Plant biomass, litter accumulation, and soil carbon storage were measured in four different-aged deciduous broadleaf forests (P1-P4) that had been clear-cut and in a mature forest (P5). Plant biomass increased as forest age increased, and leaf and fine-root biomass were almost constant. Accumulation of dead plant parts decreased at first and then gradually increased. Recovery of fine-litter accumulation was observed at an early stage of regeneration. Carbon accumulation in the soil decreased remarkably in the first 40 years and then gradually increased. Total carbon accumulation was 202 tC/ha in P1 (12 years old), 180 tC/ha in P2 (45 years old), and 323 tC/ha in P5 (mature). The forest ecosystem changed from a CO2 source to a CO2 sink, relative to the atmosphere, fairly early in the regeneration.
The soil condition of several types of Alnus japonica stands, and neighboring Ulmus davidiana var. japonica stands were studied in a small tributary basin in Miyagi Prefecture, Japan. The soil type of the study area varied from anaerobic peat soil in poorly drained areas on back swamps to aerobic alluvial soil in well drained areas. Differences in soil properties among the stands were clear in the top soils. Anaerobic soils with high clay and high organic matter contents and high CEC value were occupied by A.japonica-Polygonum thunbergii stands. A. japonica-Ilex crenatd var. paludosa stand on peat soil showed also high values for these properties. A. japonica-Filipendula kamtschatica stands and U. davidiana var, japonica stands were characterized by well drained soil which was low in organic content and contained much coarse sand. These stands were similar in such properties, but soils of A. japonica-Filipendula kamtschatica stands were lower in degree of base saturations, more acidic and more anaerobic than those of U.davidiana var. japonica stands.
Peculiarities of Japanese mires are discussed on the ground of volcanic activity and chemical properties of river waters and volcanic ash. A type of mire resembling to the "Hochmoor" developed on tephras is pointed out, and a new concept, "a tephratrophic mire" is proposed by the authors. This kind of mire is defined as a renewed, or renewing mire under the influence of tephras. This new proposal is to stress that classification of wetland vegetations should be based on water chemistry and hydrology rather than morphological features. From this point of view, mires in Hokkaido are distinguished as follows : (1) the bog and fen part of the tephratrophic mixed mire, (2) the mountain tephratrophic fen, (3) several types of lowland fens, and (4) fen swamp (alder swamp). Furthermore, the authors separate 5 mire zones vertically and horizontally in Japan by mire types and presence or absence of peat : (1) the zone of mountain bog of Hokkaido, (2) the zone of raised bog of Hokkaido, (3) the zone of mountain bog and unplad raised bog of northern Honshu, (4) the transitional zone, and (5) the zone of peatless mires of southern Japan.
The bionomics of a eusocial halictine bee Lasioglossum (Evylaeus) duplex (DALLA TORRE) was compared between two populations in northern Japan, those of Sapporo and Sendai. Although a slight phenological shift was observed, the life cycle pattern did not differ between the two populations, both of which consists of six phases involving two brood rearing periods, a spring solitary phase and a summer eusocial phase. Among other similarities, the rarity of spring semisociality and of nest reuse may be regarded as important species-specific characters. As for the differences between the two populations, (1) frequent preparation of multiple cell clusters in summer, (2) larger body size in both the queen and the workers, and (3) a smaller caste-linked size difference were noteworthy in Sendai. It is difficult to explain (1) by (3), because the social structure of summer colonies is similar in both populations, but the possibility that a social factor plays a role in (1) can not be precluded. Further comparisons of more remote populations of this and other halictine species are needed.
From species composition, it appeared that the majority of earthworms occurring in the mineral soil deeper than 10 cm and 20 cm in the ridge and bottom plots, respectively did not feed on surface litter. Based on this finding, biomass of "potential" litter-feeding earthworms was estimated and was found to differ only marginally between the two plots on the studied slope. Judging from the carbon content of earthworm faeces, earthworms in the A0-layer were considered to feed exclusively on A0-layer material. Biomass of these "actual" litter-feeding earthworms was greater in the ridge plot than in the bottem plot. Even when the biomasses of earthworms in the topsoil (0-10cm) and "actual" litterfeeding earthworms were summed, it was still somewhat larger in the ridge plot than in the bottom plot. The "mull-type" humus in the bottom plot and the "mor-type" hunus in the ridge plot indicated the relatively faster consumption of surface litter by soil organisms in the bottom plot. The estimated biomasses of both "potential" and "actual" litter-feeding earthworms were negative to a great contribution of earthworms to this differential consumption of surface litter and thus to the differential development of the A0-layer observed between the two plots.
Phytosociological vegetation mapping was conducted in 1974 and 1980 to study the successional change of community types after pine damage in and around the Natural Botanical Garden of Hiroshima University on Miyajima Island. Change of the occupying area of each community was mainly due to disturbance by selective cutting and transportation of dead pine trees. The most distinctive increase of the area was observed in the community type at an early successional stage, Typcal variant of Symploco-Pinetum densiflorae myrsinetosum seguinii, which showed a 10.2% increase during the six year period. The most declined community, which showed a 12.4% decrease at a rather late successional stage, was Neolitsea aciculata variant. The transition among the community types commonly occurred as follows : from Rhododendron kaempferi variant to Typical variant, from Neolitsea aciculata variant to Rhododendron-Neolitsea transitional community and finally to Typical variant. This finding implies that the change of the vegetation was caused by a retrogressive succession rather than by a progressive one. Change of the vegetation in 1974-1980 in relation to the microtopography was also discussed.
The reproductive behavior of the male Japanese toad, Bufo japonicus japonicus, was observed at several ponds in Kanazawa prefecture, central Japan. In the two ponds where males outnumbered females by six to one, 88% of the male toads stayed on the land around the pond, waiting and searching for females. On the other hand, in another pond, where males and females were nearly equal in numbers, 44% of the males waited and searched for females in the pond. The males' waiting position on land was located on a line between their place of residence and the pond. Males stayed longer in the breeding pond (5 to 6 days) than females (1 to 2 days). The maximum record of breeding times was 7 times in 8 seasons for a male, and 4 times in 6 seasons for a female.