January 19, 1905J 



NATURE 



lo — the first satellite — and Jupiter in a period of 

 iih. 57ni. 22-6s. 



Following this discovery came the addition, to an already 

 numerous family, of the ninth satellite of Saturn, which 

 was found by Prof. W. H. Pickering. The search was 

 ■commenced in 1888 with the 13-inch Boyden telescope of 

 the Harvard College Observatory, but was not successful 

 in bringing to light any previously unknown attendant on 

 .Saturn. On the installation of the new 24-inch Bruce 

 telescope in the clear atmosphere of Arequipa the search, 

 which was photographic throughout, was renewed, and 

 on examining the plates taken on August ib, 17, and iS, 

 1H9S, Prof. Pickering was rewarded by the appearance 

 of a short trail which apparently partook of the planet's 

 motion among the stars, and was, therefore, to be con- 

 sidered as part of its system. The story of the subse- 

 quent doubts and difficulties has been too recently told 

 (Harvard College Annals, No. 3, vol. liii.) to need re-telling 

 here, but it may be recalled to mind that the subsequent 

 ■observations showed that the satellite revolves in an orbit 

 which is far more eccentric than that of any other satellite, 

 or of any major planet, in the solar system, and that its 

 motion in that orbit is opposite in direction to the orbital 

 motions of the remaining eight of Saturn's moons. Like 

 .the fifth satellite of Jupiter, this object can only be 

 observed visually with the largest telescopes and under 

 the best conditions. .\s a matter of fact, it was not seen 

 until its position was accurately known, and even then 

 Profs. Barnard and H. H. Turner, using the 40-inch 

 refractor at Veri<es Observatory, in .August last, could not 

 feel certain that they had really observed the object which 

 had up to tliat time remained invisible to human eyes. 



Whilst our knowledge of the most recently discovered 

 satellite is as yet very scanty. Prof. Perrine's message 

 tells us that on January 4 its position angle was 269°, and 

 the dailv rate of its apparent approach towards Jupiter 

 was 45", i.e. about 100,000 miles. 



The magnitude, 14. ascribed to it is one magnitude 

 fainter than that of Barnard's fifth satellite, and this 

 primarilv suggests that the diameter may be less than 

 that of the fifth, although a smaller reflecting power, or 

 "albedo," may account for the relative faintness. Its 

 ilistance from Jupiter on January 4 would probably be about 

 !■ million miles. The statement that the motion was " re- 

 ircgrade " refers, of course, lo the apparent motion in the 

 -U\ . and must not be confounded with a retrograde orbital 

 lion similar to that followed bv Phoebe, Saturn's ninth 



I. Mite. ' W. E. R. 



ITMOSPHERIC AND OCEANIC CARBON 

 DIOXIDE. 

 T^HE carbonic acid of sea-water is usually supposed to 

 be present in combination with certain bases, which 

 constitute the alkalinity of the water, partly in the form 

 of normal carbonate and partly in the form of bicarbonate, 

 the total amount present being insufficient to convert the 

 whole of the base into the bicarbonate. Thus the water 

 of the North .-Atlantic has been found to contain 49 c.c. of 

 carbonic acid gas per litre, whilst 54 c.c. would be required 

 to convert the base completely into bicarbonate. That this 

 view is not quite correct has been shown by t>r. A. Krogh, 

 of Copenhagen, in a series of investigations on the carbon 

 dioxide of the air and ocean.' 



The reaction between carbonic acid and a normal 

 carbonate to form a bicarbonate is, like so many chemical 

 reactions, reversible, and equilibrium is established while a 

 certain amount of the carbonic acid is still free. This free 

 carbonic acid exerts a definite gaseous pressure, which varies 

 with the total amount of carbon dioxide present and with 

 the alkalinity of the water. This pressure can very readily 

 be determined by simply shaking the water with a small 

 volume of air and then ascertaining by direct analysis the 

 pressure of the carbon dioxide in this air, which is, of 

 course, equal to the pressure of that in the water, since 

 the two have heen hrought into equilibrium by the shaking. 

 This process gives excellent results both with fresh- and 

 sea-water, and can be carried out very rapidly by the aid 

 1 '' Meddelelser om Gronland," vol. xxvi. pp. 333, 409. 

