514 REPORT— 1902. 



Address and the reports of committees. Those who framed such a rule must have 

 had some unfortunate idea that the dignity of tlie Chair might be endangered if 

 some criticism happened to be expressed in the discussion of the chairman's ad- 

 dress, or that the value of the report of a committee might be endangered by some 

 adverse comment coming from outside. But it seems to me that a scientific 

 society or association, and especially one framed on a democratic constitution, 

 ought not to take such a narrow and unscientific view. I can remember several 

 Presidential Addresses which might, and probably would, have given rise to most 

 instructive debates had the rule not existed. 



Reports of committees if not suitable for discussion should not be read at all ; 

 but if read they should be open to discussion. 



I hope that to-day you will not feel yourselves bound by ancient custom, but 

 in order that, at any rate, the more scientific portion of my contribution to our 

 proceedings, should not be stained by the suspicion of immaculate conception, I 

 ■will now ask the duly constituted President of our Section to take his proper place. 

 The question I wish to bring to your notice to-day is an old one : if two events 

 tappeu simultaneously, or one follows the other at a short interval of time, does 

 this give us any reason to suppose that these two events are connected with each 

 other, both being due to the same cause, or one being the cause of the other ? 

 Everyone admits that the simple concurrence of events proves nothing, but if the 

 same combination recurs sufficiently often we may reasonably conclude that there 

 is a real connection. The question to be decided in each case is what is ' sufficient' 

 and what is ' reasonable.' Here we must draw a distinction between experiment 

 and observation. We often think it sufficient to repeat an experiment three or 

 four times to estaljlish a certain fact ; but with meteorological observations the case 

 is different, and it would, e.g., prove verj' little if on four successive full moons the 

 rainfall had been exceptionally high or exceptionally low. The cause of the differ- 

 ence lies in the fact that in an experiment we can control to a great extent all the 

 circumstances on which the result depends, and we are generally right in assuming 

 that an experiment which gives a certain result on three successive days will do so 

 always. But even this sometimes depends on tlie fact that the apparatus is not 

 disturbed, and the housemaid has not come in to dust the room. Here lies the 

 diflerence. What is possible in a laboratory, though perhaps difficult, is not possible 

 in the upper regions of the atmosphere wliere some unseen hand has made a 

 clean sweep of some important condition. 



When we cannot control accessory circumstances we must eliminate them by 

 properly combining the observations and increasing their number. The advantage 

 does not lie altogether on the side of experiment, because the very identity of 

 condition under which the experiment is performed gives rise to systematic errors, 

 which 2Sature eliminates for us in the observational sciences. In the latter also 

 tlie great variety in the combinations which offer themselves allow us to apply the 

 calculus of probability, so that in any conclusion we draw we can form an idea of 

 the chance that we are wrong. Astronomers are in the habit of giving the value of 

 the ' probable error' in the publication of their observations. Meteorologists have 

 not adopted this custom, and yet their science lends itself more readily than any 

 other to the evaluation of the deviations from the mean result on which the 

 determination of the probable error depends, AVe look forward to the time when 

 weather forecasts will be accompanied by a statement of the odds that the pre- 

 diction will be fulfilled. 



The calculation of the probability that any relationship we may trace in 

 diflerent phenomena indicates a real connection seems to me to be vital to the 

 true progress of meteorology, and although I have on previous occasions ' already 

 drawn attention to this matter I should like once more to lay stress on it. 



The particular case I wisli to discuss, though the methods are not restricted to 

 this case, is that in which one of the two series of events between which relation- 

 ship is to be established has a definite period, and it is desired to investigate the 

 evidence of an equal period in the other series. 



' CamhriHgc Phil. Trann., vol. xviii. p, 107, 



