April i, 1880] 



NATURE 



5*3 



distance, the resulting positions and distances being all 

 that is wanted. The volume is, however, a valuable 

 addition to double star measurements. 



The Ophiurida and Astrophytida of the "Challenger" 

 Expedition. By Theodore Lyman. Part 2, Bull. Mus. 

 Comp. Zool. Vol. vi., No. 2. (Cambridge, Mass.) 



This is the second part of the preliminary description of 

 the Ophiurida; and Astrophytidae dredged by the Chal- 

 lenger. Prof. Lyman issued the first part of the Prodromus 

 some time ago. The Prodromus is of course merely an 

 abridgment. Prof. Lyman's full account of the Ophiurida: 

 will appear in the large work on the Challenger Expe- 

 dition. To the present part is added an index of species 

 contained in the two parts, together with all others 

 described elsewhere by Prof. Lyman. The whole forms 

 a list of the greater portion of deep-sea Ophiurans and 

 Astrophytons known. The list comprises fifty-three 

 genera and about two hundred and twenty-three species. 

 In the present part two new genera and sixty-three new 

 species are described. Prof. Lyman considers that the 

 Ophiuran which was recently described by Prof. Martin 

 Duncan under the name of Ophiolepis mirabilis (Linn. 

 Soc. Journ. Zool, xiv. 460, 479), is a true Cphiopholis, 

 lacking none of its characters, and standing quite near 

 the typical O. aeuleata. Priority is given in all cases by 

 Prof. Lyman to specimens dredged by the Challenger 

 over those obtained by the later series of dredgings 

 carried out by the United States Government under Mr. 

 Alexander Agassiz. A similar priority has been generously 

 given by Mr. Agassiz to the Challenger Echinoidea, and 

 Count de Pourtales has shown similar consideration in 

 the matter of the corals. Owing to the delay in the pub- 

 lication of the Challenger results, the American naturalists 

 could easily have secured priority for their collections, had 

 they thought fit to do so. They have in their hands 

 almost all the forms of any importance which the Chal- 

 lenger obtained, for by their continued operations they 

 have dredged them nearly all on the United States coast 

 and around the West Indies. The thanks of English 

 naturalists is certainly due to the American zoologists for 

 their courtesy in this matter. 



LETTERS TO THE EDITOR 

 [The Editor does not hold himself responsible for opinions expressed 

 by his corresrondtnts. Neither can he undertake to 1 cturn. or 

 to correspond with /he writers of, refected manuscripts. 'No 

 notice is taken of anonymous communications. 

 [The Editor urgently requests correspondents to hep their letters as 

 short as possible. The pressure on his space is so peat that it 

 is impossible otherwise to ensure the appearance even of com- 

 munications containing interesting and novel facts.] 



The Density of Chlorine 



Tin: article on the density of chlorine, bromine, and iodine at 

 high temperatures which appeared in Nature, vol. xxi. p. 461, 

 ore your readers in the clearest manner the present 

 condition ol this important question. The conclusion hinted at 

 in the closing sentences of the article, viz. that the-e gases are 

 under certain circumstances decomposed, is however scarcely 

 warranted. Dr. Armstrong thinks that these substances may he 

 more liable to decomposition when in a nascent state. It is 

 generally supposed that in this condition the atoms of a su'bs'an :e 

 are separate, having as yet had no opportunity of selecting a 

 mate for their further career ; if therefore we could observe the 

 density of a gas in the na-cent state, we should find that it was 

 only half the theoretical density. In the case of chlorine evolved 

 1111111 chloride at a high temperature we may readily 

 imagine the emerging atom, set in rapid movement by the great 

 heat, to be unable at any time to join with another to form a 

 molecule ; we should thus have the na-cent state maintained, if 

 I may be allowed the expression, as long as the temperature was 

 high enough. It is further possible that there may be a wide 

 interval between the temperature at which chlorine gas is 

 molecular and that at which it is entirely atomic, and that in this 

 interval a certain proportion of the gas varying with the tempera- 



ture is resolved into its atoms, the rest remaining molecular. 

 The gas would then have a density intermediate between the 

 theoretical density 2-45 and its half, 1 -23, a density in fact 

 corresponding with that obtained in Meyer's experiments. 



