1. The problems of studying pollen transport and the relation of transport to the evaluation of pollen analyses of sedimentary cores are reviewed. 2. A series of eight pollen-collecting stations was established across the San Augustin Plains of New Mexico and the pollen rain examined for the 1956 pollinating season from slides exposed for 24-hour and 7-day periods. 3. Examinations were also made of moss polsters and surface silt to correlate the pollen content accumulated and preserved for the past several years with the present vegetation. 4. Arboreal pollen comprised 71-99% of the pollen rain at the eight stations, with an average of 93%, even though all stations except one were in open areas. 5. The average composition of arboreal pollen from all stations was: Juniperus, 61%; Pinus, 26%; Quercus, 9%; and other species, 4%. 6. The pollen rain at all stations was most closely correlated with relative basal area of tree species, which for the forested areas was: Juniperus, 40%; Pinus, 50%; Quercus, 9%; and others, 1%. At all stations, however, there was an overrepresentation of Juniperus and an underrepresentation of Pinus and Quercus in relation to all phytosociological measures-relative density, relative coverage, and relative basal area. 7. At Station 4, located at the site of the drill cores obtained for pollen analysis in 1953 and 1958, the entire season's pollen rain in 1956 was composed of: Chenopodiaceae and Amaranthaceae, 55%; Juniperus, 17%; Pinus, 13%; Quercus, 5%; and others, 10%. The distribution between arboreal and nonarboreal pollen for the entire season was 38% and 62%, respectively. The 38% arboreal pollen occurred in spite of an 11-mi. distance to any woodland areas to the southwest, the direction of prevailing winds. 8. On the basis of this study it is concluded that, if shrubby vegetation is involved at the site of sampling, a percentage of arboreal pollen exceeding 40-50% of the total collected must indicate a forest border within a few miles. In grassland areas, however, as at Station 2, arboreal pollen percentages may exceed 90% even when forested areas are farther away. 9. Pollen dispersal of Juniperus reached a maximum the last week in March; of Quercus the first week in May (low in concentration); and of Pinus in mid-June, with no evidence of peaks resulting from P. edulis and P. ponderosa. Chenopodiaceae and Amaranthaceae showed several peaks from mid-July to mid-August, probably because of the importance of several species, such as Sarcobatus, Atriplex, Salsola, and Suaeda. Pollen of Gramineae composed only 21/2% of the total pollen rain and was received principally in August and September. 10. Slides exposed for an entire week instead of for a single day did not have a pollen count seven times, but only three to four times, greater than that of the single-day exposures. 11. The general composition of the pollen rain was about equal for 1-day-a-week and for weekly slides. During the season at Station 4 the ratio of arboreal to non-arboreal pollen for 1-day and for weekly slides was 33:67 and 38:62, respectively. The principal genera are also equally well represented on the two sets of slides. 12. The 1-day-a-week slides do not provide as complete a record of pollen grains of rare species or of long-distance transport. For example, fifteen genera or families were tallied on the weekly slides which were absent on the 1-day-a-week slides. This omission would be important in detailed studies of long-distance transport or in the use of pollen collections as aids in detailed phytogeographical studies. By chance, two genera represented on the 1-day slides were absent on the weekly ones. 13. Unusual pollen records of species not present in the area of the San Augustin Plains and likely to have blown in from desert areas approximately 100 mi. to the southwest included: Acacia, Cercidium, Fouquieria, Larrea, and Prosopis. 14. Pollen of Abies and Pseudotsuga was low in number; the latter is recognized to be of short transport distance. Pollen of Populus and Typha was rare and could be associated with local stands. 15. Picea pollen was represented by only two grains in all samples. No spruce occurs in the San Augustin Plains drainage, but a large stand exists about 14 mi. to the southwest. The almost complete absence of Picea pollen falling into the San Augustin Plains basin, when it is so near at hand, would mean that, when Picea pollen is found in the pollen composition of sedimentary material, spruce forests must have been very near to the site of deposition. Picea is present in Clisby's pollen analyses of the San Augustin cores, where maximum concentrations of 35% are recorded. A lowering of life zones by about 1000 ft. would have enabled spruce to cross a passageway into the San Augustin Plains drainage. According to present reconstructions of post-Pleistocene shifts in zonation, this migration would easily have been possible. 16. The analysis of moss and lichen polsters revealed a good correlation with surrounding arboreal species composition. Where brush and herbaceous species were characteristic of an area, they were also found to appear in the pollen of the polsters of that area. 17. The analysis of surface-silt samples with a pH of 9+ surprisingly yielded pollen. Samples collected in July and November varied in the proportion of arboreal to non-arboreal pollen, presumably in response to the extra amount of current-season nonarboreal pollen commonly produced after July. When the pollen analysis of the silt is compared to the current pollen rain, it is apparent that Juniperus pollen is not well preserved. Pollen of the Chenopodiaceae-Amaranthaceae group and of Gramineae seems to be preserved in proportions typical of their concentration in the atmospheric pollen rain.

Text in English

Reprints Collection