Search

High rice (Oryza sativa L.) yields are closely related to plant absorption of a large amount of nitrogen (N). However, there is little information on the fate of N applied at the middle growth stages of rice. Labeled 15N ammonium sulfate was applied at the panicle formation stage in Experiment I, and 10 d after heading in Experiment II. Zeolite was also added at the concentration of 0, 0.01, and 0.1 kg kg-1 to increase the cation exchange capacity (CEC) of the soil. The amount of 15N fertilizer in the soil surface water decreased exponentially and the fertilizer disappeared within 2 d after application. The soil that received zeolite at 0.1 kg kg-1 exhibited significantly less 15NH4 +-N in the surface water and in the soil solution than the soil without the zeolite amendment. A significantly larger amount of exchangeable 15NH4 +-N was observed in the high zeolite-treatment of soil compared to the low zeolite-treatment of soil. The amount of exchangeable 15NH4 +-N increased initially, and thereaft...

The kinetics of mineralization of soil organic nitrogen (N) in samples of several air-dried and fresh soils in the tropics was studied. Four different paddy soil samples in the IRRI lowland field were used. Parameters for mineralization of soil organic N were computed by using Arrhenius' law and Michaelis-Menten's equation. The following results were obtained: 1. The mineralization pattern of organic N in the wet soil samples were represented by an exponential equation as a function of reference days, which were determined by the duration of the incubation period and apparent activation energy. Although significant differences in the mineralization rate and mineralization potential among the soils were observed, there was no appreciable difference in the value of apparent activation energy, which were relatively low compared with those of Japanese paddy soils. 2. The mineralization pattern of the dried soil samples was represented by an equation consisting of two exponential equations. When a lar...

The study has been carried out to evaluate varietal difference in leaf senescence during the ripening period in relation to sink size and nitrogen application. Leaf senescence was early in the medium-duration lines. Among the short-duration types, there were both early and late aging lines. The medium-duration lines had larger leaf area and higher sink sizes at the anthesis than short-duration types. The pattern of leaf area reduction during the ripening period in the short duration line, with relatively higher sink size, was almost similar with that in the medium-duration lines. However, in the short-duration lines with relatively small sink size, no significant leaf area reduction was observed until two weeks after the anthesis, and then the leaf area began to decrease rapidly in these lines. Decrease in leaf area during the ripening period varied with the rate of basal nitrogen (N). Higher basal N increased the rate of leaf area reduction by increasing the sink size. On the other hand, N-topdressing at the anthesis decreased the rate of leaf area reduction during the ripening period. Likewise, the reduction in sink size as imposed by removal of spikelets decreased the rate of leaf area reduction during the ripening period. Significant correlation coefficients between the sink size and the amount of N remobilized from leaves to panicles, and between the decrease in leaf area and the amount of N remobilized from leaves to panicles were observed. Therefore, the decrease in the leaf area during the ripening period is mainly governed by the amount of N remobilized from leaves to panicles, which is affected by the N requirement of grains for ripening and the amount of N absorbed during this period.

Field experiments on the response of rice plants to slow release fertilizer (Meister 10, MS10) were conducted from the 1986 wet season (WS) to the 1987 WS at IRRI, with ammonium sulfate (AS) as control. IR64 and IR36892-163-1-2-2-1 were planted under two N levels and three spacings. In all experimental plots regardless of season, the amount of N in the plants increased exponentially at the early growth stage (y = abx) and linearly at the middle and late growth stages (y = a + bx), as previously reported. MS10 lengthened the exponential phase and increased parameter 'b' during the linear phase. The amount of N in the plants was higher in AS than in MS10 plots up to 5 weeks after transplanting. Comparing narrow and wide spacing, the difference in the amount of N in the plants was greater in the MS10 plots. At flowering and maturity, the amount of N in the plants was higher in the MS10 plots due to the higher percentage recovery of basal N and the higher rate of N absorption during the linear phase. The amount of N in the plants grown under MS10 was almost equivalent to those in the plants grown under a double dose of basal AS. Tillering pattern and achievement of the maximum tiller number stage varied with spacing and fertilizer types, as reflected in the N absorption pattern of the plants. Yield, sink size and potential sink size of the plants reflected the N absorption pattern and the amount of N in the plants. The yield of 6g N-MS10 was almost equivalent to the yield of 12g N-AS. The yield of MS10 plots was especially higher, with narrow spacing due to N absorption during the early growth stage.

