It is commonly accepted that percentages of pollen in a pollen diagram do not express the exact composition of forests in earlier times. This inaccuracy is due to several factors, for instance the different quantities of pollen produced by plants, the distance of transport etc. A pollen diagram tells us only the change in pollen rain on the locality where we collected soil samples. In studying a pollen diagram we find a close relation between the variations in the percentages of a certain species and the area occupied by this species in the vegetation. When the percentage of pollen of a species increases, we conclude generally that the relative area occupied by this species in the vegetation increases too. However, such a connection might be doubted. The variety of factors controlling the dispersion of pollen is so great that the interpretation of a pollen diagram often meets with great difficulties. The connection between pollen rain and the composition of the vegetation is a simple one in the cases where we are dealing with a region of uniform vegetation. A diagram taken from a region in which the vegetation varies from place to place has to be regarded with some caution. Unfortunately such a heterogenity of the vegetation exists on the very place, where we want to compose a pollen diagram. The pollen rain which falls into a bog arises from two sources: a pollen rain from the local vegetation of the bog itself and one from the surrounding vegetation. When we are dealing with great bogs, the pollen produced by the vegetation of the bog itself will be mostly that of herbaceous plants, shrubs, and spores of the Bryophyta and the Pteridophyta. It is the rule rather than the exception that the bog will be treeless. The tree pollen in such a bog mostly takes its origin from the surrounding forests. It is a fortunate circumstance in a diagram that pollen of trees is separated from other pollen. However, one exception is seen in the way in which Iversen composes a diagram for late glacial times. This method, commonly used for late glacial times, embraces a pollen sum not only containing trees but also some herbaceous plants. The origin of the latter can, with some certainty, be accepted as from outside the bog. Therefore the local vegetation of the bog does not influence the percentages of tree pollen. The pollen sum thus comprises pollen of plants which grow under the same biotic conditions.