Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Optimum Plot Size for Experiments on Sugarcane

M. A. Ghafur
Sugarcane Research and Training Institute, Ishurdi, Pabna, Bangladesh.

S. A. Imam
Sugarcane Research and Training Institute, Ishurdi, Pabna, Bangladesh.

M. Shahjahan
Sugarcane Research and Training Institute, Ishurdi, Pabna, Bangladesh.

Abstract/ Executive Summary:

In 1983. a uniformity trial was carried out at the Sugarcane Research and Training Institute to determine the optimum size and shape of plots for use in sugarcane field experimentation. The results showed that, the coefficient of variation (cv) decreased with: 1. the increase of plot size, 2. a decrease in length of the plot along the row and at the same time. an increase in length of the plot across the row. 3. rectangular plots compared with square ones, and 4. square and nearly square blocks compared with rectangular ones. The coefficient of variation was found to be minimum for 38.4 sqm. to 115.2 sqm. sizes of plots but the optimum plot size was found to be 96 sqm. (12m x 8m). Coefficient of variation was found maximum for smaller plots compared to optimum (96 sqm) sized plots, In an experiment, smaller plots require a large number of replications compared to optimum ones for the same precision.

Keywords: Optimum plot size, Number of replications, Statistically precise results, Sugarcane field experiments

Location: Sugarcane Research and Training Institute farm

Start Date:

End Date:

Program Area: Knowledge Management

Commodity: Sugarcane

Objectives:

To study an optimum plot size, and number of replications required per treatment to obtain statistically precise results, in all sugarcane field experiments

Methodology:

A uniformly grown sugarcane plot having Co. 1158 variety planted in 1982 crop season was taken for the experimental purpose at the Sugarcane Research and Training Institute farm. A net area of 0.432 sqm. standing cane from about one hectare crop was taken for the experiment. All recommended cane production practices were followed throughout the growing period of the crop. The crop was planted mechanically with row to row spacing of 1.20 m. The net area was composed of 60 m in length along the rows and 72 m. across the rows. Each 60m long row was divided in 30 equal sections each 2m. in length. These sections were then harvested by sickles and weights were recorded separately. Finally, the yield records of 1800 (30x60) ultimate units were obtained. The yields of adjacent plots were then combined to estimate the yields of different plot sizes and shapes. The data were examined statistically for the study of: 1. the coefficient of variation (cv) of plot and block sizes and shapes 2. the effective number of replications and the total area required per treatment for specified accuracy 3. optimum plot size. The coefficients of variation were calculated by the formula s/x. x 100 where "s" is the x standard deviation and "x" the mean of the plot yield. The efficiencies of the blocks were obtained by the ratio of error variance before and after eliminating the differences due to blocks. If the ratio is greater than one, blocking is efficient. As the ratio decreases towards one, the advantage due to blocking decreases (Kempthorne, 1973). For the estimation of minimum number of replications, the table 2.1 of Cochran and Cox (1964) were used to detect a mean difference as small as 20%; of the experimental mean for 90% probability of obtaining a significant result at 5% level. The minimum area required per treatment were then obtained by the product of the number of replications and the corresponding plot size. The optimum plot size was deter- mined by studying the variations of different sizes of plots for the same size and number of blocks.

Source of Information: Bangladesh J. Sugarcane, 6: 1-7 July, 1984

Article Link (Online Journal):

Funding Source:

Total Budget (Tk.) :

Achievement/Findings:

It revealed that for smaller plots, a large number of replications were required for the same size of the block. As the plot size was increased from 2.4 sq.m. to 48.0 sq.m. the number of replications decreased from 22 to 5 for ten plots block. The experimental plot size should be optimum or around the optimum for efficient use of the plot and block.

Knowledge Management: Journal