The author tried, in this study, to examine the proportional relation between the scale of conflagration in Japan and the wind-velocities shown at the time of the spreading of the fires.<br>First, he prepared the scale of 7 gradings ( I to VII ), by which the conflagration in Japan which the conflagration in Japan were grouped according to their scales. Then for the purpose of calculating the intensity of burning, he made the following formula ;<br>b_a=A/t<br>b_a : intensity of burning (m²/min).<br>t : duration of conflagration (from the time when the fire broke out to the time when it was extinguished). (min.)<br>A : total floor area of the buildings burnt at the conflagration. (sq. m.)<br>After having examined the intensity of burning of each conflagration by applying the above formula, he could produce 5 tables and 5 graphs.<br>And on this substantial study, he found the scale of conflagrations is not always proportionally subject to the wind-velocities shown at the time of spreading of the fires, after having taken into consideration of the intensities of burning.<br>What is the reason why the intensities of buring were not proportional to the wind-velocities at these occasion ? There seemed to exist several factors to prevent proportional burning with the wind-velocity. In the study, he could find those factors as follows :<br>1) In the scale of conflagrations, from I to III grades ;<br>The wined-velocities have a little influence to spread the fires, but factors, other than wind, have much larger influences.<br>2) In the scale of conflagration, from IV to VII grades ;<br>The influence of the wind-velocities was slightly more than the above item : but much stronger factors which had influence to spread the fires were recognized.<br>After all, he recognized, in the study, the above-mentioned factors as follows :<br>a ) Value of the wind-velocities, used in this study, was not the same as those at the exact spots where the fires were burning.<br>b ) Burning conditions of the fire were not always uniform, which was proved experimentally by the physical trials of the burning conditions of model fires.<br>c ) Buildings, of the town was composed, were not similar in their sizes and qualities.<br>d ) There were lots of varieties of the organizations of fire-fighting and fire-extinguishing system in Japan.<br>e ) There were many varieties of the time of the breakingout of the fires. (day and night).<br>In conclusion, he may add that, within the study of the large scale of conflagrations, the wind conditions on those occassions should not be neglected, but that other stronger controlling factors on intensities of burning are observed as important elements as he just mentioned above.