November 24, 1923 



NA TURE 



765 



tained, and may result in agitated^ unnatural move- 

 ments, giving a totally false impression of the true 

 natural characteristics of the animal. 



This has been demonstrated on several occasions in 

 making records of some of the shyer animals in the 

 collection of the Zoological Society of London. In 

 obtaining successful records of the rare and interesting 

 maned wolf of South America, the kinematograph 

 apparatus had to be set up again and again and the 

 mechanism run without any film, before the animal 



could be induced to tolerate its presence or move about 

 in a natural manner. On the other hand, the anthropoid 

 apes, like children of the human race, are so intensely 

 curious and interested in what is going on, that they 

 will cease playing about in their normal fashion until 

 they have been permitted thoroughly to examine the 

 apparatus and satiate their curiosity. 



We are indebted for the accompanying illustrations 

 to the directors of British Instructional Films Ltd., 

 26-27 D'Arblay Street, Wardour Street, London, W.i. 



Meteorological Perturbations of Sea-Level. 



By Dr. A. T. Doodson. 



IT is always understood that the predicted heights 

 of high and low tidal water do not take into 

 account the variations in the height of the sea due to 

 wind and to air-pressure, and that the errors due to 

 these causes may be of considerable magnitude. With 

 the large ships that are now in common use the margin 

 between sea-bottom and ship-bottom is small, and 

 since many of the largest ports in the world are situated 

 in comparatively shallow water, navigation, both in 

 channels and into dock, is carried on only with constant 

 reference to the state of the tide. A particular example 

 of the problem is that of loading a vessel in dock : how 

 much cargo must be left on the quay-side so as to leave 

 sufficient clearance for the vessel to get safely out of 

 dock ? The cargo so left has afterwards to be trans- 

 ported by lighter, with consequent increase of expense. 

 If the tide is lower than was expected there is increased 

 risk to the vessel, and if the tide is higher than was 

 expected needless expense has been caused through 

 leaving cargo to be transported by lighter. It is 

 therefore obvious that a forecast of the effects of wind 

 and air-pressure on sea-level and tides would be of very 

 great advantage to navigators in and near a port, and 

 for this reason much attention has recently been given 

 to the subject. 



The effects of wind and air-pressure on sea-level are 

 also important factors for engineers engaged in the con- 

 struction of harbour works. Again, they are of im- 

 portance in connexion with geodetic surveys, since 

 sea-level is an obvious datum from which to take 

 measurements ; but it has been shown by the Ordnance 

 Survey (" Second Geodetic Levelling of England and 

 Wales," p. 34) that measurements by levelling gave 

 mean sea-level at Dunbar and Liverpool respectively 

 0-8 ft. and 0-4 ft. higher than at Newlyn. These 

 discrepancies cannot be attributed wholly to errors of 

 levelling, and there is reason to believe that part of 

 the explanation is connected with climatic causes. 

 Investigations as to the variation of sea-level with wind 

 and pressure have been made by Mr. H. L. P. Jolly, 

 of the Ordnance Survey, and are referred to below. 



Most investigations on this subject have been con- 

 cerned with air-pressure and not with wind, the sea 

 being regarded as a negative water barometer ; the 

 " constant " for the barometer, however, varies much 

 from place to place, and even according to the numerical 

 method used in obtaining it. A British Association 

 Committee in 1896 reported that the effects of wind 

 and pressure were real, but no law could be established ; 



NO. 2821, VOL. 112] 



the methods of investigation, however, were faulty, 

 A successful reduction to law for both wind and pressure 

 in connexion with tides at Ymuiden was published by 

 Ortt in 1897, his method being to collect together 

 observations for given ranges of values of pressure, 

 wind direction, and strength. This method has been 

 used, in essence, by other continental workers. Prof. 

 R. Witting (Bulletin de la Societe de Geographic de 

 Finlande, Fennia, 39, No. 5, 1918) has elaborated a 

 method of comparing the gradients of the sea-level in 

 the Baltic Sea with the gradients of the pressure- 

 system over the sea ; this method is strictly in accord- 

 ance with theoretical considerations, but it requires 

 a large number of observing stations, and is most 

 confidently applied to narrow seas. His use of pressure 

 gradients instead of wind-strength and direction of 

 wind is very commendable, and was utilised by Mr. 

 Jolly in his investigations, leading to the simple 

 formula 



^= k(B - B) -I- A(E - E) -f- /x(N - N), 



where ^ is the meteorological disturbance of sea-level ; 

 B, E, N are the values of the local barometric pressure 

 and its gradients to the east and north respectively ; 

 bars denote means in the interval of time con- 

 sidered, and K, X, /A are constants determined from 

 observation. 



This formula is valuable because it lends itself very 

 easily to numerical methods, and fairly accurate values 

 of the constants may be obtained from observations 

 extending over only a month, whereas an elaborate 

 method like Ortt's requires far more observations and 

 much more labour. It represents the perturbations 

 of mean sea-level with a fair degree of accuracy. 



The formula has been used extensively at the Tidal 

 Institute at Liverpool, and has yielded some very 

 interesting results. It is easy to deduce from it the 

 direction of the most effective wind for raising sea-level 

 at the place considered, and this has been evaluated 

 from a month's observations at various places on the 

 British coast, the results being illustrated in Fig. i. 

 The arrows give the direction from which the most 

 effective winds blow, and the lengths of the arrows are 

 proportional to the effects for a given strength of 

 gradient in the most appropriate direction. Many 

 previous investigators dealing with the perturbations 

 of mean sea-level on the Continental coast of the North 

 Sea have found that the most effective winds for 

 raising sea-level there arc those which blow towards 