It may be urged that this attempt at an explanation, necessi- 

 tating as it does a density varying with the temperature, is 

 incompatible with the facts, since Meyer obtained a uniform 

 density of about 16 in all his experiments. It must however be 

 remembered that these observations cannot lay claim to great 

 accuracy, and that the recurrence in several experiments of the 

 same observed density may often be ascribed to chance. 



Of this we hive an excellent example in the experiments 

 recently recorded in the Proceedings of the Royal Society by 

 Prof. Dewar and Mr. Scott. The densities required were those 

 of the vapours of potassium and sodium. In a first series ot 

 experiments which were made in an iron vessel the mean density 

 of potassium vapour (referred to hydrogen) was found to be 

 40'8, that of sodium vapour 25-33, whence it was naturally 

 inferred that those vapours were normal in character. In a 

 second series of experiments, in which a platinum vessel was 

 used, the densities 21 and 13 were found for potassium and 

 sodium vapours respectively ; from this it was with equal rea-on 

 inferred that these metallic vapours were atomic, and resembled 

 that of mercury. Unless platinum has a special dissociating 

 effect on the molecules it must be admitted that in the one series 

 or the other (since they were both made at similar temperatures) 

 the concordance of the results was due to chance. 



That the density of chlorine is really subject to gradual varia- 

 tion as the temperature increases, is rendered very probable by 

 the results obtained by Meyer with iodine ; the table of these 

 results given in the article referred to (Nature, vol. xxi. 

 p. 461) shows clearly that the density of iodine decreases gradu- 

 ally, and there would seem to be no reason whatever for the 

 assumption that it is complete at about 1,500" C. 



I fear that I trespass much on your space in thus trying to 

 point out that the otherwise inexplicable density 1 '6 most probably 

 represents only a stage on the road to the complete dissocia- 

 tion of the molecules, a stage more readily reached by a nascent 

 gas than by one in which the molecules have to be dissociated ; 

 the importance of the subject must, however, be my excuse. 



Clifton College, Bristol, March 21 Frel>. V. Brown 



The Annual Variation of the Barometer in India 

 It has been pointed out by Mr. Archibald, in Nature (vol. xx. 

 p. 54), that the late Mr. J. A. Broun, F.R.S., was probably 

 mistaken in supposing (see vol. xix. p. 6) that there is no direct 

 causal connection between the annual variations of temperature 

 and atmospheric pres ure in India. Mr. Broun appears to have 

 adopted this opinion because, at all places in India where the 

 annual oscillations of temperature and pressure are considerable, 

 their turning points are not the same. The highest pressure 

 usually occurs about the middle of December, a'id the lowest at 

 the end of June, while the lowest temperature is reached during 

 the first ten days of January, and the highest in the latter half 

 of May. 



Having been employed a short time ago in calculating the 

 Constants of Be sel's formula; for the annual variations of temper - 

 ature and pre sure at Allahabad, I noticed that the first term of 

 the pres-ure formula, which includes nine-tenths ot the total 

 variati n, reaches its maximum almost exactly at the time of 

 lowest annual temperature. The value of this term at the 

 middle of January is -271'' sin IOI° 32', and its maximum 

 theref >re falb about II J days before the middle of January, that 

 Bon January 4th or 51I1. The same tenn of Bessel's formula 

 for Benares is represented by -279" .sin 102' 34', and for Roorkee 

 by '258' sin 103° 12'. The maximum pressure at these two stations 

 therefore falls about the 3rd of January, if we take the oscillation 

 of annual period ah me. The first periodic term of the formula 

 for the annual variation of pressure at Bombay is given I y Mr. 

 C. Chamhers(" Meteorol igy of the Bombay Presidency," p. 16) 

 as '1405' sin 87° 2", the angle being counted from the 3rd of 

 January at the rate of 30° for a month. This throws the maxi- 

 mum forward to the 5'h January. 



The pressure oscillation of full annual period may be sup- 

 posed to represent the most important part of the effect of the 

 annual variation of temperature, freed from all minor inequali- 

 ties due to changes of wind and other causes. The close 

 coincidence of the time at which this pressure o dilation attains 

 its maximum with the time of the temperature minimum at the 



