Even in summer, the cold air, lower than average annual temperature by 10°C, is exhausted from underground through spaces within debris at the sites called “Fuketsu”. At the Fuketsu, the subsurface air circulation occurs. In winter, the air is inhaled from lower part of a slope, and exhausted at upper part of a slope. During this kind of the air circulation, the debris is frozen to even 50 m depth. In summer, the air circulation is reversed. The inhaled air is cooled to almost annual average temperature before arriving frozen debris. Due to these processes, debris remaines to be frozen untill autumn and this is the reason why the exhausted air in summer is abnormally cold. The common ground condition of the “Fuketsu” is studied by relative comparison of 20 samples, and the result is as follows. 1) Snow covered districts in winter. 2) North facing slopes. 3) Abundantly debris accumulated slopes in front of steep scarps, gigantic landslide area is favorable to this condition.
In the mountainous regions, so-called wind-shaped trees are frequently used as an effective indicator of wind. But it is a matter for regret that their causality is not always apparent. Therefore, the main purpose of this paper is to determine what is actually the causative factor for deformation of larches (Larix leptolepis) on Mt. Fuji. In the first place, the following types of deformation are classified. 1) C type: This is generally low in stature and the trunk seems to creep on the ground. 2) F type: This is a type of deformation which has its crown on one side only. So-called flag-shaped tree. 3) S type: This type shows only a slight degree of deformation. Then, the directions and types of these deformed trees are surveyed along 30 routes.(Fig. 1) Their distribution is shown in Fig. 2, which is presented on a circular gragh as a model of Mt. Fuji. But, I can't relate this result with any particular cause. It is only noticeable that there is a great difference in directions between F type and Stype as shown on the 15th route (Subashiri) as an example. On the 2nd May, 1970, I observed that the reddened bark on one side of a trunk (Photo. 1), which had been probably formed in winter, corresponded almost exactly to the side on which branches and boughs are lacking. On the other hand, for the purpose of determining wind action, painted plates were set on the trunk of F type tree in four directions so that one of them was faced to the direction of the expected deforming agents.(Photo. 2) They were exposed to wind and preciptation during summer or winter. Consequently after a winter exposure, the plate corresponded to the direction of F type tree was most damaged.(Photo. 3) From these evidence, F type deformed larches are expected to be formed during winter time. Now, on the top of the mountain, strong westerly or northwesterly wind prevail in winter.(Fig. 5) But as a result of detailed investigation of meteorological data reported by extraordinary observations on southeast slope during January 1951, it seems that the wind coincided with the direction of deformed trees blows during the night when northwesterly monsoon slightly weaken.(Fig. 8) For the duration of blowing this wind, the vapour tension and air temperature become lower. This colder and drier northwesterliesin winter are thought as an important cause for the F type deformation. Different from the other types, the direction of S type deformation indicate southward on the whole. This direction is in good agreement with southwesterlies accompanied with passing of trough or depression. From a point of view described above, the distribution of expected wind direction is drown in Fig. 9. Emphasizing only on ridges because of probability representing larger scale wind system than in valleys, I drew streamlines along the slopes of Mt. Fuji.(Fig. 10) The figures clearly show the existence of up-current in “summer” and down-current in winter. Further, I tried to investigate about the variation of form of trunks with altitude or direction of slopes. The ratios of the height of trunks to the diameter of breast high (H/D) on each routes are calculated and averaged every 50 m high.(Fig. 11) Some character are pointed out in this figure. Namely, two groupes are distinguished by the H/D values around tree limit. Refering to Fig. 12, the extent of C type range likely contributes to the difference between both groupes. In addition, the H/D values do not increase in both groupes from about 100 m below the tree limit. It appears independently of the types of deformation or direction of slopes. This tendency is determined by the distance from the tree limit, and likely related to the increase of tree densities with descending the slopes.