The behavior or basal nitrogen and absorption of nitrogen by the rice plants in two different paddy fields, Odawara and Takadate, were compared. The results obtained are as follows; 1) The soils of both fields belonging to Gray-Brown Soil were poor ill soil oraganic matter, and were poorly drained. However, the clay contents and CEC values of the plowed layers were much greater in Takadate than in Odawara. Furthermore, the clay fraction of Takadate soil was dominated by montmorillonites. On the other hand, the amounts of nitrogen mineralized by incubation at 30°C were treater in Odawara soil than in Takadate soil. 2) The amount of soil ammonium nitrogen in both fields decreased exponentially after transplanting, and almost all of it was disappeared at the accumulated effective thermal index (AETI) of about 400. The decrease of soil ammonium nitrogen in both fields was shown as an exponential equation of AETI. Soil ammonium nitrogen derived from basal nitrogen was disappeared about the same time. 3) The amounts of nitrogen per unit area absorbed by the rice plants (y) in both fields were indicated to be closely related to the AETI (x). An exponential equation (y=cdx) was obtained for the early growth stage, and linear one (y=ax+b), for the middle and late growth stages. The crossing point of these equations was the AETI of about 400. However, there were some differences in the parametcrs between Odawara and Takadate. It was assumed that the limiting factors were the ability of nitrogen absorption of the rice plants, for the exponential part, and the rate of mineralization of soil organic nitrogen, for the linear part. 4) The average recoveries of basal nitrogen (60-70 kg-N/ha applied) in the rice plants were 43% in Takada, and 27% in Odawara, reflecting the ammonium absorption of the plowed layers as mentioned in 1). On the other hand, the amounts of soil nitrogen absorbed by the rice plants at the ear mergence were 71 kg-N/ha in Odawara, and 44 kg-N/ha in Takadate, reflecting the amounts of nitroten mineralization by incubation. 5) The amounts of residual basal nitrogen in plowed layers were 16% in Takadate, and 19% in Odawara. Therefore, the total recoveries (the recovery of basal nitrogen in the rice plants and residual basal nitrogen in the plowed layers) were 68% in Takadate and 46% in Odawara. Since the water permeability of both fields was extremely small, it was assumed that almost all of unrecovered basal nitrogen was lost by denitrification.

Field experiments were carried out in order to study the relationships between the nitrogen absorption by rice plants grown under various cultural conditions and the growth of the plants. The results obtained are as follows; 1) It was found that the rate of nitrogen absorption by the rice plants was the greatest just before the neck node differentiation stage throughout all the growth stages. The greatest rate of nitrogen absorption by the plants at this stage was positively correlated to the length of culms, the sum of lengths of lower three internodes, the sum of lengths of upper three leaf blades (excepting those of the plots of nitrogen top-dressing), and the number of spikelets per panicle (excepting those of the plots of no basal nitrogen, nitrogen top-dressing at the neck node differentiation stage and at the stage of reduction division), respectively. 2) The amount of plant nitrogen per unit area at the early stage of spikelet differentiation was positively correlated to the number of panicles per unit area. Furthermore, the amount of nitrogen in plant per unit area at the ear mergence was positively correlated to the number of spikelets per unit area. 3) It was previously shown by Matsushima et al (3, 4, 5, 6) that lodging resistance and light utilizing efficiency of rice plants were improved by the decrcase of the length of culms, the sum of the lengths of lower three internodes, the sum of the lengths of upper three leaf blades and the number of spikelets p6r panicle and by the increase of number of panicles and spikelets per unit area. In order to obtain the growth of rice plants mentioned above, the rate of nitrogen absorption by rice plants must be decreased just before neck node differentiation stage and be increased at the other growth stages. It was recognized that early transplanting and high plant density was effective means to permit the desirable rate of nitrogen absorption by rice plants and to increase the rice yield per unit area.

The influence of a thermal factor on the soil ammonium nitrogen and its absorption by rice plants grown under the field and controlled conditions were studied. It was shown that the absorption of nitrogen by the rice plants was closely related to the accumulated effective thermal index (AETI) given by T. Hanyu and T. Uchijima (1962). The amount of nitrogen per unit area absorbed by the plants (y) grown in the field was closely related to the AETI consisting of an exponential equation (y=cdx) for the early growth stage and linear one (y=ax+b) for the middle and late growth stages. The crossing point of these equations was the AETI of about 400. The similar equations excepting the b parameter were obtained for both low and high dosages of basal nitrogen. But the crossing point of these equation was varied by the cultural conditions, especially plant density and placement of fertilizer. Soil ammonium nitrogen decreased exponentially after transplanting. Almost all of the nitrogen disappeared at the AETI of about 400, or at the commencement of ear-primodia formation. Therefore, it was assumed that the rate of nitrogen absorption of the rice plants was the limiting factor in the exponential part. On the other hand, the rate of supply of soil ammonium nitrogen, or the rate of mineralization of soil organic nitrogen was assumed to be the limiting in the linear part.

The amount of N in plants increased exponentially at the early growth stage (y=abx) and linearly at the middle and late growth stages (y=a+bx). The crossing point of the two equations coincided with the maximum tiller number stage. Narrow spacing shortened the exponential phase. On the other hand, wide spacing and slow release fertilizer lengthened the exponential phase. Parameter 'b' at the linear phase was not affected by spacing and rate of basal fertilizer. However, it varied with the kind of basal N and with soil chemical characteristics. Sink size and potential sink size were correlated with the amount of N in the plant at the late spikelet initiation and flowering stages. However, contribution of N in the plant to sink formation varied with spacing and growth duration of varieties. This can be explained by the duration of the vegetative lag phase of the plant.

