February i8, 1909J 



NATURE 



459 



1 do not know in what way divers prepare themselves for 

 such feats, but presumably it is by a preliminary forced 

 breathing- only, without oxygen inhalation. Hence this 

 record is probably comparable with my record of 4m. 53., 

 and in that case it follows that forced breathing, together 

 with oxygen inhalation, might enable some individuals to 

 stay under water for nine or ten minutes. Moreover, they 

 could achieve such times without any risk of loss of con- 

 sciousness. Even at the end of my eight minutes' record 

 the air in my lungs still contained 46 per cent, of oxygen, 

 or three times the normal amount. 



The practical applications of this method of forced breath- 

 ing and o.xygen inhalation are obvious. Prof. Herdman 

 states (Report of Ceylon Pearl Oyster Fisheries, part i., 

 p. (13 ; part ii., p. 13) that the maximum time the best 

 pearl-divers (the Arabs) remain under water is, in his 

 experience, only ninety seconds, whilst the Tamil and 

 other divers vary from thirty-five to fifty seconds. Of 

 course, one would not for a moment expect them to attain 

 the times above mentioned, as they are performing violent 

 muscular work whereby the rate of production of carbon 

 dioxide by the body is greatly increased. Still, there is 

 little doubt that if they performed about two minutes' 

 forced breathing, and took a single deep breath of oxygen 

 at the end of it, they could, without risk, double or treble 

 their average time under water. This might be of especial 

 value to them when fishing in the deeper waters. Prof. 

 Herdman says that while the usual limit of the divers is 

 about nine fathoms, exceptional divers could go to fifteen 

 fathoms, " but they had barely time to secure a single 

 handful of the bottom before having to come up in an 

 exhausted condition." The method might also be of value 

 to sponge-divers, and to some extent also for rescue work 

 in mines and drains poisoned by foul air, when proprr 

 rescue apparatus was not available. 



22 \orham Road, Oxford. II. M. \'i.rno\. 



The Isothermal Layrr of the Atmosphere. 



The difference of opinion between Mr. Hughes and 

 myself apparently comes to this; he considers (February 11, 

 p. 429) that radiation plays an important part in the 

 temperature that is recorded by meteorographs sent up 

 with a balloon, and I think that, save in exceptional circum- 

 stances, radiation may be neglected. We are agreed in 

 stating that the temperature of the metal strip can only 

 be altered by contact with the air and by radiation, and 

 the only question is the relative values of these two causes. 

 Furthermore, I gather that Mr. Hughes thinks that 

 whether the ascent be by night or day, after a certain 

 height the temperature is unduly raised by radiation from 

 what he calls the hot planet. 



Now, first, the thermograph is made of polished metal, 

 and is protected by a polished metal case, and it is well 

 known that a polished metal surface is not susceptible to 

 radiation. One need only mention the double vacuum 

 bottle in which liquid air is kept, the commercial " thermos 

 llask." Loss or gain of heat by radiation is practically 

 excluded by silvering the internal surfaces. 



Secondly, it must surely be admitted that radiation must 

 be very different by night from what it is by day. It is 

 true that the sun subtends but a small solid angle, and the 

 earth an angle of nearly 2ir, but the power of radiation 

 varies as the fourth power of the absolute temperature. 

 In saying that all radiation was insignificant compared 

 with that of the sun, I was thinking of ordinary experi- 

 ence. In the tropics a man protects himself against the 

 sun ; to quote a very ancient writing, " there is nothing 

 hid from the heat thereof." There are places in the high 

 valleys of Switzerland where in calm, sunny weather a 

 person may sit out of doors in the sun in perfect comfort, 

 although the country round is deeply covered with snow 

 and the temperature is far below the freezing point. On the 

 other hand, in the Arctic and .Antarctic winter it is pro- 

 lection from the wind that is sought ; all accounts agree 

 that if there be no wind extremely low temperatures are 

 not unpleasant, and Toss of heat by radiation is not feared. 



