March 1, 1901.1 



KNOWLEDGE. 



01 



1S36, 1S47, 1S58. The ceutval Hue in 1921 crosses the 

 island of Lewis, aud ouw.irds to the Lofodeu Islands. 



The eclipse of 1925 is tctal near Boston, U.S.A., but 

 docs not appear to touch land on this side of the 

 Atlantic. 



In 1927 we have the next totality in the British 

 I.slcs. The central line runs from near Anglesea to 

 North Yorkshire, where totality will last about 21 

 seconds. Dr. Hind gives it as nine seconds, but this 

 seems to me to be clearly too small. The track then goes 

 right up the backbone of Norway, and passes out near 

 Yadso, which was occupied as an eclipse station in 1896, 

 and may again be occupied, as the duration of totality 

 there is considerably longer than in southern Norway ; 

 our weather experiences m 1896, however, were not very 

 encouraging. 



The eclipse of 1936 may be well observed from Con- 

 stantinople; that of 1915 from Nonvay, Sweden or 

 Finland. 



In 1954 we may have another totality in the British 

 Isles, as Dr. Hind considered that the northernmost of 

 the Shetlands would lie within the shadovi^ track ; in 

 any case it will be observable in southern Norway.* 



The remaining eclipses do not need much comment. 

 The eclipse of 1961 may be well observed in Italy and 

 Turkey. In 1966 there will be a very brief totality in 

 Greece. It is not shown on the map. In 1999 occurs 

 the third totality of the centiu'y in the British Isles, and 

 the most favourable of all. 



The north limit of totality passes approximately 

 through Tintagel Head, Exmouth and Weymouth ; the 

 central line from St. Ives (Cornwall) to Prawle Point. 

 Hence the shadow will cover nearly the whole of Corn- 

 wall and the southern portion of Devon. The duration 

 of totality will be just two minutes. This eclipse will 

 also be visible in North France (the central line runs 

 from St. Valery to Laon), Germany, Austria and 

 Turkey. 



Many interesting illustrations of the Saros cycle or 

 period of 18 years 11 days, after which eclipses recui', 

 may be derived from oiu" eclipse map. This cycle was 

 explained in an ai'ticle by Mr. Walter Maunder in 

 Knowledge for 1893, Vol. XVI., p. 181. Each recurring 

 eclipse moves westward about 120° of longitude, so that 

 after 54 years we get an eclipse in about the same longi- 

 tude, but north or south of the first according as the 

 Moon is at a Descending or Ascending Node. 



The eclipses of 1882 (Egj'pt), 1900 (Spain), 1918 

 (United States), 1936 (Greece), 1954 (Shetlands), are an 

 example of an eclipse moving slowly northward. 



Again, the eclipse of 1860 was visible in Spain; after 

 three more Saroses it crosses Norway in 1914, while 

 three more bring it to north-east Russia in 1968; after 

 one more return in 1986 west of Iceland this eclipse 

 ceases to be total.- 



As an example of a southward moving eclipse, that 

 of 1851 was total in South, Norway ; after three Saroses 

 it crosses Spain in 1905, while in 1959 it will cross the 

 Sahara. 



Again, the eclipse of 1927 returns in 1945, and after 

 three more Saroses it givss us that of 1999. 



While conveying a general idea of the locality where 

 an eclipse will recur, the Saros is not exact enough to 

 predict the track of a futui-e eclipse with great precision. 

 Major-General Strahan, in his report on the Indian 



eclipse of 1898, attempts to predict the eclipses of the 

 next century by the Saros cycle alone, but many of his 

 tracks {e.g., 1961) are considerably in error. 



It would be useless to completely classify the non- 

 European totalities for 100 years, but the more impor- 

 tant eclipses of the next 30 years, some of which have 

 been already mentioned, arc given in the following 

 list : — 



1901. Mauritius, Sumatra, New Guinea (long 

 totality). 



1904. Pacific Ocean (long totality). 



1905. Labrador, Spain, Tunis, Egypt. 



1907. Caspian, Turkestan, Mongolia. 



1908. Pacific. 



1911. Pacific. 



1912. Soutli Aniei-ica (Bogota to Kio Janeiro).* 



1918. North America (Vancouver to Florida). 



1919. Brazil, Gold Coa.st, Lake Tanganika (long 

 totality). 



1922. Australia (long totality). 



1923. North America (San Francisco to Jamaica). 



1925. North America (Lake Superior to Boston). 



1926. Victoria Nyanza, Amirantc Islands, Sumatra. 

 1929. Sumatra, Malay Peninsula, Philippines. 



The next thirty yeai-s afford on the whole a more 

 favourable series of eclipses than the last thirty years, 

 and as many of them arc in reasonably accessible locali- 

 ties, it is to be hoped that they will be extensively 

 observed, and unravel many problems of solar physics 

 that are still veiled in mystery. 



* I -,1111 inliinijiii 'rp\ Eev. S. 3. Juiiu-.m luat iJl-. Iliii.l ■.iiij?.-..i.i.;liily 



foncludctl that the shadow would not touch the Shetlands, but would 

 lie fiiHl.^"- \..rt1i. 



SUNRISE ON THEvSEA OF PLENTY. 



By E. Walter Maunder, f.r.a.s. 

 The region of the Moon shown in the present plate, 

 which is taken from the fourth number of the magnifi- 

 cent Atlas Photographique de la Luue, published by 

 the Paris Observatory, exhibits under the double 

 influence of strong foreshortening and sunrise illumina- 

 tion, a striking amount of relief; and the nearness of 

 the terminator to the limb accentuates a feature, 

 common indeed to the whole lunar surface, but here 

 seen most strikingly, namely the tendency for the chief 

 formations to dispose themselves along meridians. Thus 

 we find the narrow strait of light, lying between the 

 terminator and the limb, is almost exactly bisected by a 

 succession of great wailed plains, — Petavius, Vendelinus 

 and Langreuus, — all on the 60th meridian. This meridian 

 in its sweep northward, traces out the western boundary 

 of the Mare Fecuuditatis, cuts through the ring-plains, 

 Webb and Apollonius, and bisects the Mare Crisium. 

 In this latter Mare, we see the influence of these 

 meridianal forces marked in still greater detail, since 

 it is crossed by a number of ridges, — one of which is 

 especially distinct on the photograph, — which streak it 

 from north to south; a clear indication that llic tidal 

 attraction of the cartii when the moon's crust was still in 

 a plastic condition, was a chief agent in moulding the 

 surface into its present shape. 



The Sea of Conflicts, from the beauty of its surround- 

 ings and the strong relief into which it is thrown whilst 

 the crescent moon is still young, has always been a 

 favourite object of study. Its size too is convenient ; it 

 can be examined as a whole, and the definiteuess of its 

 outline makes it the more suitable for such study. In 

 length, east and west, it reac hes to 355 miles, whilst its 



* There will thus in 1012 be two total eclipses six months ajMrt, 

 which is a very i-Jiv occurrence. 