After classifying the rice plants which grown under several cultural conditions in last few years into the several groups according to the number of spikelets per unit area, the authors studied the percentage of ripened grains from an analytical investigation of the number of spikelets per unit area, obtaining the following results. 1. When the number of spikelets per unit area was extremely small, no definite relation was found between the percentage of ripened grains and any component of the number of spikelets per unit area (i.e. the number of panicles per unit area, that of spikelets per panicle and that of primary and secondary rachis-branches per panicle). 2. The percentage of ripened grains showed a tendency to decrease progressively with an increase in the number of spikelets per unit area, spikelets per panicle and secondary rachis-branches per panicle, while it seemed to increase with an increase in the number of panicles per unit area. 3. When the number of spikelets per unit area was extremely large, the percentage of ripened grains was positively correlated with the number of panicles per unit area and negatively correlated with the number of spikelets per panicle as well as the number of secondary rachis-branches. 4. In case of the number of spikelets per unit area being large, the fact that the percentage of ripened grains decreased with an increase of the number of spikelets per panicle could be explained from the fact that the number of spikelets per panicle increased with a decrease of the amount of carbohydrates stored in sheaths and culms before heading and that of photosynthetic products after heading.

In the previous papers (Report nos. 74, 69, 67), Matsushima et al. stressed the necessity of the ideal plant types for maximizing grain-yields of rice. In the present paper the authors further studied a method no obtain high yields by making the plant into ideal plant types. Two conditions are subjected to making the ideal plant type, i.e. (1) ensuring the number of spikelets per unit area corresponding to a target yield by the time of 69 leaf-number index (which is the time just before the neck-node differentiation and roughly corresponding to 43 days before heading) and (2) restricting nitrogen supply during the period from 69 to 93 in the leaf-number index (the period is approximately corresponding to the duration from 43 days to 20 days before heading) in which the plant type is definitely predetermined. For ensuring the number of spikelets correspondig to high yields, much ammonium nitrogen must usually be supplied, but when much ammonium nitrogen has been used in a paddy field, the control of nitrogen absorption by the plant during the critical period in which the plant type is definitely predetermined is hardly possible on account of the ammonium nitrogen having been adsorbed by the soil. On the othr hand, when nitrate nitrogen is applied to a paddy field, it is by no means adsorbed by the soil, but it is always dissolved in irrigation water so far as the irrigation water remains in the field. In this occasion, if one wants to stop the supply of nitrogen, one can easily do so at any time by changing the irrigation water which remains in the paddy field for new water which contains no nitrogen. Getting a hint from a circulating irrigation method in which identical water is repeatedly irrigated by a pump and which has long been practised economically in Toyama Prefecture, the authors devised a method by which one can easily take off the nitrogen from a paddy field at any time as follows. Putting much nitrate nitrogen into water in a paddy field and making water circulate by pumping up ditch-water which comes out through underdrains of the paddy field, the authors could always supply much nitrate nitrogen to rice plants for encouraging their tillering with little leaching of nitrogen (some amount of nitrate nitrogen is, however, lost by denitrification). By changing the irrigation water for new water at the time of 69 leaf-number index (approximately 43 days before heading), they succeeded in stopping the supply of nitrogen to the plant and thereafter they irrigated the plant with nitrogen-free water until the time of 93 leaf-number index (approximately 20 days before heading). At the time of 93 leaf-number index and again at full heading time they top-dressed with ammonium nitrate to encourage the carbon assimilation of the plant. By this method they succeeded in, (1) obtaining the necessary number of tillers per unit area corresponding to a target yield by the time of 69 leaf-number index, and (2) decreasing the nitrogen content of the plant, paling of leaf-colour, decreasing the dry weight increment, of the plant at the critical period, decreasing the length of leaves, culms and internodes, and as a result, (3) raising the grain-yield by 5-27% as compared with the control plot respectively and obtaining 7.1 tons of brown rice per hectare as the highest yield which has never been obtained by any other means so far in their experimental paddy fields. By the way, the amount of water consumed for evapo-transpiration in the circulating irrigation method was less by 14-23% than the ordinary irrigation (flooded) method, and a tentative proposition was made on the application of nitrogen for ensuring the number of spikelets per unit area to maximize the yield of rice.

Along the line previously reported (67, 69), the authors further studied the principles for maxmizing yield of rice and tried to demonstrate them, obtaining the following results. 1) A target number of spikelets per unit area could easily be obtained in most treatments by increasing the number of panicles, which was brought about by making the plant produce as many tillers as possible at the early growth stages. And a pretty high percentage of ripened grains could also be obtained in most treatments by decreasing the number of spikelets per panicle, controlling excessive growth and improving plant form, which were easily achieved by decreasing or stopping nitrogen supply during the critical period in which the plant form after heading is definitely predetermined. As a result, a yield of 91.9kg of brown rice per are was obtained as the maximum in the present experiment, and 8 treatments yielded more than 75kg per are. 2) A strong positive correlation was found between the number of spikelets per unit area and the amount of nitrogen absorbed by the stage of spikelet differentiation or heading time. From the correlation one can easily estimate the necessary amount of nitrogen to be supplied for producing a target number of spikelets. 3) Decreasing the nitrogen supply in the critical period, in which the plant form after heading is predetermined, made the plant proportionally decrease the amount of phosphate, potassium and silica absorbed, resulting in a decrease in the number of spikelets on the other hand, however, made the plant increase the amount of carbohydrate, resulting in favouring ripening and preventing the plant from lodging. The degree of decreasing the nitrogen supply in the critical period, therefore, must be determined by taking the number of spikelets per unit area, eventual percentage of ripened grains and the resistance of lodging of the plant into due consideration. 4) The optimum leaf-area index of the rice plant at heading stage had hitherto been reported to be about 5 to 6 under the light intnsity of near 320 cal. per day per cm2, while in the present experiment it had likely proved to be about 7 to 8 for producing a maximum dry matter after heading. The leaf-area index, therefore, seems not to be constant even under a given light intensity, but to be considerably variable according to the other factors, i. e. the form and healthiness of the plant etc. 5) From the analytical investigation of high yields in the experiment it had been clarified that the percentage of the amount of carbohydrate accumulated in straw before heading to the final grain yield was always less than 20. Accordingly, the importance of the amount of carbohydrate assimilated after heading can not be overestimated for maximizing yield. For increasing the amount of carbohydrate assimilated after heading, the rice plant must be subjected to the sufficient amount of solar radiation for 25 days just after heading. For that purpose one must find out the most sunny period in the ripening season in each place from the meteorological data investigeted for many years, and must select the variety to be used and determine the planting date so that the rice plant may effectively utilize the most sunny period for carbon assimilation after heading.

