NA TURE 



97 



THURSDAY, DECEMBER 4, i9°2 



DR. NANS EN'S OCEANOGRAPHY OF THE 

 NORTH POLAR BASIN. 

 The Norwegian North Polar Expedition, 1893- 1896, 

 Scientific Results. Edited by Fridtjof Nansen. Vol. iii. 

 Published by the Fridtjof Nansen Fund for the Ad- 

 vancement of Science. Pp. xii + 428 ; 88 Plates. 

 (London : Longmans, Green and Co., 1902.) 



THIS volume discusses at great length the not very 

 numerous observations on the physical conditions 

 of the water in the Arctic Sea which were made on board 

 the Fram during her memorable drift. The matter is 

 dealt with in two memoirs, both by Dr. Nansen, the first 

 on the oceanography of the North Polar Basin, the 

 second on hydrometers and the surface tension of liquids. 



The second memoir may be considered as an appendix 

 to the first, a great part of which also deals with methods 

 and discusses experiments carried out after the return of 

 the expedition. 



We must, in the first place, express gratitude to Dr. 

 Nansen for his choice of the English language as the 

 medium by which the official account of his expedition 

 is made public ; a choice which brings at least this 

 reward, that his volumes will find their natural place 

 beside the magnificent record of the Challenger expedition. 



It is natural to look at the contents of this volume 

 from the separate points of view of methods and 

 results. As to methods, it must be confessed that 

 the key-note is one of regret mingled with hope. 

 "What might have been" if the experience gained 

 during the voyage had been available before it started 

 is insisted upon almost too much ; and we are 

 sometimes tempted to forget, in dealing with the results, 

 that the high degree of possible precision lay latent in 

 the methods and was not actually attained. Dr. Nansen, 

 indeed, loses no opportunity of disclaiming higher accu- 

 racy than his instruments as they were handled could 

 give, and we have rarely seen a scientific man more 

 candid in blaming himself for neglecting precautions 

 which, after all, few specialists in his subject, if any, 

 thought necessary at the time of his departure. 



The methods most fully dealt with are those of deter- 

 mining the temperature and density of sea-water at 

 various depths. Two methods of observing the tem- 

 perature were usually used, one involving the use of 

 the Pettersson insulating water-bottle and the other the 

 reversing thermometer on Negretti and Zambra's principle. 

 The insulating water-bottle is an apparatus which encloses 

 at a given depth a large sample of water, partly in a 

 central tube, partly between the members of a series of 

 concentric outer tubes having no communication between 

 them. It is thus apparent that before the temperature 

 in the inner tube can change, the temperature of each 

 of the concentric water-jackets must be changed to a 

 greater degree. The important question is how long a 

 time may be allowed to elapse before the temperature in 

 the inner tube changes by an amount appreciable on the 

 thermometer. The actual water-bottle used on the voyage 

 was lost, and its constants could not be tested. Two other 

 water-bottles of apparently identical construction gave on 

 NO. 1727, VOL. 67] 



examination distinctly different results, and differences 

 were also found when the external temperature was 

 changed in various ways. It seems probable, however, 

 that with one of the improved water-bottles on this prin- 

 ciple, the constants of which have been elaborately 

 determined before use, it may be possible to arrive at 

 temperature readings to something approaching the 

 hundredth of a centigrade degree ; but this will be the 

 result of applying many corrections, some of considerable 

 magnitude. The question is not definitely settled, but 

 it is clear at least that no error exceeding one-tenth of a. 

 centigrade degree should occur with a water-bottle of 

 this description when skilfully handled. The peculiar 

 virtues and failings of the Negretti and Zambra reversing 

 thermometer are familiar to all who have had occasion 

 to use that beautiful but capricious instrument for deep- 

 sea work. Dr. Nansen discusses the various corrections 

 which have to be applied, and in their case also he 

 suggests improvements which should lead to increased 

 precision and certainty. For the Fram expedition, the 

 reversing thermometers gave readings the error of which 

 in almost all cases could be guaranteed not to exceed 

 ±o D 'iC. 



The determination of density received a great deal of 

 attention, for just as Dr. Nansen believes that observ- 

 ations of sea-temperature should be correct to one- 

 hundredth of a centigrade degree, so he believes that 

 the density of sea-water should be obtained with a degree 

 of precision sufficient to indicate a difference as small as 

 one unit in the fifth place of decimals, when the density 

 of distilled water is taken as unity. But in the case 

 of density, as in that of temperature, the results of 

 the Fram did not realise the desired ideal. The in- 

 vestigation as to why this was so occupied much time 

 and led to the interesting study in hydrometry which 

 forms the second paper. We cannot follow the experi- 

 ments in detail, and it must suffice to say that the 

 villain of the piece was finally exposed and found to be 

 grease. Variations in surface tension due to observing 

 with unwashed hands or wiping the hydrometer with a 

 towel not above reproach led to the most distressing 

 irregularities. It was shown that an ordinary stem- 

 reading hydrometer could give good results if the glass 

 was perfectly clean and the surface of the water swept 

 free from impurities by careful brushing with a piece of 

 clean paper. But better results can be obtained by using 

 hydrometers of total immersion, which act in the heart of 

 the liquid untroubled by surface tension or capillarity. 

 Dr. Nansen finds the best results when using a jar with 

 vacuum jacket on the principle employed by Prof. Dewar 

 for handling liquid air, thus preventing change of tem- 

 perature by radiation. The stemless hydrometer is 

 weighted so as just to float in the sample of sea-water 

 the temperature of which is read by means of a very 

 sensitive thermometer. The temperature is then raised 

 by stirring with a tube containing warm water, and the 

 exact temperature at which the hydrometer begins to 

 sink is noted ; a tube of cold water is then used as a 

 stirring rod until the hydrometer begins to rise again, and! 

 thus by one or two operations the temperature at which 

 the sample has the precise density of the weighted hy- 

 drometer can be ascertained with high precision. Other 

 methods of determining salinity were tried onboard, the 



F 



