Nov. 5, 1874] 



NATURE 



15 



Islands, north of the Sandwich Islands, through the 

 southern point of California, through Mexico and Cuba, 

 and just north of Madeira. In fact, through its whole course 

 it lies just south of the isoclinic line of 60" inclination and 

 between this and that of 50° ; a fact forcibly illustrating 

 Prof. Fritz's remark that the isochasmic curves lie nearly 

 parallel to those of equal magnetic inclination. For this 

 curve we have for the value of M in ]\Iadeira, Cadiz, 

 Naples, Smyrna, Teneriffe, and Cuba o'l, for the Azores 

 ©•IS, for Barnaul 07, and Nertschinsk o'6. 



It is well known that both in ancient and modern times 

 polar lights have been seen occasionally south of this line, 

 as for instance in 'the year 502 at Edessa, in 1097,1098, 

 and 1 1 17 in Syria, in 1621 at Aleppo, and in 1S72 over 

 most of North Africa and India. 



North of this line their frequency rapidly increases, 

 and we ha\e M = i beginning at Bordeaux, through 

 Switzerland and north of Cracow, south of Moscow and 

 Tobolsk, and north of Lake Baikal, through Udsk and the 

 southern point of Kamtschatka, through northern Cali- 

 fornia and the north of Florida. For the values of M 

 for this zone we have for Perpignan, Marseilles, Bordeaux, 

 La Rochelle, and Mviers, a mean of ri, for Moscow i, for 

 Tobolsk o'9, Barnaul 07, and Sacramento o"S. Singularly 

 enough, probably from climatic or other local causes, the 

 value of M for New Orleans is only o'i4. 



The zone for IM = 30 passes through the north coast of 

 Ireland, through Scotland near Edinburgh, through the 

 White Sea and the Gulf of Obi, where it attains a lati- 

 tude of 70 ', and then tends a little southward through 

 Werchni, Kolymsk, and the Bay of Anadyr, near Sitcha, 

 Cumberland House, Quebec, and the north coast of Nova 

 Scotia, to the north coast of Ireland. 



North of this the frequency of aurora rapidly increases. 

 The zone of M = 100 passes through the Hebrides, 

 Shetland, near Drontheim and Wardon, through Nova 

 Zembla, across Behring's Straits, just south of the Arctic 

 Circle, south of Lake Athabasca, through Hudson's Bay, 

 and just north of Newfoundland. 



Only a little further north we reach a zone of maximum 

 frequency, beyond which the intensity of auroral display 

 again declines, contrary to the old idea that its intensity 

 increased up to the poles. This zone passes just north of 

 Faroe and of the North Cape, through the northern part of 

 Spitzbergen, and just north of the Siberian coast, near 

 Point Barrow, Great Bear Lake, and Nain on the coast of 

 Labrador. Iceland, Spitzbergen, and Greenland lie con- 

 siderably to the north of this zone, and aurora; are not 

 there so frequent, nor especially so brilliant as at Faroe, 

 the north coast of Norway, and Labrador. Of this Prof. 

 Fritz adduces much evidence, and in addition draws 

 attention to the important fact, that while south of this 

 zone of maximum frequency the arches are generally 

 north of the observer, from the north of it they appear to 

 the south, and upon it, indifferently, north, south, or 

 overhead. 



It will be noticed that the system of curves tends strongly 

 southward in North America, while in the Atlantic and 

 Pacific Oceans the curves pass rapidly northward and 

 reach their highest latitudes in Central Asia. This is 

 borne out by the fact that the great aurora; of Aug. 28 

 and Sept. i, 1S59, were not noted in the meteorological 

 registers either of Nertschinsk, Barnaul, or Jekaterinburg, 

 nor were they seen at Tigris in Yozgat (39° N.), Mosul 

 (36° N.), or Kharput (33'' N.) ; whilst in the Atlantic Ocean 

 they were visible at least to 12^ N., in Africa to St. George 

 del Mina (28^ N.), and in America iluring the maximum 

 they were frequently observed in the Antilles (20'-' N.) 