With a view to making clear the difference in yield-determining process between transplanted rice plants and directly sown ones, the authors carried out nitrogen top-dressing experiments on directly sown plants in 1962 and '63 along the same line as in the previous experiments (LI) in which transplanted plants were dealt, obtaining the following results. 1. The most effective time for top-dressing to increase the magnitude of each yield-component was found as follows: (1) The number of panicles per unit area is most effectively increased at the most active tillering stage, (2) the number of spikelets per panicle is at the neck-node differentiation stage, (3) the percentage of ripened grains is at the full heading stage and (4) the weight of 1, 000 grains is just before the reduction division stage. 2. On directly sown rice plants the number of degenerated spikelets was much less than that on transplanted ones, so the nitrogen top-dressing just before the reduction division stage was hardly effective for decreasing the number of degenerated spikelets. (On transplanted plants the nitrogen top-dressing at this time is quite effective for decreasing the number of degenerated spikelets.) The reason why the number of degenerated spikelets on directly sown plants was much less than that on transplanted could be ascribed to the fact that the number of differentiated spikelets was much less on account of much more number of panicles per unit area being produced on directly sown plants than that on transplanted ones, respectively. 3. There was no top-dressing method which consistently gave either the highest yield or the lowest yield under all conditions, which suggested that the best top-dressing method varies with the cultural conditions and seasons. The four components of yield have their individual best time for top-dressing a time when each component is most easily increased, while the yield which is a product of these four components has no one definite top-dressing time at which it is most easily increased. 4. Examining the results, the authors reached the following conclusion which was the same as that obtained from the previous experiments on transplanted plants. The most suitable method of top-dressing differs with conditions under which the plants grow. Under conditions which produce plants with a high percentage of ripened grains and a small number of spikelets, a top-dressing which makes the plants increase its number of spikelets is suitable, while under conditions which produce the reverse effect, a top-dressing which makes the plants increase its percentage of ripened grains is suitable. Accordingly, it is necessary for the growers to adopt the most suitable method of top-dressing by examining the rice plants in their own paddy fields. However, in case of the percentage of ripened grains and the number of spikelets per unit area being both moderate in magnitude or both components being well balanced (and such cases seem to be common), the methods of top-dressings which increase the number of spikelets per unit area as well as the percentage of ripened grains (such as Treatment No. 7 and 10) would be the most suitable.

The efficiency and absorption of basal nitrogen (ammonium sulfate) enriched with 15N was studied in the paddy field in 1969 and 1970. Rice plant was transplanted at two different times, in early and late May, and was grown under the two rates of basal nitrogen, 60-70 kg/ha and 150 kg/ha. It was ascertained that the absorption of basal nitrogen by rice plant almost finished at the end of June in all the experimental plots. After that, the rice plant absorbed only soil nitrogen. The rate of plant absorption of fertilizer nitrogen was low soon after transplanting and increased gradually. The maximum rate was found in the early of June in the early-transplanted plots and in the middle of June in the late transplanted plots. According to nitrogen analysis at different growth stages, the ratios of the amounts of fertilizer nitrogen to those of total nitrogen absorbed by the plant were the highest soon after the transplanting and decreased gradually. They were higher in the plots with the high rate of basal nitrogen than those with the low rate. The recoveries of nitrogen in the plant which were calculated from the tracer were 22 per cent in 1969 and 26-27 per cent in 1970. On the other hand, the gross recoveries of nitrogen were ranged from 43 to 50 per cent in 1970. The ratios of the amounts of fertilizer nitrogen to those of total nitrogenoin the plant at maturity were about 25 per cent in the high nitrogen plots and less than 15 per cent in the low nitrogen plots. The amount of soil nitrogen absorbed by the plants was greater in the plots with basal nitrogen than in those without basal nitrogen. It was assumed that the effect of basal nitrogen contributud to the gross recovery of nitrogen in the plant which was substantially greater than the recovery calculated from the tracer.