 NO. 1838, VOL. 71I 



of the apparatus of Haldane or Petterson and Sonden for 

 the estimation of small quantities of carbon dio.xide. As 

 the result of a careful study of the behaviour of sea-water 

 in this respect, it appears that a comparatively large amount 

 of carbon dioxide may be absorbed, whilst the correspond- 

 ing pressure only undergoes a very small absolute change, 

 provided that the alkalinity remains constant. .A water, 

 for example, which has the alkalinity 23, and contains 

 367 c.c. of carbon dio.xide per litre, is capable of absorbing 

 43 c.c. of the gas per litre, whilst the pressure, measured 

 as described above, only rises from 0015 per cent, to 00295 

 per cent, of an atmosphere. This means that the air shaken 

 up with the original water would be found to contain 1^5 

 parts of carbon dioxide per 10,000, whilst after the further 

 absorption the air similarly treated would contain 295 parts 

 per 10,000. 



Owing to this pressure of carbon dio.xide constant inter- 

 change takes place between every water surface, whether 

 of sea-water or of fresh-water, and the air above it, result- 

 ing in evolution from the water or absorption by it accord- 

 ing as the pressure of carbon dioxide in the water or the 

 air is the greater. The effect of this is that the ocean acts 

 as a regulator on the amount of carbon dioxide in the air, 

 tending to compensate for any deviation from the normal 

 proportion. The pressure of carbon dioxide in the air is at 

 present about 003 per cent, of an atmosphere (3 volumes 

 per 10,000), the absolute amount in the whole atmosphere 

 being calculated as 24x10'" tons, whilst the quantity con- 

 tained in the entire mass of the sea may be taken as twenty- 

 seven times as great as this. 



In order to increase the proportion of the atmospheric 

 carbon dioxide to 004 per cent, it would be necessary, of 

 course, in the first place to add one-third of the amount 

 already present. The pressure thus attained would, how- 

 ever, be gradually decreased by absorption by the sea, and 

 it follows from the author's experiments that in order to 

 bring the ocean into equilibrium with the altered atmo- 

 sphere a further addition of twice the amount originally 

 present would be required, a total change involving the 

 production of 5-6x10'" tons of carbon dioxide! A calcula- 

 tion of this kind goes far to explain the constancy of com- 

 position of the atmosphere, which at first sight appears so 

 remarkable, and to indicate the enormous changes required 

 to produce any considerable variation in it. 



The interchange of carbon dio.xide between sea and au-, 

 moreover, is by no means a slow process, but takes place 

 with remarkable rapidity. Thus a pressure difference 

 between sea and air of only 0001 of an atmosphere, i.e. the 

 presence in the air of an additional o^i part of carbon 

 dioxide per 10,000, leads to the absorption of 0525 c.c. of 

 this gas per square centimetre of ocean surface per year, 

 or a total annual absorption of 3^85Xio'' tons. 



The author considers from this point of view the effect 

 on the composition of the atmosphere of the combustion of 

 coal, which annually throws into the air about one- 

 thousandth of the carbon dioxide already present in it, so 

 that, apart from any regulating action of the sea, in a 

 thousand years — if the coal lasted — the percentage propor^ 

 tion would be doubled, rising from 3 to 6 volumes per 

 10,000, and rendering the air almost unfit for continued 

 respiration. Before the proportion rose to 31 volumes per 

 10,000, however, the sea would be able to absorb the gas 

 as fast as it was produced, and, owing to the large volume 

 required to bring the ocean water into equilibrium with the 

 air, it is probable that at the expiration of the thousand 

 years the proportion of carbon dioxide in the air would not 

 be more than 35 volumes per 10,000. 



Many other interesting questions of great importance in 

 the economy of nature are capable of being attacked from 

 this point of view and subjected to experimental investi- 

 gation. Such are the rate of deposition of calcium 

 carbonate from hard waters, the rate of solution of limestone 

 and chalk in natural waters, the absorption of carbon 

 dioxide by rocks and soils, &c. 



On the great question as to whether the production ol 

 carbon dioxide is on the whole greater or less than its de- 

 composition nothing certain is known. Indications are not 

 wanting, however, that this constituent of the atmosphere 

 is increasing in quantity. The chief evidence to this effect 

 is derived from the fact that over the sea the pressure of 