But Mr. Hughes admits that radiation is stronger by 

 day, and says that if it is not apparent on the trace it 

 must be because the traces differ so much inter sc. This 

 is a question of fact, and I can only refer him to the pub- 



NO. 2051, VOL. 79] 



lished records and to my previous statement that traces 

 made in the day do not differ from those made at night. 

 There is a striking similarity about the general form of 

 all the traces, excepting those obtained in the daytime 

 from a balloon which did not burst. 



With regard to the vertical speed, we know that the 

 time occupied in falling is about half that occupied in 

 rising, because we have been able to ascertain by observa- 

 tions with a theodolite that the horizontal distance passed 

 over during the fall is about half that passed over during 

 the rise. We do not now use any parachute, and we 

 used not to use one of more than i sq. foot area. The 

 cross-section of the balloon before bursting is probably 

 25 sq. feet to 30 sq. feet. Inasmuch as at 20 km. height 

 the air density is only one-si.xteenth that at the ground- 

 level, the initial rate of fall will be four times the final 

 rate, and must therefore be greatly in excess of the rate 

 of ascent. Notwithstanding this, the temperatures recorded 

 are to all intents and purposes identical. Differences ex- 

 ceeding 3° C. between the up and down trace at the same 

 height are very rare ; 6° C. is the maximum recorded, and 

 anything more than 4° is only produced by change of level 

 of an inversion surface in the lower strata occurring during 

 the ascent. W. H. Dines. 



Barometric Oscillation. 



In N.\ture of December 3, 1908 (p. 130), Mr. Dines, in 

 reference to a previous note upon the semi-diurnal baro- 

 metric oscillation, gave as his opinion that the semi- 

 diurnal temperature oscillation is the result of pressure 

 variation. 



In connection with this question, it seems to me of some 

 value to give here a couple of results derived from the 

 Batavia observations. They are related to pressure oscilla- 

 tions of extra-terrestrial origin, like the semi-diurnal varia- 

 tion, and show a pressure change followed by a change of 

 temperature. 



.At Batavia the well-known barometric oscillation with 

 a period of 3^ years is very definite. It is followed by an 

 equally distinct temperature oscillation of the same period. 

 The difference in time is 6j months. The temperature, 

 moreover, shows the remarkable fact that the seven-yearly 

 means from 1871-1905 are increasing regularly from 

 o°o8 C. to o°'io C. every seven years, so the temperature 

 of the air increases by about o°-oi a year. 



In the second place may be mentioned the influence of 

 the 26-day period of the sun's rotation on meteorological 

 data. \ corresponding pressure oscillation is clearly shown. 

 It is followed after nine days by a variation, of the same 

 kind, of the temperature and the daily range of pressure. 



There seems to be a close connection between the above- 

 mentioned pressure oscillations and the influence of the 

 sun's prominences on the earth. Both coincide in relation 

 to time. ■ C. Braak. 



Observatory, Batavia, January 11. 



Electrons and Atomic Weights. 



LoTi«R Meyer suggested that the slight divergences 

 between the theoretical and actual atomic weights in the 

 periodic system might be due to the imprisonment of a 

 quantity of the a;ther within matter ; Lord Kelvin ascribed 

 to the tether a weight of one-thousand billionth of a gram 

 per cubic meter. Meyer's suggestion is hardly acceptable. 



In the light of present-day theories of the perpetual dis- 

 integration of matter, it seems more likely that the atomic 

 weights vary through loss of electrons ; when the loss has 

 reached a certain critical point a re-adjustment may take 

 place, resulting in transmutation to a lower element. 



If there be any truth in this theory, it may be supposed 

 that the atomic weights of the elements may vary in 

 different worlds of space, the more or less uniform weights 

 found on the earth being due to the fact that the period 

 of formation was identical in all cases. In this connec- 

 tion, it would be interesting to determine the atomic 

 weights of the elements in meteorites, an investigation 

 which I am unable to undertake at this time, but which 

 I hereby suggest. Alfred Sang. 



q6 Boulevard dc \'crsailles, St. Cloud, S. et O., 

 January 12. 



