460 



NATURE 



[Oct. 8, 1874 



in use on the clock of the apparatus of Cape Bon, Tunis, an 

 apparatus exactly sinular to that now standing in the Inter- 

 national Exhibition. It consists of a shaft making 170 revolu- 

 tions per minute, to which the balls a a are hung, and on which 

 the disc /' /' can slide, guided by a feather key. When the clock 

 is below speed the disc rests upon a collar fixed on the shaft, the 

 pull exerted by the balls through the links dd being insufficient 

 to raise it; but as soon as the proper speed is attained, the disc 

 rises and comes in contact with the screws c ,\ which are tipped 

 with leather and fixed to the frame of the clock. Spaces are cut 

 out of the disc to admit the balls, avoiding unnecessary height. 

 The screw / serves as a brake to stop the clock at pleasure. I 



calculate that work to the extent of five foot-pounds per minute 

 must be done on the governor to accelerate the clock one second 

 per hour. This form possesses two advantages over that in 

 which the rubbers are carried by the balls— I. It checks any 

 acceleration of the clock more powerfully : 2. It is easier to 

 adjust. In the older form it is necessary to ascertain by careful 

 experiment that eacit ball shall bring its rubber into contact 

 exactly when the speed is correct, whereas in this it is immaterial 

 that tlie arms of the balls should be exactly equal ; it is only 

 needful that they should together raise the disc to contact when 

 the speed is right. J. HoPKlNSON 



Glass Works, near Birmingham, Sept. i 



Rainbows 



As a pendant to my note inserted in Nature, vol. x. p. 437, 

 I may mention that an exceedingly fine lunar rainbow was 

 observed here at S.40 r. M. on September 29. 



Though the moon was near the last quirter, the bow was 

 bright enough to appear reddish on one side and greenish on the 

 other. It is the only one, of some five or six lunar rainbows I 

 have seen, which appeared to show any trace of differences of 

 colour. 



I may also mention that about the end of August I saw, two 

 hours after sunrise, a dazzlingly bright and gorgeously coloured 

 parhelion in a small ice-cloud to the right of the sun, the rest of 

 the sky being almost perfectly clear. There had been a sudden 

 and considerable fall of temperature during the previous night. 



St. Andrew's, Oct. 2 P. G. Tait 



In Nature, vol. x. p. 438, Mr. Schuster complains that in 

 text-books no mention is made of supernumerary rainbows, and 

 that the theory of them is to be sought in original memoirs, not 

 generally accessible. Allow me to mention that in Sir John 

 Herschel's Meteorology (a little work published by Black, price 

 three and sixpence, and originally an article in the Encycl. 

 Britann.), a complete explanation of the rainbow, and of the 

 supernumerary bows as well, on the principle of interference, is to 

 be found. F.M.S. 



U.S. Weather Maps 

 In Prof. Loomis's "Results of an Examination of the 

 U.S. Weather Maps for 1S72 and 1873" (published in the 

 Aviericari Juuntal of Science and Arts, and recently noticed 

 in Nature, I am stnick not only by the general agreement 

 but by the almost verbal coincidence of one or two of his 

 " Results " with some of the rules laid down in my work on the 



"Laws of the Winds 'prevailing in Western Europe," which was 

 published in the beginning of 1S72. 



In " Laws of the Winds," Part I. p. 56 and following, I have 

 shown that "we are unable to account for the eastward progress 

 of depressions by attributing it to prevailing westerly upper- 

 currents," but that " each system of depression appears to travel 

 eastward with a kind of self-developed motion," anl that the 

 precipitation on the east side of the centre " is the principal agent 

 in producing the change of geographical position." Prof. Loomis 

 writes: "The progress of a storm eastward is not wholly due 

 to a drifting Jresulting from the influence of an upper-current 

 from the west, but the storm works its way eastward in conse- 

 quence of the greater' precipitation on the eastern side of the 

 storm. " 



Prof. Loomis also appears to attribute the formation of some 

 depressions, primarily developed in the United States, to the 

 collision of moist air from the Pacific with the mountains in the 

 norlh and west, in the same way as I have attributed the primary 

 formation of some of our depressions to the collision of the 

 vapour-laden atmosphere from the Atlantic with the high-lands 

 in the west and north of the British Isles. 



I am glad to observe that Prof. Loomis is no advocate of the 

 "circular theory " of storms as still held by some meteorologists. 

 He intimates the mean inclination of the wind towards the lower 

 isobars as "more than 45°" in the United States. In the 

 Journal of the Scottish Meteorological Society, No. xxxix. I 

 have shown that at stations in the British Isles tlie mean inclina- 

 tion is 21°, but that it appears to be considerably higher in con- 

 tinental Europe. 



In the work previously alluded to I have shown that depres- 

 sions appear to travel most to the south when the atmosphere is 

 warmer in the west than in the east, and most to the north under 

 contrary circumstances, but that this influence is interfered with 

 by another, viz., the tendency of depressions to travel so as to 

 have the highest general pressures on their right. A less limited 

 acquaintance even than I can claim with the U S. Weather Maps 

 would go far to show which of these two influences is the pre- 

 dominant, the general atmospheric conditions of the United 

 States presenting a better field for their investigation than is to 

 be obtained in Europe. Prof. Loomis finds that in North 

 America storms tend most to the south in July and to the north 

 in October. It would be interesting to mquire whether this 

 observation holds good of depressions on tse Pacific coast, as 

 well as near the Atlantic. But a two years' average is insufficient 

 to settle such questions. 



On the whole it is satisfactory, to find that some important 

 results obtained from a study of European weather-charts are 

 found, on good authority, to be in accordance with those derived 

 from the U.S. maps. At the same time some ot the theoretical 

 remarks made by Pruf. Loomis will not, I think, be generally 

 endorsed by meteorologists. The statement that "it needs no 

 argument to prove that when the winrl is flowing from all quarters 

 in\^'ards upon a central area, there is a rapid accumulation of air, 

 which can only escape by an upward motion," is incorrect ; the 

 depression of the barometer in the centre showing that there is no 

 accumulation, but a rarefaction, produced in part, as Prof. Loomis 

 has himself previously shown, by precipitation, and wliich is 

 itself the cause of tlie influx. 



Under the present conditions of anemometry all endeavours to 

 calculate the upward movement in a storm from anemometrical 

 data should also be accepted with much reserve. .Still more 

 hazardous (considering the inclination of depression-axes and the 

 frequent dilTerence ol direction between currents at small and 

 those at great elevations) is the attempt, in such an inquiry, to 

 correct the observed velocities at sea-level by those on the summit 

 of Mount Washington. With a depression in Eastern Canada a 

 west wind not uncommonly blows on Mount Washington while 

 more southerly airs are felt at the three nearest stations. If in 

 such a case we calculate the amount of influx towards the depres- 

 sion-centre simply from the ratio between the velocity at sea-level 

 and that on Mount Washington, it is obvious that the result will 

 be the reverse of accurate. 



Aug. 2j W. Cle.ment Lev 



Aurora 



On Sept. II I was at Kyle Akin (Skye). The day h.ad been 

 wet and stormy, but towards evening tlie wind fell and the sky 

 became clear. About 10 r. M. my attention was drawn to a 

 beautiful auroral display. No crimson or rose tint was to be 

 seen, but a long low-lying arc of the purest white light wa 



