622 



NA TURE 



[October 25, 1900 



numeral may be connected with the Australian. On the opposite 

 New Guinea coast the numerals for ' one ' are very ditTferent to 

 the Saibai urapon. They are : natnbi (Morehead River), atnbior 

 (Wasi Kasa), iarangesa (Bugilai Tribe), iepa (Kunini Tribe), 

 alanok (Dabu Tribe), netat (Murray Island), nao (Kiwai Island), 

 vionoii (Purari Delta), farakeka (Papuan Gulf), aia (Yule 

 Island). Mr. Mathew asks us to believe that the Kalkadoon 

 numeral " hiadi" (two) is a Melanesian numeral used in 

 Australia 150 miles south of the Gulf of Carpentaria, when the 

 only comparable form in Melanesia is the Duke of York Island 

 rnadi (second). This is a flagrant example of the method 

 adopted, though disavowed, by Mr. Mathew. 



The point missed by Mr. Mathew in discussing tlie phrase 

 " ori kaiza," which I called mongrel, is that he has no proof 

 thui' ori' and ' w'oi' are the same words. Considering that 

 the Gulf tribes have a word (in'va) for ' cassowary ' (the emu is 

 not found), and that nowhere else in New Guinea is there a 

 word similar to orz, meaning ' bird,' and also that it requires a 

 Torres Straits word to give 'ori' an Australian meaning, it is 

 highly improbable that it can explain words like waitch, wadgie, 

 tvarritch for "emu." The Saibai for 'big bird' is "■ koi urtii,' 

 for 'cassowary,' ' satmi.' 



In his letter, Mr, Mathew objects to my calling comparisons 

 of Australian with Malay and New Hebridean " absurd and 

 misleading." He suggests, without any warrant, that I find 

 the absurdity in analogies between Malay and New Hebridean, 

 whereas I have directly affirmed the connection of Malay with 

 Melanesian (including New Hebridean) in the Joicrnal of the 

 Polynesian Society for 1896. The real absurdity is that of 

 supposing that there is a relationship between Australian, Malay 

 and New Hebridean, because " the terms for father, skin are 

 the same." 



In conclusion, I must again express my regret that Mr. 

 Mathew should regard my criticism of his work as ' mere fault- 

 finding and ridicule.' I have studied these languages for many 

 years (without postulating theories), and have much material yet 

 unpublished. To point out the weakness of Mr. Mathew's 

 argument, with regard to method and deductions is fair criticism, 

 and should not lead to a charge of unsoundness and inaccurate 

 knowledge. If, as Mr. Mathew states, his materials are im- 

 perfect, why found a theory upon them ? 



218, Balfour Road, Ilford, Essex, Sidney H. Ray. 



October 14. 



RECENT AND PROPOSED GEODETIC 

 ME A S UREMENTS. 



