PEPPEll TREE AS A CAUSE OF HAY FEVER. 161 



According to Dr. Scheppegrell's experience these densities 

 should be sutiicient to cause ha.y fever of considerable severity, 

 and, in point of fact, severe cases of hay-fever were prevalent 

 in the town at this time. As little or no other pollen was being 

 found on the plates it is to be presumed that the attacks were due 

 to pepper tree pollen. 



It is worthy of note that the occasion on which the largest 

 number of grains of pepper tree pollen was counted on a plate, 

 viz., '205, on October 28, works out, when expressed in Dr. 

 Scheppegrell's unit, to the lowest density shown in the table. 

 This is because of the longer period of exposure, viz. 74 hours. 

 It should be emphasised, however, that the densities are to be 

 regarded as purely local and not as being general for the whole 

 town. Thus, whilst a plate exposed at the E amblers' Club on 

 November 1 gave a density of pepper tree pollen represented by 

 206 (a very high figure), no pollen whatever was found on a plate 

 exposed for exactly the same period at Government Buildings, 

 less than half a mile distant. Also on November 20, whilst no 

 pollen was found on a plate exposed at Park Road, another ex- 

 posed at the Ramblers' Club for exactly the same period had 

 pepper tree pollen present in amount equivalent to the very high 

 density of 150. 



The irregular occurrence of pepper tree pollen on the plates 

 presents some curious anomalies. One would hardly expect to 

 find pollen in the air in wet weather, nor in the absence of wind, 

 but an examination of the tables will show that pepper tree 

 pollen was frequently not found on the pollen plates in weather 

 that seems suitable for the dispersal of pollen. Subsequent ex- 

 perience showed that the weather has a determining influence 

 not only on the dispersal of the pollen of the pepper tree, but also 

 on the opening of its flowers and the drying of the pollen : two 

 essential preliminaries for dispersal. In other cases the direction 

 and strength of the wind, and position of the pepper trees in the 

 vicinity in relation to the place of exposure of the pollen plates, 

 offer a sufficient explanation — see notes regarding sites of 

 exposure. 



The circumstance that there were no pepper trees to the west 

 of and near the place of exposure of the first five plates shown 

 in Table IV is, in the light of subsequent experience, a sufficient 

 reason for the absence of pepper tree pollen from these plates. 



Oilier Pollens: 



Regarding pollens other than that of the pepper tree: 

 "1. Pine pollen was found twice, both dates being in October. 

 As a cause of the epidemic it need not be further considered, as 

 its principal flowering season in Bloemfontein is the last fort- 

 night in September; and, except for isolated trees, it has ceased 

 to provide pollen before the hay fever epidemic commences. The 

 species is Pintis haicpensis. 



2. Chilianthus arborens Benth. — This pollen was also found 

 only twice. The plant is locally known as the Vaal Bos and is 