On the basis of the fact that the nitrogen content was negatively correlated with the starch content in rice plants, some trials had hitherto been conducted which had intended to diagnose the nitrogen requirement of rice plants at the panicle initiation stage by knowing the relative starch content which was easily known by the iodine reaction in leaf-sheaths. The authors, therefore, examined from 1958 to 1965 the relation between the nitrogen content and the starch content at the neck-node initiation stage (the initial stage of panicle differentiation) as well as the heading stage with a special view to evaluating the diagnosis of nitrogen requirement by iodine reaction as above mentioned. The results may be summarized as follows. 1. A high and positive correlation coefficient was always found between the starch content in the rice plant and the staining degree of a leaf-sheath by iodine solution. In particular, the strongest correlation was obtained when the third leaf-sheath counted from the uppermost developing leaf was used. 2. In cases of an identical variety being used and transplanted on an identical date under an identical season with different amounts of basic ferfilizers without top dressing, a strong and negative correlation coefficient was always found between the nitrogen content and the starch content in the rice plant. However, in case of varieties, transplanting dates, seasons and topi dressings being different, the correlation coefficients markedly decreased in most cases. 3. The nitrogen content seemed to be negatively correlated with the starch content in the rice plant by nature. The correlation coefficient between them, however, decreased markedly when the rice plant had been subjected to abnormal weather conditions or top-dressed with nitrogenous fertilizers just before the plant was sampled, presumably because the nitrogen content had not yet come to be balanced with the starch content in the plant. 4. The effect on yield of top-dressing at the panicle initiation stage had no connection with the content of nitrogen or starch in the plant at the neck-node initiation stage, but it was closely connected with the number of spikelets per unit area, i. e. the effect was progressively remarkable with the decrease in the number of spikelets. 5. Only in the following case one might decide, from the diagnosis by iodine reaction, the top-dressing at the panicle initiation stage to be necessary for the rice plant, i. e. only whenth e rice plant showed not only a severe iodine reaction but also a small amount, of nitrogen absorbed at the neck-node initiation stage and the plant could also be estimated to absorb a small amount of nitrogen during the period from the neck-node initiation stage to the final stage of spikelet initiation : in other words, only when the plant might be expected to have a small number of spikelets per unit area : and moreover the plant could be expected to be subjected to fine weather conditions after heading. Accordingly, the diagnosis by iodine reaction in the rice plant to judge the necessity of the top-dressing to the plant at the panicle initiation stage may be said to be of little utility value.

Since the authors reported the favourable effects of the nitrogen top-dressing at full heading time on the rice plant in the previous paper (XL VII, Vol. 27, No. 1), they have further been examining them for those several years under the various conditions differing in the amount of basic fertilizer, time of planting, spacing and variety, and obtained the following conclusions. 1. Nitrogen top-dressing at full heading time caused the plant to increase the rate of photosynthesis and amount of dry-matter as well in the period after heading except the cases of extremly unfavourable weather conditions and extremely dense spacing. 2. When the amount of dry-matter was increased by nitrogen top-dressing at full heading time, it was generally observed that yield as well as the percentage of ripened grains were clearly increased independently of the time of planting, spacing and variety, except the case of the control (non-top-dressed) plot being high enough in its percentage of ripened grains. 3. No effect of nitrogen top-dressing at full heading time was found on the plant in the case of the translocation of carbohydrates being disturbed by cool temperature in autumn or the enough amount of nitrogen being supplied to the plant from the soil even in the period after heading.

The fate of basal nitrogen (ammonium sulfate) enriched with 15N was investigated in the paddy field in 1969 and 1970. The paddy soil was relatively poor in organic matter, fine-textured and poorly drained. Rice plant was transplanted at two different times and was grown with two rates of basal nitrogen addition as previouly reported. The soil ammonium nitrogen derived from the basal fertilizer was more in the plots with the high rate of basal nitrogen than in those with the low rate. It decreased gradually and almost disappeared in every experimental, plot at the end of June. The fact above-mentioned was coincident with the results of a previous paper indicating that the absorption of basal nitrogen was finished in all the experimental plots at the end of June. The quantities of fertilizer nitrogen left in the soil after harvest ranged from 18 to 21 per cent. On the other hand, 22 to 27 per cent of the nitrogen was absorbed by rice plant. Therefore, the recovery of basal nitrogen was less than 50 per cent. According to the frequent determinations of the amounts of nitrogen absorbed by rice plant and soil ammonium nitrogen, it was assumed that ammonium formation in the soil had two maxima during the growing season -one, from the early to middle June and another, from the end of July to early in August.

It has hitherto been considered that even if the number of spikelets per unit field area is excessive, the percentage of ripened grains may be reduced, but the yield will be nearly constant so long as the plants hold the constant amount of carbohydrates. The present paper, however, has clarified that in fact there is an optimum number of spikelets per unit field area and an optimum percentage of ripened grains equivalent to the amount of carbohydrates per unit field area for maximizing the yield. An optimum number of spikelets per unit field area is represented as a linear function of the amount of carbohydrates but an optimum percentage of ripened grains is found to be nearly constant (about 80 to 85 %) for any amount of carbohydrates to obtain the highest yield. So, the degree of the percecentage of ripened grains can be taken as an index of the diagnosis of the defect in rice cultivation. When the rice plants have less number of spikelets than an optimum, their grain yield decreases in proportion to the number of spikelets, however high the percentage of ripened grains may be, and the amount of carbohydrates which remaines in shoot at maturity increases in inverse proportion to the number of spikelets. While, when the rice plants have more number of spikelets than an optimum, their grain yield decreases in inverse proportion to the number of spikelets, because of the number of imperfectly ripened grains being increased. The main reason for this fact can be explained by that as the number of spikelets increases, the given amount of carbohydrates must be shared with the more numerous spikelets, causing a more than proportional increase in number of grains which have an insufficient share of carbohydrates and consequently resulting in proportional decrease in the number of fully ripened grains.

