This experiment was carried out to ascertain the hypothesis that the total yield of a crop community will descend if the agricultural materials such as fertilizer are unevenly distributed such as the case an over and under supply of resources. An attempt to observe the relationships between the rice yield and rice plant communities with variant production structures which were caused by uneven fertilizer application in 1991 and 1992 in Tsukuba. The degrees of variation of crop growth, measured by variance, mainly occurred on the growth area closed to the rows where fertilizer had been applied. The total yield of the rice plant communities barely decreased in 1991, while, increased in 1992 with the variance. It was difficult to find a clear relationship between the yield of rice plant communities and uneven growth, with the variance. One of the factors affecting the higher productivity found in the rice plant community with the large growth variation might be the acceleration of the leaning effect of co-existence of easily-lodging among hard-lodging hills. It was estimated that the promoted yield in a rice plant community was not in proportion to the number of rice plants. In addition, we showed that there might be a correlation between the yield in the rice plant communities with variant production structures and the distribution of nitrogen density as an assumption of meteorological potentiality for productivity. The stratified random sampling used in these experiments was recognized as a proper method for the pattern found in a rice plant community.

View full abstract

Recently, identification (discriminant) models based on neural network are being introduced in several agricultural researches. In this study, we discussed the method to select the supervisor data set from whole data to perceptron neural network for the evaluation of soybean plant shape. Because a perceptron neural network is trained to respond to only a supervisor data set, the general identification (discriminant) efficiency of the neural network for non-supervisor data strongly depends on the method to select the supervisor data. Though the method to select the supervisor data takes one of the most important roles to develop neural network models, it has not been fully established especially in such a case as soybean plant shape evaluation where the features (variables) to identify the shape are not clear and the distributions of those features are hardly known. In this study, we applied neural network model to evaluation of soybean plant shape for the substitution of human visual judgments and examined several supervisor data sets to find the method to select the most effective supervisor data set. The results of the examinations indicate two strategies to select the supervisor data set. The first that the distribution of target output in the supervisor data set is not biased and the second is that the supervisor data set contains the largest and smallest values for each component of input vectors. The neural network trained to the supervisor data set selected based on those strategies showed the identification (discriminant) efficiency 20% higher than that given by the neural network trained to ordinarily selected data sets.

View full abstract

Recently, Internet is growing fast. The scientist put the net to use for electric mail, file transfer, network news, sending out the results of study, etc. It becomes indispensable tool for the academic research field. The computer center for AFFR (Agriculture, Forestry and Fisheries Research) investigated the academic research networks in Europe, South-East Asia and Australia. This report is the state of the academic research networks in South-East Asia and Australia in 1995. Generally speaking, Asian countries are behind Western countries in the field of the academic research network. But those countries are rapidly constructing infrastructure under recognizing important basic social works. Research cooperation between Japan and those countries is expected. Especially the mutual exchange of agricultural research and technical information is involved. Then it will be very important infrastructure for an international joint research project, etc. that the computer network for ministry of agriculture, forestry and fisheries is connected with those countries by very fast line.

View full abstract