On the Surface Tevvperafurc of A'orfh Sea and North Atlantie. 5 



proceed with confidence to " interpolate," to fill in the missing fact 

 or figure in harmony with the others, and so to fill up the gap. It 

 goes without saying, of course, that this procedure of " interpolation " 

 is to be employed with all possible care and caution : it is to be 

 used but not abused ; and it must be more and more dispensed with 

 as our work proceeds, as knowledge accumulates, and as we aim at 

 greater and greater accuracy in detail. 



Throughout all our study of temperature, or of other physical 

 conditions of the sea, we are on the look out for abnormal conditions, 

 which may help us to understand and to explain such abnormal 

 conditions of fish-life as appear from time to time, and aff'ect seriously 

 the fortunes of the fisherman. History tells us that, among various 

 tishes (the herring in particular and also the haddock and a number 

 nf others), there have been great temporary fiuctuations, sometimes 

 in the direction of scarcity, sometimes in the direction of extra- 

 ordinary plentifulness. Such fluctuations were experienced even 

 I in days when the toll taken of the stock by man was so small 

 that of itself it could not possibly affect the common stock ; and 

 even nowadays, when the catch is multiplied many fold, and is 

 no longer insignificant, we can still in many cases recognise years 

 of plenty alternating irregularly with years of scarcity, and the 

 reasons for this alternation we are bound, apparently, to look for 

 I in some abnormal changes of physical conditions. But it is obvious 

 * that w^e have no right to speak of abnormalities, no right to assert 

 that here or there the sea is warmer or colder, more or less salt, than 

 usual, until we have a very thorough knowledge indeed of the normal 

 ronditions, and of the periodic changes which are a part of the 

 normal phenomenon. And so it comes to pass that the very kind 

 of knowledge which, for our practical ends, we most desire to 

 have (namely, the knowledge of abnormal conditions), is bound to 

 be postponed till we have prepared the way for it by years of 

 work, employed in gathering knowledge which is of little practical 

 utility in itself, but which is the indispensable preliminary step to 

 that other knowledge which we have every hope of turning to 

 practical account. 



Among other charts and diagrams, the present paper contains a 

 series of twelve charts, in illustration of the mean or normal tem- 

 perature of the North Sea and of the other waters round our coasts, 

 for each month of the year. For several years past I have had these 

 charts in preparation, and I have more than once been about to 

 publish them, with such additions or corrections as I was able to 

 make for the time being. But again and again I put the task aside, 

 waiting for still more information. Now, however, it seems to me to 

 be time to publish these charts ; not that they are to be considered in 

 any way complete or final, but simply because they now give, as it 

 seems to me, just such a representation of our seasonal surface- 

 temperature phenomena as ma}^ usefully serve a temporary purpose, 

 and be gradually improved and corrected by those workers who are 

 making a special study of our sea temperatures in particular areas. 



The observations on which this paper and the accompanying 

 charts are based are drawn from many sources, of which the following 

 are the chief : 



1. The mean surface temperatures for the Atlantic, in the pos- 



