64 FOREST DESTRUCTION IN NEW SOUTH WALES. 



very much within not very well defined limits, always swinging 

 back as it were but with the greatest possible irregularity, and 

 the changes are not coincident either in time or sequence with 

 any known local terrestrial or cosmical changes whicli might be 

 supposed to have produced them. The reason of our inability to 

 forecast the rainfall of any particular place, even for one year or 

 one day, must be that the causes which determine such rainfall 

 are infinitely numerous, and their interaction on each other so 

 complicated that results are very rarely repeated. If this were 

 not so, we would before now have gained even a little foothold of 

 safe standing ground. As the ages roll on, if civilization and 

 progress continue, gradually accumulated records and experiences 

 may enable the scientist of the future to do what for us is 

 impossible, but at present I am afraid the outlook is not hopeful. 

 Our work is to accumulate the records and experiences. 



Of course it may be contended that rainfall having been 

 determined by other remote and complicated causes, might be 

 slightly increased or reduced in quantity by the presence or 

 absence of growing trees at any particular place. To prove or 

 disprove this is impossible, until we are able to say beforehand in 

 the presence or absence of trees what would have been the exact 

 rainfall of any particular place on the earth's surface for any year 

 or any series of years. In the absence of any such knowledge 

 one way or the other, I think we may safely consider our 

 convenience, and disregard even this much modified claim, which 

 is by no means what is usually meant when people assert that 

 forest destruction causes drought. There is one line of inquiry 

 which I have often thought might possibly, if patiently and 

 laboriously followed up, lead to some practical result in the way 

 of enabling us to make some approach to the prediction of seasons 

 for any particular place. It has not yet as far as I know been 

 tried. With the most careful research, the finest instruments, 

 and the greatest intelligence which the nineteenth century has 

 produced, we cannot find for any given place in the varying 

 records of the occurrence of droughts and floods, any order or 

 sequence which will enable us to predict the season even for one 

 year or one day in advance, or to say whether the coming year 

 will be a year of drought or of flood ; yet there may be a way 

 different from that generally tried by which such knowledge 

 might in some measure be gained. The attempt to find a saros 

 for the seasons, like that discovered by the Chaldean priests for 

 the moon's changes, has always failed, and there does not seem to 

 be now even the slightest indication that such a thing will ever 

 be found ; but though we cannot predict the changes of the 

 seasons for a century in advance as the moon's changes of position 

 may be predicted, yet there may be a possibility of our attaining 



