

From Table 2 it is clearly seen that direct selection for grain yield per unit area by weight and number of seeds per plant is not possible due to the high environmental variability of traits (55 and 60%, respectively), i.e. the size of the signs to a large extent

determined by environmental conditions.,

The maximal genetic and minimal values of the environmental coefficient of variation were met to the maximum extent by the indices (Table 2): seed weight / plant weight, seed weight / number of fertile nodes and seed weight / number of beans.

A study of the genetic correlation dependencies between yield and indices having a satisfactory ratio of environmental and genotypic variants showed different

yield per unit area. The closest relationship is observed in indices having in their structure such characteristics as the mass and number of seeds from the plant.

However, on the studied indices, only the harvesting index and the seed mass / number of fertile nodes index have the optimal ratio of all three parameters, i.e. VCg exceeds VCe, and rg 0.7.

when studying the harvesting index by the method of analysis of variance on the entire collection of the axis axis, it turned out that Rfact. 3.50 FTeop. 3.92 (Table 3). This suggests that all the variability of this index lies in

limits of random error, i.e. within the limits of error of experience and index it is impossible to use as a marker.

From table 4 it is seen that the variability of the other selected index significantly exceeds the variability of the random error and it can be used as a marker of productivity per unit area.

Table 5 shows the indices — seed weight / number of fertile nodes (LSF), which best meets the above requirements compared to the remaining 231 indices and 22 characters. For comparison, the characteristics of the trait are given - plant productivity (direct selection). Then, the found index was used for selection in a collection nursery (Table 6).

From the above data it can be seen that the marker index for selection found by the proposed method effectively works in the breeding program of one of the leading breeding centers in the country for self-assessment.

Example 2. Field experiments with peas were carried out in 1980 - 1982. in the city of Krasnograd and in the city of Omsk, and the data obtained were processed on a Nairi-C computer at the VNIIZBK (Orel).

74 pea samples from the world collection were grown as a collection nursery, i.e. in a plot of 80 plants with a feed area of 5 x 60 cm in repeatless crops. Plant analysis and subsequent calculations were carried out similarly to soy experiments.

A direct assessment of pea productivity by the method of analysis of variance showed an unreliable difference between the studied samples in both Krasnograd (Table 7) (Rfact. 0.215 is significantly less than Rteor. 1,000) and Omsk (Table 8).

The variability of the selected index, the mass of seeds / number of productive nodes, significantly exceeds the variability of the random error and can be used as a marker of productivity per unit area (Tables 9, 10),

PRI me R 3. Field studies were carried out in 1983 - 1987. in the Lithuanian Research Institute of Agriculture (Dotnuva).

As a material for research, 231 forage bean samples from the Worldwide VIR Collection were used.

Samples were sown in the field without repetition in divisions 1.5 × 2 m in size with a density of one hundred and 40 plants per 1 m2. Ausra standard variety was seeded through ten numbers. Each year, 10 plants were removed from each plot, for which 18 quantitative traits were measured and all simple indices were calculated.

The experimentally obtained data were processed by analysis of variance and

standard method (x, o2, vc).

All calculations were carried out at the LITNIIZ computer center on the SM - 4 computer.

Ecological and genotypic coefficients of variation, as well as the genotypic correlation coefficient, were calculated similarly to the soybean example.

; From the presented data it is seen (Table 11) that direct selection for yield by characteristics: weight and number of seeds and number of beans, is not effective due to the large influence of environmental conditions. Ecological and genotypic coefficients of variation are almost equal,

Thus, according to the parameters corresponding to the first restriction, the following features and indices were distinguished: plant height, stem thickness, seed weight / seed number, seed weight / number of beans, seed weight / number of productive nodes, stem thickness / plant height

The second limitation is relatively

close Positive correlation with yield per unit area is a prerequisite for the effective evaluation of samples by yield. Of the above characteristics and indices, this requirement is most met by: plant height, stem thickness, seed weight / number of seeds, seed weight / number of beans, stem thickness / stem length.

Of these, the stem length is unlikely to be used as a marker for assessment, since its increase will lead to a prolongation of the growing season, due to the correlation between

the height of the plant and the duration of the growing season (g 0.56).

Thus, as markers of seed yield in fodder beans in collection and breeding nurseries

- only the thickness of the stem can serve reliably in the characters, and from the indices: the mass of seeds n / number of beans, the mass of seeds n / number of seeds and the thickness of the stem / length of the stem.

Analysis of variance of 80 samples from

of the studied collection over three years according to the distinguished features and indices, as well as productivity (control) showed that the most reliable genetic differences in the samples were obtained by the mass index

sem n / the number of beans in which Rfact. exceeded rteor. 12.5 times (tab. 12). A direct assessment of plant productivity for three years of study does not give significant differences between the samples (Rfact. FTeop.), Which is confirmed by the data in Table 11, and therefore productivity cannot be used for mass assessment of the collection of fodder beans, except for extreme options between which the differences exceed

x ± per.

Formula, finding

The method of evaluating samples of the world collection of agricultural crops for the collection of products per unit area, including the sowing of agricultural samples, determine the environmental coefficient

.. of crops with a feeding area close to 55 production, characterized in that, in order to increase the efficiency of the method, a number of morphological characters are measured in the samples, for each character

variations (VCe,%) and additive coefficient of variation (VCg,%), find marker features that have a low environmental and high additive coefficient

Dispersion analysis of the productivity of collection soybean samples (Note NIIISH, 1987 - 1988)

The value of the environmental (VC «,%) additive (VCg,%) correlation coefficient with yield (fa) for some signs and indices of soybean (Note NIISH 1987-1988); /

Analysis of variance of the index-mass of seed n / mass of soy plants (Note. NIISH, 1987 - 1988)

with close correlation with the collection of products per unit area, and samples are evaluated according to the maximum severity of the marker trait, Table 1

Table 2

Table3

Analysis of variance of the index mass of seed n / number of soybean fertile nodes (Note. NIISH, 1987-1998)

Index characteristic mass of seed n / number of fertile

nodes in soybeans that most satisfy the constraint

for a marker of high productivity (Note NIIISH, 1987-1988)

The value of the marker index and the yield of the released soybean samples (Note NIGISKh, 1987-1988)

Table 4

Tableb

Tableb

Variance analysis of collection pea samples (Kraenograd, 1980 - 1982,).

Dispersion analysis of the productivity of collection pea samples (Omsk, 1980-1982)

Analysis of variance mass of seed n / number of productive nodes of peas (Kraenograd. 1980 - 19.82)

Analysis of variance of the index mass of seed n / number of productive nodes (Omsk, 1980 - 1982)

Table

Tableb

Tableb

Table 10

Coefficients of genetic L / Cg,%), environmental (VCe,%) variation and

genetic correlation (rg) with yield per unit area of characters

and indices of samples of fodder beans (LitNIIZ, 1983 - 1987).

The list of signs and indices in fodder beans, to the greatest extent

performing restrictions on seed yield markers

(LitNIIZ, 1983-1986)

For the correlation coefficients, the comma and zero are reduced.

Table 11

Table 12