3o8 



NATURE 



[July 30, 1903 



Black, and Brown, from their characteristic aspects ; the 

 first being snow-covered, while the two others displayed 

 the bare basaltic rocks of which they are composed. In 

 September and October minimum temperatures of —51° 

 and —57° F. were experienced during two of the journeys. 



During a visit to Mount Terror, the eastern slopes of 

 which are terribly wind-swept and bare to the summit, Mr. 

 Skelton made a perilous descent to the sea-ice. and was sj 

 fortunate as to discover a breeding-place of the Emperor 

 penguin, obtaining several specimens of the young in down, 

 besides photographs and notes. The attacks of scurvy which 

 occurred about this time, brought about by severe work 

 and exposure, were in reality very slight, and their im- 

 portance has been much exaggerated, all symptoms quickly 

 disappearing when the diet was restricted to seal-meat. 

 Skua gulls, which were also obtained, were found to be 

 excellent eating. 



The southern sledge expedition undertaken by the Com- 

 mander, with Lieut. Shackleton and Dr. Wilson, was 

 carried out entirely on the surface of the great ice-sheet, 

 it being found impossible to reach the land, though it was 

 sufficiently near to allow of observations as regards the 

 bearing and altitude of the different land-masses, as well 

 as sketches and photographs. A remarkable feature seems 

 to be the fiord-like openings by which it is penetrated at 

 various points, though the intervening volcanic masses rise 

 into magnificent ranges of mountains. These openings 

 had the appearance of straits, nothing being seen behind, 

 though the state of the ice-sheet opposite them showed that 

 ice must be pressing out through them. On approaching 

 the land at the furthest south, the ice-sheet was found to be 

 separated from it by an immense chasm, the ice-foot re- 

 sembling that seen elsewhere at the sea-margin of the 

 lands, and forming a complete bar to further progress. 

 The return journey was rendered difficult by the nature of 

 the surface and the prevalent mist. 



The further details supplied of Lieut. Armitage's western 

 expedition show that, after proceeding up one large glacier, 

 lying between precipitous granite mountains, a ridge 

 was crossed by a pass 4000 feet above the sea to a 

 second glacier, which had a general trend from south- 

 west to north-east. Its right-hand branch was ascended 

 to a range of remarkably bare granite mountains, the 

 ice surface being much crevassed. A line of sticks set 

 up during the ascent showed a maximum motion of 3 feet 

 8 inches in twenty-three days. On gaining a height of 

 9000 feet a smooth, open snow-covered plain stretched to 

 the westward, its surface being soft, with successive crusts 

 nine inches or a foot apart. There were no sastrugi. The 

 whole horizon to the west was clear and unbroken, and the 

 plain appeared to have a slight fall in this direction. 

 Running streams, 7 feet wide, with occasional pools some- 

 times a mile in diameter, were seen on the return journey, 

 and Bergschrunds 150 feet deep were found at the base of 

 the mountains. Among the other journeys described, those 

 of Dr. Koettlitz for the investigation of the ice and esker- 

 like lines of debris in the neighbourhood of the Black and 

 Brown Islands are of most interest. 



An interesting complement to the narrative of the British 

 expedition has been supplied by the report of Dr. von 

 Drygalski, leader of the German expedition in the Gauss, 

 which was published as a supplement to the official 

 Keichsanzeiger on July 10. It describes the voyage from 

 Kerguelen -Jid Heard Island to the supposed position of 

 the non-existent Termination Land, the southward advance 

 to a previously unknown land, in the vicinity of which 

 winter quarters were established, the scientific work done 

 at the winter station, and the sledge journeys undertaken 

 during the stay. The Gauss was frozen fast in the ice to 

 the north of this land, the pack there remaining stationary 

 owing to the shallowness of the sea over the " Continental 

 Shelf.'; Only a few miles to the north it appears to be 

 kept constantly in motion by the heavy swell caused bv the 

 westerly storms, which would have seriouslv impeded the 

 scientific work, besides endangering the ship. From the 

 land rose a bare volcanic peak 1200 feet high, which was 

 named the Gaussberg. The Gauss was set free on 

 February 8 by a strong easterly wind, but was caught again 

 temporarily in a somewhat lower latitude, the final start 

 northward being made on April 8, when the lengthening 

 nights were already making navigation difficult. 