1. The number of differentiated spikelets is firstly affected by the amount of nitrogen absorbed by the time of late spikelet differentiation and secondly by the amount of carbohydrates produced during the period from the neck-node differentiation stage to the late spikelet differentiation stage. 2. The number of degenerated spikelets is found to be determined by the interrelationship between the number of differentiated spikelets and the amount of carbohydrates produced during the period from the late spikelet differentiation stage to the heading stage. The amount of nitrogen in leaf-blades during the period of determining the number of degenerated spikelets has no immediate effect, but has a mediate effect on determining the number of degenerated spikelets by influencing the rate of carbon assimilation. 3. Both the number of differentiated spikelets and the number of degenerated spikelets have nothing to do with the amount of carbohydrates accumulated in the rice plant. 4. It has already been pointed out by the other research workers that the nitrogen content in leaf-blades is strongly correlated with the number of spikelets per panicle, and also the weight of leaf-blades per hill or the amount of total nitrogen in leaf-blades per hill at the heading stage is closely correlated with the number of spikelets per hill, but it has also been clarified by the present experiments that the above-mentioned facts can hardly be generalized and they hold true only in the following cases:-(1) an identical variety is used and it is sown or transplanted on an identical date, (2) the period of determining the number of panicles is not overlapping with the period of determining the number of spikelets, and (3) the amount of dry-matter increased during the period from the initial stage of reduction division to the heading stage is positively correlated with the amount of nitrogen in leaf-blades at the heading stage.

As the plant chamber of the apparatus reported in the previous paper was higher in temperature and in humidity than under natural condition, the authors made an effort to improve the defects by setting up a refrigerator (2.H.p.), resulting in a good success. Using the improved apparatus, they made further experiments and tried to apply the results to the yield-forecast. 1. Relation between the rate of carbon assimilation and light intensity. The relation of light intensity to the rate of carbon assimilation in different growing stages showed as much the same relation as that of the previous paper, as shown in Fig. 1. Light saturating points were found at the intensity of approximately 0.6 cal./cm2./min. in the period before the heading time, and in subsequent period the intensity seemed to begin to fall off day by day, as seen in the former report. In spite of using 6 varieties differing in earliness and in ecotype, varietal differences of saturating light intensity could hardly be seen so far as this experiment was concerned. An there were no clear differences regarding saturating intensity among plants grown in differently fertilized conditions, as shown in Fig. 2. Moreover, even in the conditions differing in spacing and in temperature differences of saturating light intensity could also hardly be distinguished, though the experiments were not enough to ascertain. 2. Diurnal changes of the rate of carbon assimilation. Diurnal changes of the rate of carbon assimilation on fine days were closely resembling to those of the previous paper. From this experiment it was reaffirmed that the stronger light intensity than 0.6 cal./cm2./mln. is of no use for carbon assimilation. Diurnal changes on partly cloudy days or cloudy days were invesigated, as shown in Fig. 3. From these figures it can be drawn out that the assimilating activity holds almost constant in the higher light intensity than the saturating point (0.6 cal.), but it changes almost completely parallel to light intensity in the lower intensity than the saturating one. 3. Seasonal changes of the rate of carbon assimilation. Seasonal change of the carbon assimilation on middle maturing variety (Norin-No. 15) was traced up in 1953-'54, and the results in Fig. 4 were obtained. As the curve of 1953 in Fig. 4 was obtained under the higher condition by 10°C than the normal and the weather of the year was abnormally bad, the curve seemed to be less reliable than that of 1954 as the seasonal change in general. The curve in 1954 shows that maximum rate of photosynthesis as a whole plant occurs at the maximum stage in number of tillers........the stage is aiso the beginning stage of young panicle formation........and the second maximum occurs at heading time, and the former is much higher than the latter. 4. Application to the yield-forecast. From the resuts above mentioned it was made clear that relative amount of carbon assimilation in rice-plants at a given stage can be estimated by measuring light intensity to a considerable extent. Using daily data of sunlight intensity obtained from "Robich solarimeter", covering the period from transplanting time until ripening in 1951-'54 the authors tried to calculate the index of daily relative amount of cabon assimilation and arranged them in mean value for 10 days as written in Table 1. For instance, in the first decade of July in 1951 the relative amount of carbon assimilation is 69.8% as compared with that of the case of complete fine weather. The index of relative amount of carbon assimilation has to be corrected by the differences of the day-length and of the rate of photosynthesis in each decade. Moreover, the inexed at the stage of reduction division and of milk-ripening are necessary to be weighted, because the amounts of carbon assimilation at those two stages have always marked effects upon the yield of rice. [the rest omitted]

