POPULATION-DYNAMICS OF ALNUS-INFECTIVE FRANKIA IN A FOREST SOIL WITH AND WITHOUT HOST TREES

Little is known about the seasonal dynamics of Frankia populations in soils. For this purpose, we sampled forest soils at a site in northern Sweden (66-degrees-31'N) twice during the snow-free period: in June and September 1991. Several treatments were established at this site during 1977-1980, including planting lupine (Lupinus nootkatensis Donn), alder (Alnus incana (L.) Moench)-an actinorhizal host, liming and NPK fertilization. We collected soil samples from control, lupine, alder, alder limed, alder plus NPK, and alder limed plus NPK. Populations of Alnus-infective Frankia were measured in two ways: (i) a plant bioassay MPN to determine nodulation units (NU) of Frankia, and (ii) the polymerase chain reaction with oligonucleotide primers specific to Alnus-infective Frankia to quantify the number of Frankia genomic units (GU). Numbers of NU ranged from 25 to 182 g-1 soil and GU from 1700 to 3400 g-1 soil. There were statistically-significant treatment and seasonal effects on NU. The control and alder limed treatments had about 2.5 times higher NU than the lupine and alder limed plus NPK treatments. Frankia NU decreased almost 3-fold from June to September. There were no significant treatment or seasonal effects on Frankia GU. These data suggest that the nodulating ability of Frankia is controlled largely by the physiological status of Frankia, as indicated by infectivity, rather than total population size, as indicated by GU.