The geographical extent of great displays of polar 

 lights is very significant. That of Sept. i, 1S59, was 

 visible in the Sandwich Islands (20" N.), Sacramento 

 (20"' N.), San Salvador (13' N.), in the whole Atlantic 

 Ocean to 12^-' N., in Western Africa to 14" N., and in the 

 whole of Europe. At the same time the southern lights 



were seen in Australia, South America to 33" S., and in 

 the Indian Ocean to 39" S. 



For the southern hemisphere there are as yet too few 

 observations to calculate the distribution as has been 

 done for the north. For Hobarton (43° S.) M = 6, and 

 for Melbourne 15. In low latitudes they have been seen 

 at Cusco (12° S.) in 1744, at Rio Janeiro (23° S.), 1783, 

 at Bloemfontein (29° S.), and Vaal-Fluss (28' S.); in Africa 

 and at Reunion and Mauritius in 1870 and 1S72. 



Dr. Fritz remarks that his zone of greatest frequency 

 nearly coincides with that given by Munckc (in " Gehler's 

 Worterbuch), and that the whole curve-system has great 

 similarity to the zone-system of Loomis in Sillinuni's 

 JoHiiial, vol. XXX. The curves cut the magnetic meridians 

 in most places at right angles, and are very similar to the 

 isoclinic curves constructed by Hansteen in 17S0, while 

 they noticeably deviate in places from those of Sabine of 

 I S40, and approximate, at least in the best determined 

 portions in East America, the Atlantic Ocean, and Europe, 

 with the isobaric curves of Schouw. It may here be 

 remarked that the curves of increasing frequency in the 

 Atlantic Ocean tend towards the point of lowest baro- 

 metric presure. 



It is also noticeable that throughout the greater part of 

 the northern hemisphere the curves tend to follow the 

 form of the continents, and the limits of perpetual ice 

 which depend upon it ; and Prof. Fritz points out that in 

 mean latitudes the magnetic meridians and the direction 

 of visibility of the aurora are coincident, and are mostly 

 (viz., from the Atlantic Ocean to the Asiatic Icy Sea) 

 normal to the limit of ice. The greatest deviations from 

 this rule exist in places where the ice-limit is most irre- 

 gular, as, for instance, in Hudson's Bay and the Gulf of 

 Labrador. It may here be noted that at Fort Franklin, 

 Fort Nermann, and Wardoehus the northern lights begin 

 in spring to be seen most frequently in the south at the 

 same time as the ice-limit deviates furthest in the 

 same direction. At Bossekop, according to the report 

 of the Scientific Commission, the northern appearances 

 are to the southern ones as 3'6 to i during the four last 

 months of the year, but only as 2 to i in spring. 

 Wrangel, from his observations on the coast of the 

 Arctic Ocean, concludes that the freezing of the sea is 

 favourable to aurora ; but remarks that in the east of Asia 

 the appearance is more frequent as the coast is ap- 

 proached, and is most so during the increasing cold of 

 November, while it becomes rarer in January, when the 

 coast ice extends further to the northward. M'Clintock 

 notices that aurora was most frequently visible when 

 water was in sight ; and Hayes, that it was more fre- 

 quently seen in the direction of some piece of open water 

 than of the magnetic north. These observations would 

 rather support a belief common in Scotland that the fre- 

 quency of the aurora varies with increase and decrease of 

 the Greenland ice, and render it probable, at least, that 

 ice-formation is one of the most prominent local inllu- 

 ences by which auroral distribution is affected. It seems 

 not unlikely that the neighbourhood of the Alps may 

 influence the frequent displays in North Italy. These 

 and other points, however, require more systematic obser- 

 vation, and it is especially desirable that some notice 

 should be taken of the relative intensity of different dis- 

 plays. H. R. P. 



EDWIN LANKESTER, M.D., E.R.S. 

 T T is with great regret that we have to announce 

 -'■ the death, from diabetes, on Fi'iday last (October 30), 

 at Margate, of Dr. Lankester, the Coroner for Central 

 IMiddlesex. 



Dr. Lankester was born April 23, 1814, at Melton, near 

 Woodbridge, in Suffolk, at which latter town he received 

 his early education and commenced his medical studies. 

 In 1834 he entered University College, London, as a 