In the previous report the authors reported that one could easily obtain high yields by using a circulating irrigating method combined with the application of nitrate nitrogen. By this method one could exactly take the nitrogen off the soil at a given growth stage for controlling the growth of rice plants. In the present experiment, furthermore, the authors tried to apply the same principle as the above to an ordinary paddy field without the circulating irrigation method. The main points in conducting the experiment were the following three points: (1) In order to ensure the necessary and sufficient number of bearing-tillers for obtaining a terget yield by the time of 70 in 'leaf-number-index', and to take nitrogen off the soil without difficulty during the critical period in which the plant type after heading is definitely predetermined, much amount of nitrate nitrogen in stead of ammonium nitrogen was given at the early growth stage. (2) In order to take nitrogen off the soil immediately during the critical period, logged water which contained nitrate nitrogen in the paddy field was completely drained out, and fresh water which contained no nitrogen was again irrigated. (3) In order to increase the rate of carbon assimilation after the critical period, nitrogen was again top-dressed at the end of the critical period (about 20 days before heading) and at the full heading time. Water permeability in the experimental field was at a rate of 10∼15 mm per day. The results might be summarized as follows. 1. Nitrate nitrogen applied at the early stage of growth was quite effective for increasing the number of tillers, and it could be taken off the soil by exchanging logged water in the field for fresh water at the begining of the critical period. The success in restricting the nitrogen supply in the paddy field mentioned above could be recognized directly by the remarkable decrease in the nitrogen content of leaf-blades, and be checked indirectly by the discoloration of the plant and the retardation in increasing rate of the number of tillers and the dry-matter during the period, and furthermore by shortening of leaf-blades, culms and internodes in length at maturity. 2. Accordingly, the type of plants was improved to a near ideal type by the above-mentioned treatment, and, as a result of it, treated plots were higher than the control in the percentage of ripened grains and in the grain yield as seen in table 3. 3. It was quite noticeable that the treated plants (1) had gained a high lodging resistance which was mainly ascribed to the shortening of lower internodes in length, and (2) has yielded high percentages of ripened grains which was largely ascribed to abundant accumulation of starch in culms and leaf-sheaths before heading. 4. From the results of the present experiments it was strongly proved that if one would utilize nitrate nitrogen at the early stage of growth, one could make the plant type ideal with ease by taking nitrogen off the soil at the critical period, and could obtain high grain yield through the "ideal plant".

In a field experiment long-term varieties yielded greater total dry weight than short-term varieties but had similar dry weight of ear and grain yield to short-term varieties except a few treatments, and the dry weight ratio of ear to shoot or grain yield to total dry weight was greater for short-term varieties than for long-term varieties at maturity. A positive correlation coefficient was found between the dry weight ratio of ear to shoot at maturity and the ratio of dry weight produced after ear emergence to dry weight at ear emergence, number of spikelets per unit dry weight at ear emergence, RGR. or NAR. during ripening stage. A negative correlation coefficient was found between the dry weight ratio of ear to shoot at maturity and dry weight at ear emergence or number of days from transplanting to ear emergence. These facts might be ascribed to that the weight of ear was determined by the amount of dry matter produced after ear emergence or the number of spikelets and that there was no definite relation between dry weight produced after ear emergence or number of spikelets and dry weight at ear emergence, in case of varieties and transplanting dates being different.

The distribution of absorbed nitrogen derived from basal and top-dressed fertilizer was investigated by tracing 15N. In the basally fertilized plots, at rooting stage, nitrogen derived from fertilizer was distributed apparently more in leaf sheaths than in laminae, and in the top-dressed plots, it was distributed apparently more in leaf sheaths than in laminae just after top-dressing, because there were new growing organs in leaf sheaths. The nitrogen derived from fertilizer in rice plants translocated gradually from sheaths to laminae with emergence of new laeves. So, it could be assumed that a large amount of absorbed nitrogen distributed in new growing organs. At the later growth stage the rice plants were top-dressed, it took the less number of days to accumulate the maximum amount of nitrogen in the laminae. Nitrogen applied at the ear emergence was firstly found in leaf sheaths, culms and laminae, and it translocated to ears with the progress of ripening. At maturity, nearly 70 percent of absorbed top-dressing nitrogen was found in the ear. At the earlier growth stage of rice plants fertilizer was applied, the higher ratio of tagged nitrogen in the laminae to that of the whole plant was there at the ear emergence, and the lower ratio of tagged nitrogen in the cars to that of the whole plant was there at maturilty.

For maximizing the yield of rice it is quite necessary to obtain high percentages of fully ripened grains under the condition in which too many spikelets per unit area have been borne. Conducting many experiments for long years, Matsushima, one of the authors, realized that it could by no means be achieved without using productive soil, but recently he noticed an idea that it might be achieved not by using any productive soil, but by controlling growth habit (i. e., shortening the lengths of upper 3 or 4 leaf-blades and those of inter-nodes for avoiding the mutual shading of individual leaves) and by increasing the rate of carbon assimilation per unit leaf-area after heading. The authors, therefore, tried to maximize the yield by growing rice plants by water-culture under community conditions, and compared them with those grown in ordinary paddy fields in which a large amout of compost was applied. (By using water-culture, one can easily control the growth habit of rice plants at any growth stage by increasing or decreasing the concentration of nitrogen in the solution.) As a result of it, the rice plants grown by water-culture yielded the amount of 101.6 kg and 77.2 kg of kernels (brown rice) per are, while those grown in paddy fields produced only 57.6 kg at the most. In the present experiment a main control of growth habit was done by decreasing the nitrogen concentration in the solution or taking off nitrogen from the solution during the period from 41 days to 20 days before heading, making the plants short in the lengths of upper 3 or 4 leaf-blades and those of inter-nodes and also thick in the thickness of leaf-blades. On the basis of these facts, the authors concluded the meaning of the application of organic fertilizer as follows; i. e., under the condition of no harmful effects caused by decomposing organic matters in soil being found, organic fertilizers (1) supply the necessary and sufficient amount of all kinds of nutrient elements to the plants through their whole lives, and (2) act as a "buffer" not to supply nitrogen to an excess during the period in which the lengths of upper 3 or 4 leaf-blades and those of inter-nodes are determined.

