1) Permanent quadrats of 1 sq.m. were laid in the abandoned field in the grounds of the Chiba University. On laying quadrats, the surface soil was cultivated, mixed, and weeded at the begining of the year. The kinds of permanent quadrats are as follows : [table] 2) A part of the floristic composition of each plot is shown in Tables 1-6. The second-and third-year fields (plot 53) were dominated by Erigeron annuus. The chief dominant in the first year Ambrosia elatior remains still, but it is present in smaller sizes in the second-year fields. The first-year fields (plot 54a) were dominated by Ambrosia elatior as stated in Report I, where in autumn the second-year dominant Erigeron annuus germinated in abundance. However about one fourth died in winter. In the plots 54b and c, Erigeron dominates faster than usual by cutting Ambrosia. Early succession in abandoned fields includes the following stages in our experiments : summer annual weed (Ambrosia), winter annual or biennial weed (Erigeron), perennial grass (Imperata and Miscanthus). Imperata invades the early stages very slowly, progressing inward chiefly around the margins of the plots. 3) The yearly variation of biological type spectra (Table 7) shows the prograssion of succession in decreasing Th, and e, and increasing H, G, I, and p. This means the increment of the spatial utilization of species and the development of the community structure. The seasonal variation of biological type spectra (Table 8) shows by decreasing e that some of the erect types become prominent and the number of those species decreased, while the number of those individuals increases. The increment of t means the progression of the spatial utilization of species. In the comparison of the community types of 54a, b, and c (Table 9,Figs. 1-2), increasing H, G, t, 1,and r and decreasing Th and D_<1-2> seem to show the development of communities. However, decreasing R_<1-3> and pr, and increasing e indicate the opposite tendency. In such a way, the cut dominant or dominant group bring a kind of pseudo-developmental process. 4) When the organizing process of weed communities is shown by the number of individuals-rank of species relations, there are three important types, linear, L-type, and S-type as shown in Fig. 3. In such a figure, the angles of the subordinate groups against the y-axis (Table 10) correspond to the power relations of components. Small angles (sheer linear relations) show a severe interspecific competition. The dominant ratio (Table 11) varies seasonally accompanying with the progression of the spatial utilization of species. The linear, L-type, and S-type relations are based on MOTOMURA's law of geometrical progression, WILLIAMS' law of logarithmic series, and PRESTON's lognormal type of population respectively. Linear relations are said to satisfy in case of a small size of the sample. Besides this, however, the stratification by the growth stage or the community layer is very important biologically. Even the small size of the sample does not show a linaer relation when not stratified (Table 12,Fig. 4). 5) Ecological parameters concerning the speciesarea or species-individual relation vary seasonally and yearly. For instance, the index of diversity does not vary as the number of individuals (Table 13). The species-log. individual relation is curved (Fig. 5). The reasons why such curves are present even in case of small α will be based on the contagious distribution and themendous increment of the number of indiduals of partial rosette types in autumn. 6) The mode of distribution is examined based on the dispersion diagram (unit area : 5×5 sq. cm.). Especially POISSON types and THOMAS' distribution are examined (Table 14). It is said that abundant species such as Ambrosia, Erigeron, and Setaria are distributed after THOMAS' series fairly well, and fewer species such as Digitaria and Imperata afte
(View PDF for the rest of the abstract.)