THE STRUCTURE OF SPECTRA.^ 



nPHIS paper gives a very lucid account of the structure- 

 -'■ of various types of spectra, special attention being 

 directed to the work on " series " which has been per- 

 formed during recent years. 



„ After referring to the splendid work performed by 

 Angstrom ^nd Rowland in establishing trustworthy tables 

 of standard wave-lengths, the author passes on to the evolu- 

 tion of the definite laws which have been found to govern 

 the distribution of lines in the spectra of many elements, 

 comparing the occurrence of similar definite groupings of 

 lines in the spectrum of a substance to the " harmonics " 

 obtained in acoustics. 



In 1863 Mascart found that certain groups of lines of 

 characteristic aspect were reproduced in different parts of 

 the spectrum of the same metal, e.g. he found that similar 

 triplets to the " b " group of magnesium were reproduced 

 in the ultra-violet region of the spectrum of that metal. 

 To-day it is known that altogether there are fourteen such 

 groups in the magnesium spectrum, one in the infra-red, 

 the " b " group and twelve in the ultra-violet. 



Similarly in the spectrum of sodium there are twelve 

 such " doublets " as that commonly known as Dj and D, 

 in the solar spectrum. If these " triplets " and " doublets " 

 are represented on a scale of wave-lengths, they contract 

 as they approach the ultra-violet, but if they are represented 

 on a scale of frequencies, the groups of the same metal' 

 become identical, and are absolutely superimposable. 

 Similar groups have been found for a large number of 

 metals by Kayser and Runge. 



The alkaline metals, like sodium, give a series of 

 "doublets," as also do copper and silver, whilst the diva- 

 lent metals (Mg, Ca, Sr : Zn, Cd, Hg) give triplets, 

 although some of them, e.g. Hg, are so mixed up with 

 other groups that at first this arrangement is difficult tc 

 recognise. Here then we have a simple law, which should 

 be credited to Rydberg, viz. " In the spectra of a large 

 number of elements there exist groups which are repro- 

 duced several times, the interval which separates the in- 

 dividual lines of each group (when represented on a scale 

 of ' frequencies ') being exactly the same for all the groups." 



For the alkaline metals the length of the interval which, 

 separates the doublets varies as the square of the atomic 

 weights, as is shown in the following table :— 



It was then found that these groups arrange themselves- 

 in regular series capable of mathematical expression, and 

 in 1885 Balmer found that on plotting the thirteen lines- 

 of hydrogen on a curve which had " m " (the number of 

 the line counting in order from " 3 " in the red to " 15 " 

 in the ultra-violet) for its abscissa and N (the frequency) 

 for its ordinate, he obtained a perfectly regular curve which* 

 could be exactly expressed by the formula 

 N = B/4-B/w2, 



where B was a constant. Later, Messrs. Hales and Des- 

 landres discovered sixteen other hydrogen lines in the- 

 spectra of prominences and various stars, and it was found" 

 that these also might be represented by the above formula. 



The spectra of metals also arrange themselves in similar 

 series, although the relations are not so easily seen atj 

 first, because of the overlapping of the other lines. How- 

 ever, the spectrum of potassium may be taken as an ex- 

 ample, and we find that on plotting the lines in a similar 

 manner we obtain three such series, known respectively 

 as " the principal," " the first subordinate," and the 

 " second subordinate " series. In each of these series the 

 brightness of the lines decreases as they approach the more 



1 "La Structure des Spectres," by Prof. Ch. Fabry, Marseilles, in the 

 Kevue G^neraU des Sciences, No. 5, March 15. 



2 The "doublets" for lithium have not been observed, but if the law is- 

 applicable in this case the interval would only be 0016, which is too smali 

 to be observed. 



0. 1 76 1, VOL. 68] 



