&o8 



NA TURE 



[April 30, 1908 



application of Kirchhoff's law sometimes leads students to 

 imagine that the two quantities are identical, but Kirch- 

 hod's law applies only to purely thermal radiation (c/. 

 Wood's " Physical Optics," chapter xix.)- A gas, such 

 as hydrogen, in its non-luminous condition does not absorb 

 selectively the light emitted by luminous hydrogen. I do 

 not think that any absorption bands in non-luminous 

 hvdrogen have been detected ; they are probably far in the 

 ultra-violet, and there is no reason for supposing that 

 their wave-lengths will be connected by any formula similar 

 to that of Balmer. 



The reason for the difference in the frequencies of the 

 absorption bands and the lines in the emission spectrum is 

 sufficiently obvious. The emission of light by a gas is 

 doubtless connected with the ionisation of its atoms. 

 But, when an atom is ionised, the electrons in or surround- 

 ing that atom are subject to forces entirely different from 

 those which act upon them when the atom is ionised ; 

 there must be a corresponding difference in the periods of 

 free vibration. The absorption bands probably represent 

 the vibrations of the electrons in the neutral atom, the 

 emission spectrum those of the electrons in or around the 

 ionised atom. 



Nor is there any reason why the refractive index of a 

 hnninoiis gas should differ greatly from that of a non- 

 luminous gas, except in the immediate neighbourhood of 

 a line in the emission spectrum. (It is relevant to note 

 that Drude's formula cannot be applied to such regions.) 

 Only a very small proportion of the total number of atoms 

 present is ionised even under the most favourable experi- 

 mental conditions ; most of the atoms are not ionised, and 

 affect the light in the same way as those of a non-luminous 

 gas. Of course, if the refractive index due to the luminous 

 atoms were really infinite, the refractive index of the 

 luminous gas would be infinite, however small the pro- 

 portion of luminous atoms. But it is impossible that it 

 should be infinite ; if Balmer's formula were accurately 

 true for all values of »«, there would be an infinite number 

 of lines in the emission spectrum, implying an infinite 

 number of degrees of freedom in the vibrating system. 

 .According to the modern view, which seems to be accepted 

 by Prof. Schott, this system is composed of discrete charged 

 particles possessing a finite mass ; the number of such 

 particles must be finite, and they can only have a finite 

 number of degrees of freedom. Experiment can never 

 demand an infinite number of lines, for, if m is very great, 

 the lines are so close as to be beyond the range of resolu- 

 tion. Norman R. Campbell. 



Trinity College, Cambridge. 



The Oligcchsetous Fauna of Lake Birket el Qurun 

 and Lak3 Nyassa. 



In N.ATurjE of August i, 1907 (vol. Ixxvi., p. 316), Messrs. 

 Cunnington and C. L. Boulenger wrote a preliminary 

 account of the fauna of Lake Birket el Qurun. I am 

 indebted to these gentlemen for the opportunity of supple- 

 menting their account by a note upon the Oligocha?ta of 

 that lake. They were so good as to send to me two tubes 

 with a large number of specimens of a small oligocha?tous 

 worm collected in the lake. These specimens were found 

 to belong, without exception, to the species Paranais 

 littoralis. The occurrence of this Naid in northern Africa 

 is a new fact in its distribution. It has hitherto been met 

 with in many parts of Europe, both in fresh water and in 

 brackish, even salt, water. .As to its marine habitat, it 

 has been collected on the shores of Denmark and near 

 Odessa. 



It is clear from the fact that this was the only aquatic 

 Oligochaete met with by Messrs. Cunnington and 

 Boulenger that it must at least be a prevalent form in the 

 lake. 



Mr. Cunnington has also kindly placed in my hands 

 some examples of aquatic Oligochaeta from Lake Nyassa-. 

