Journal of Taiwan Agricultural Research

Author:Hui-Chin Wu, Lit-Fu Chan, Meng-Li Wei, and Hsiu-Ying Lu*

Abstract:

    Leaf area is a valuable index in identifying muskmelon (Cucumis melo L.) growth and development.  A simple technique was developed for estimating leaf surface area (A) of musk- melon with measurements of leaf length from the sinus base to the apex of leaf along midrib (L) and/or maximum leaf width (W) without the use of any expensive instruments.  Leaves for three muskmelon varieties, Chioumih, Zhufen and Tianhun were collected from different levels of the canopy of plants during the growing season in 2007 at Taiwan Agricultural Research Institute, Wufeng, Taichung, Taiwan.  Each leaf was measured for L, W and A.  Data analysis were per- formed by regression technique.  Results showed that differences in leaf shape among these varieties affected the performances of prediction accuracy in different models for leaf area estimation.  The linear equations having W² as the independent variable provided the most accurate estimation of A for Chioumih and Zhufen muskmelons, with leaf area coefficients of 0.70 and 0.62, respectively.  For Tianhun muskmelon, the equation of A = 0.67×L×W exhibited the highest accuracy of leaf area estimation, and the equations of A = 0.62×L² and A = 0.71×W² could also be adopted.  With these models, a high degree of 1:1 relationship was obtained between measured and predicted leaf areas, and they produced satisfactory results of lower mean squared deviation (in the range of 451 to 762), lower percentage of deviation for total leaf area (≦1.34%), and lower mean percentage of deviation for individual leaf area (≦0.25%).  The linear regression model was simple, rapid, inexpensive and accurate for estimation of leaf surface area of musk- melon and nondestructive to plant growth.

Key words:Surface leaf area, Nondestructive estimation, Linear regression model, Leaf shape, Prediction accuracy

Download: PDF Links