The fate of top-dressed nitrogen (ammonium sulphate) enriched with 15N was investigated in the paddy field. The amounts of nitrogen applied were 30 kg per hectare in 1969 and 25 kg per hectare in 1970, respectively. The earlier was the top-dressing at the growth stage of rice plant applied, the longer stayed the ammonia derived from the top-dressed fertilizer in the plowed layer. This fact is approximately consistent with the duration of absorption of top-dressed nitrogen by rice plant. After harvest only 3 to 10 per cent of top-dressed nitrogen was found in the plowed layer. Since the drainage of the paddy soil was very poor, it was assumed that almost all of the unrecovered top-dressed nitrogen was lost by denitrification. The recoveries of top-dressed nitrogen were higher for the plots applied at the young panicle formation stage than for other ones. The earlier was the top-dressing applied, the longer the duration of top-dressed nitrogen uptake by the plant. And the later was the top-dressing applied, the higher the absorption rate of nitrogen by the plant. The amounts of soil nitrogen abosrbed by the plant were greater in all the plots with top-dressing nitrogen than in those without top-dressing. Therefore, the gross recovery of nitrogen in the plant was greater than the recovery calculated from the tracer.

1) Two further experiments have been carried out to confirm the hypothesis, derived from the previous experiments (XLI, Vol. XXVI. No. 1.), that the gradient of sugar concentration in a rice plant is one of the most influential factors in determining the rate of translocation. From the experiments following facts have newly been made clear, as can be seen in Table 1: (1) the rate of translocation is markedly disturbed even in a midday by shading the rice plants, and (2) sugar concentration in a rice plant gradients in the order of leaf-blade>sheath>culm>panicle, and the gradient between a leaf-blade and a panicle is the larger at the time when the translocation is the more active. 2) Entire plants being subjected to different temperatures for 20 days after heading, the relation of the rate of translocation to the temperature has also been experimented. From the experiment it has been elucidated that the rate of translocation increases with the temperature up to about 25°C, but it seems to keep almost constant in higher temperatures than 25°C, as can be seen in Fig. 1 and 2. 3) Enabling the rice plant to favour its photosynthetic activity, the nitrogenous top-dressing at the time just after heading has been proved to play an important roll in increasing the percentage of ripened grains on many cases where rice plants have less contents than 1.25% in total nitrogen at heading time, as seen in Table 2.

Using the same apparatus reported in the previous papers, the authors made further experiments. 1) The differences of the light curves of carbon assimilation in nursery beds between thinly and thickly spaced rice-seedlings could hardly be found. 2) Nitrogen supplies at different growing stages always increased the photosynthetic activity of rice plants. Especially, the top-dressing at the beginning stage of panicle differentiation caused the rice plants to make a marked increase in the rate of carbon assimilation per unit leaf-area. 3) Drought and shading treatments decreased the rate of carbon assimilation, but the assimilating rates of shaded plants were much more active than the normal ones at a certain period after finishing. the treatments. 4) Fertile tillers (which bear heads) were superior to sterile tillers (which fail to head) in the photosynthetic activity at the time shortly after the stage of maximum number of tillers. 5) Rice plants subjected to strong wind decreased their photosynthetic activity. 6) Carbon assimilation of rice plants in actual fields were measured also. The results were nearly the same as those of previous papers, except for the seasonal changes of carbon assimilation.

Investigating the vertical distributions of CO2 concentrations on actual crop-fields, and applying their values to the formula (1) derived from Thornthwaite-Holzman (1942; DSDA Tech. Bul., 817), the authors tried to establish a new method for measuring the rate of carbon assimilation under actual field conditions, and got the results suggesting the method would be a promising one.

1. A noticeable fact was drown out from the present experiment that the rice plants, which possess the same amount of carbohydrates and the different number of spikelets, always differ not only in the percentage of ripened grains but also in the actual number of ripened grains and in the yield of grains, and the excessive spikelets in number always act on the plant to decrease the number of ripened grains and the yield of grains on account of causing the plant to increase the number of imperfectly ripened grains which are to be eliminated in the course of threshing, husking and cleaning. From this point of view it was emphasized that the excessive growth of rice plants often brings about a rather much reduction in yield of rice-grains. 2. It was made clear that a high content of nitrogen ih rice plants at heading time, which has hitherto been considered to be more injurious than beneficial to ripening, does not damage the ripening of grains, but rather favors it, enabling rice plants to raise their photosynthetic activity. The reason why a high content of nitrogen at heading time has been considered to damage the ripening could be explained by the fact that a high content of nitrogen brings about a large number of spikelets per plant and a large number of spikelets per plant are liable to act on the plant to decrease the percentage of ripened grains. 3. The amount of carbohydrates accumulated after heading, which was proved to be most inportant in constituting the yield of rice in the present experiment, was strongly correlated with the total amount of nitrogen which is the product of the dry weight of rice plants and the nitrogen content in rice plants. And the product of the total amount of nitrogen in rice plants and the total amount of solar radiation expressed in calorie was ascertained to be highly correlated with the grain-yield. From these facts, it was pointed out that the total amount of nitrogen in rice plants at heading time would be of great importance in predicting the grain-yield of rice.