 These belong to three species, and the contrast with the 

 oligochsetous fauna of the North African lake is very 

 marked. The genera represented in Nyassa are Dero, 

 Nais, .-ukI Pristina. Unfortunately, none of the examples 



NO. 2009, VOL. 77] 



submitted to me are sexually mature. The Pristina 1 

 identify with Pristina longiseta, a widely spread form. 

 The genera Nais and Dero are also found in many parts 

 of the world, and as all three genera have already been 

 recorded from tropical East Africa (Michaelsen in Zeitschr. 

 f. wiss. Zool., Bd. Ixxxii., 1905, p. 289), there is no cause 

 for surprise at their occurrence in Nyassa. Still, the fact 

 seemed to me to be w'orth putting on record. 



Frank E. Beddard. 

 Zoological Society's Gardens. 



THE FORTHCOMING DUBLIN MEETING 

 OF THE BRITISH ASSOCIATION. 



THE British Association will hold its fourth meet- 

 ing in Dublin on^ September 2-8 of this year. 

 The three previous meetings took place in 1835, in 

 1857, and in 187S. The 1878 meeting drew an attend- 

 ance of 2578, which was well above the average, and 

 it was marked by the presence of an unusually large 

 number of inen distinguished in scientific or other 

 work at the time and since. In his presidential 

 address before the Anthropology Section, Prof. 

 Huxley spoke prophetically about those " who may 

 be here thirty years hence — I certainly shall not be," 

 little realising how both his prophecies were destined 

 to come true. 



Appropriately enough (though, I believe, not con- 

 sciously in connection with Huxley's forecast), the 

 invitation for 1908 originated with Prof. W. H. 

 Thompson, of the physiological laboratory of Dublin 

 University. It was originally intended for 1907, but 

 1908 was found to be more suitable, and the invita- 

 tion was formally accepted at York in 1906, where 

 a deputation attended consisting of the Provost of 

 Trinity College, Prof. Thompson, Dr. Tarleton, Mon- 

 signor Molloy, Rev. Dr. Delaney, and the Lord Mayor 

 of Dublin. The invitation was formally renewed at 

 Leicester by Prof. Thompson, Rev. Dr. Delaney, 

 and Sir Howard Grubb. At a first meeting at 

 the Dublin Mansion House, general and executive 

 committees were appointed, and these have been at 

 worlv ever since, with the result that the arrange- 

 ments are in an advanced state of preparation. Sub- 

 committees were appointed to deal with finance, 

 entertainments, hospitality, and the handbook re- 

 spectivelv. The four local secretaries are Dr. Joseph 

 McGrath, secretary of the Royal LIniversitv, Prof. 

 W. E. Thrift, Prof. W. H. Tho'mpson, and Mr. John 

 Mulligan, of the Hibernian Bank. About 3000/. has 

 already been subscribed towards the expenses of the 

 meeting. The reduced fare tickets will be available 

 for a month, so as to include the excursions subse- 

 quent to the official meeting. Day excursions will 

 be arranged during the week of the meeting to the 

 Devil's Glen and Glendalough in county Wicklow, 

 to Powerscourt Waterfall and the Dargle near Bray, 

 to the Boyne Valley, and to the Shannon Lakes. 



The presidential address will be delivered by Mr. 

 Francis Darwin, F.R..S., on the evening of Wednes- 

 day, September 2, in the large hall of the Royal 

 University. Here also w-ill be delivered two of the 

 evening discourses, one on " Halley's Comet," by 

 Prof. H. H. Turner, F.R.S. (Friday, September 4), 

 the other on " The Lessons of the Colorado Cation," 

 bv Prof. W. M. Davis, of Harvard LIniversity 

 (Monday, September 7). The third evening discourse 

 will be delivered to operatives on Saturday, 

 September 5. To this discourse ordinary members 

 will not be admitted. The lecture hall has not yet 

 been decided upon. 



The serious work of the sections will, for the most 

 part, be over each day by two o'clock, leaving the 



