10 Transactions of the South African Philosophical Society. 



phenomenon subject to local influences. For the purpose of com- 

 parison with the Kenilworth numbers I venture to quote the con- 

 stants given by Eiggenbach for Basle and Vienna." Those for 

 Basle are computed from hourly statistics of 1473 hours of thunder- 

 storms observed between 1826 and 1888, supplemented by 171 

 others of which the duration and time of day is somewhat doubtful.. 

 Those for Vienna are quoted from Hann. Both the Basle and 

 Vienna amplitudes are reduced to proportional parts of a thousand, 

 the Kenilworth amplitudes are not. Eiggenbach notes the near 

 resemblance of certain of the phases of the two European stations. 

 As it happens, the first two phases of the Kenilworth formula fall 

 about as much later than Basle as Basle is than Vienna. Some 

 correction, however, will have to be made for a departure from local 

 time at Kenilworth of about seven minutes, the longitude of this, 

 place being about 24° 27' E., while the time system is reckoned from 

 the meridian of 22° 30' E. Whether the Basle and Vienna con- 

 stants refer to local or zone time I do not know T . 



It is important to determine in what manner particular wind- 

 directions are associated with the monthly rainfalls. To do this, 

 the monthly average number of hours of wind blowing in each 

 direction during the six years 1897-1902 has been first determined 

 from the automatic records taken at Kenilworth. The departures, 

 from these monthly means are then tabulated for each month, and 

 it is also marked whether the concomitant rainfall is greater or less 

 than the mean. Such departures of monthly wind-direction from 

 the normals are then arranged into two sets according as the 

 accompanying rain was greater or less than the normal rainfall.. 

 The result is condensed into Table 21, for each quarter of the com- 

 pass. A plus sign indicates that in any set, for any month, there 

 were more hours of wind than the average, a minus sign that there 

 were less hours. For example, in January, with deficient rainfall, 

 the number of hours of wind from N.-E.N.E. was thirteen hours 

 short cf the average ; whereas when the rain was in excess the 

 wind from the same quarter was twenty-four hours greater than the 

 average. It appears from this that, taking the year as a whole, in 

 the more rainy months the winds with an easterly component 

 increase ; those from the west increasing in dry months. In 

 individual months there are some exceptions to the rule : in Decem- 

 ber, e.g., a rainy month and an increase of wind from the north- 

 west go together ; in October the increase is from the south-west ; 

 in June and July the increase is with the winds having a southerly 



* A. Eiggenbach, " Resultate aus 112 jahrigen Gewitteraufzeichnungen in 

 Basel," T. viii., 1889. 