IN the history of geodesy and the discussion of the 

 problem of the figure of the earth, the measurement 

 of three arcs of meridian stands out conspicuously. These 

 are the Peruvian and Lapland arcs in the Northern 

 Hemisphere, and that of Lacaille in the neighbourhood of 

 the Cape of Good Hope. Those who took part in the 

 original measures worthily distinguished themselves, 

 but it is inevitable that with the demand for greater 

 accuracy in geodetic inquiries the necessity for repeating 

 the ancient measures should be acknowledged. Without 

 experience and with imperfect instruments, it is remark- 

 able that the old astronomers accomplished what they 

 did. To determine the approximate figure of the earth, 

 and to derive a fairly accurate value of the compression 

 was no mean achievement. But Lacaille's arc of meridian 

 soon fell into disrepute, and other measures have had to 

 be substituted by Maclear and others. Svanberg 

 repeated the original work of Clairant and his colleagues, 

 in Lapland, soon after its completion, and within these 

 last few years a new determination of the Peruvian arc 

 has been imperatively called for. There is no doubt but 

 that the French nation, who have so honourably distin- 

 guished themselves in the difficult task of geodetic 

 measurement, will be able to undertake this work, and 

 increase their scientific reputation. From 1734, when 

 the enlightened Government of the day undertook the 

 measurement of two of the aforementioned arcs, down to 

 our own time, when we have seen the Mediterranean 

 successfully bridged by a geodetic survey, and the interior 

 of Africa connected in one unbroken chain with our own 



Shetland Islands, French inen of science have played a 

 conspicuous part in all questions connected with the true 

 figure of the earth. At a moment therefore when the re- 

 measurement of the famous arc of Peru has been forced 

 upon us, It would have been with a feeling akin to dis- 

 appointment if we did not find the French nation eager 

 to repeat the historic work with all the skill that long 

 experience has suggested, and all the accuracy that 

 modern science demands. The report of a committee of 

 tlic French Academy of Sciences, however, assures us 

 that the ardour displayed by the French in the past is nO' 

 whit abated, and that though the sacrifice of time and 

 money is considerable, these drawbacks will be cheerfully 

 borne. For the arc in Lapland another has been substi- 

 stuted on the Island of Spitsbergen, and the necessary 

 work of triangulation has been for some time quietly 

 carried on under the auspices of the Russian and Swedish 

 Governments ; while at the same time, the reports of Her 

 Majesty's Astronomer at the Cape of Good Hope tell us 

 what is being done in the way of supplementing 

 Lacaille's arc in the Southern Hemisphere. With this 

 activity before us, it seems a fitting opportunity to com- 

 pare the aims and the motives that inspired those who 

 inaugurated the earliest geodetic expeditions with those 

 that will influence and guide the latest surveys in the 

 same or similar regions. 



In the middle of the last century a distinct issue was 

 presented to the scientific mind. Then the general 

 figure of the earth was an undetermined problem, and 

 whether the polar or the equatorial axis was the longer 

 was a vexed subject of controversy. It is amusing 

 enough now to read of the disputes between the partisans 

 of defective observations on the one hand, and the up- 

 holders of an incomplete theory on the other. To-day 

 no such broad issue is before us, the differences are of a 

 more subtle character, demanding great nicety of observ- 

 ation and more effective analysis. Then the true 

 difficulties of the problem were scarcely apprehended. 

 Only a iew years earlier, Fernel had attempted to deter- 

 mine the length of a degree by counting the number of 

 revolutions of his carriage-wheel between Paris and 

 Amiens, a method which recalls the earliest attempts of 

 Eratosthenes. Picard, it is true, had recognised the 

 necessity of employing means of greater accuracy, and 

 had taught the true principles of geodetic measurement ; 

 but the methods pursued by his descendants, and the 

 precautions taken to ensure accuracy and recognise the 

 surface to which the measurements are referred, con- 

 stitute almost a new science. For with accumulating 

 materials and greater experience, it has become necessary 

 to distinguish between three surfaces. First, the ellip- 

 soid of revolution, which corresponds most nearly to the 

 form of the earth ; secondly, the true geoid, that is to 

 say, the surface of equilibrium of water at rest under the 

 influence of centrifugal force, and the attraction of the 

 earth's mass ; and, thirdly, a corrected geoid, diftering 

 but slightly from the true, in which it is attempted to 

 eliminate the effect of unequal masses on the earth's 

 surface and in its interior. Theory shows that the true 

 length of the arc of meridian, measured on the corrected 

 geoid, will be given in terms of the measured base, if the 

 effect of local attractions has been correctly determined. 

 It is precisely in overcoming the difficulty of correctly 

 eliminating the effect of local attractions, and of reducing 

 the length of the measured base to the level of the water 

 surface of the geoid, that the measurements of this 

 century will mainly differ from those of the last. 



A preliminary survey of the district undertaken by 

 Captains Maurain and Lacombe has shown that the 

 country of Peru possesses peculiar difficulties for an 

 accurate geodetic survey. The levels vary very consider- 

 ably with the distance from the coast, while here and there 

 mountains of a volcanic character rise to a height of 6000 

 metres. But for the interest attaching to a historic site,. 



NO. I 61 7. VOL. 62] 



