171 The N.Z. Journal of Science and Technology. [Aug. 
period yields a rather unexpected result, as follows: Auckland, 43*87 in.; 
Wellington, 48*57 in. ; Dunedin, 38-07 in. 
Coming now to the question of the 
the results have been fully analysed i 
centages — 10, 20, 30, and 40 per cent. — 
of excess and of deficiency in rainfall 
have been determined (Table I). These 
results can either be expressed as a 
frequency of occurrence of such excess 
or deficiency per century, or as the 
average cycle or period of years which 
has elapsed between the occurrence of 
such years of deficiency or excess. For 
practical purposes the latter method of 
expression is the most useful, as it pro¬ 
vides a number of years within which 
the necessary financial provision for 
such lean or excess year has been neces¬ 
sary in the past. Table I gives mean 
cycles of years, which are obtained by 
dividing the whole period by the fre¬ 
quency of the occurrence of a rainfall 
of various percentages—10, 20, 30 ; and 
40 per cent.—above and below the 
average rainfall for each place. Thus 
in Auckland the rainfall has exceeded 
30 j^er cent, above the average six times 
in the sixty-eight years : that is to say, 
the mean cycle of 30 per cent, excess 
rainfall is 11-3 years for Auckland. 
Similarly, the mean cycles of 30 per 
cent, deficiency below the average rain¬ 
fall for Auckland has been 22-6 years, 
the rainfall less than this having occur¬ 
red in only three years out of the sixtv- 
eight. Averaging the corresponding 
cycles for each of the three stations, we 
find that an excess of 30 per cent, has 
occurred on the average in the three 
places on a cycle of 12-5 years, and a 
deficiency of 30 per cent, below the 
average rainfall has occurred on an 
average cycle of 16-2 years, and an ex¬ 
cess of 40 per cent, above the average 
has occurred on a mean cycle for the 
three stations of 51-7 years. These 
figures, of course, refer only to the 
past, but as referring to the past they 
are quite definite. They do not mean, 
of course, that a deficiency in the 
future of 30 per cent, is going to occur 
only once in 16-2 years or at an inter¬ 
val of 16-2 years. Such deficiency may 
occur at a very much shorter inter¬ 
val, or even in consecutive years, in 
the future. But with the actual 
figures of the past before him it will 
be possible for the farmer or the 
annual fluctuations of these figures, 
nd the frequencies of various per- 
Fig 
