386 



NATURE 



\^Feb. 24, 1 88 1 



of the climatic features of British Columbia, may be found in 

 an appendix written by me for the Canadian Pacific Railway 

 Report of 1880, p. 107. 



The mean temperature of Tongass at the southern extremity 

 of Alaska, from two years' observations, is stated as 46°'5. 



Observatioas have been maintained at Sitka with little 

 interruption for a period of forty-five years. The latitude of 

 thii place is 57° 3', or about one degree north of Glasgow. The 

 mean temperatures are as follows : — spring 41' '2, summer 54° 6, 

 autumn 44°'9, winter 32°"5, and for the year 43°"3. 



According to the Pacific Pilot above qujted, that p)rtion of 

 the Kuro-snvo, having a temperature of 55° F. or more, 

 approaches the coast in the vicinity of Vancouver Island. 

 Temperatures not much lower than this however prevail much 

 farther nor 111. The average temperature of the surface of the 

 sea during the summer months in the vicinity of the Queen 

 Charlotte Islands as determined by me in 1S78 ("Report of 

 Progress, Geological Survey of Canada, 1878-79") is 53''8. 

 Obseri-atious by the U.S. Coast Survey in 1867, in the latter 

 part of July and early in August between Victoria and Sitka, 

 gave a mean surface-temperature of 52°'I. 



George M. Dawsom 



Geological Survey of Canad i, February 1 



" Ttie New Cure for Smoke" 



It was not my intention to trouble you further on this subject 

 at present, but as Dr. Siemens has been good enough to notice 

 the result of my trials with the coke-gas grate, and has asked a 

 question with reference to the grate used by me, it is due to that 

 gentleman that I should at once explain that the grite in w'jicU 

 the trial* were male is of nudern coistruction and permaneally 

 fitted with side-cheeks and back of fire-clay lumps, and tliat 

 when in use with thi coke and gas the back was fitted with a 

 copper plate, and in all other respects the grate was arranged in 

 the manner described and illustrated in Nature, vol. xxiii. 

 p. 26. J. A. C. Hav 



On the Space Protected by Lightning-Conductors 

 The very interesting article by Mr. W. H. Preece on the 

 " Space Protected by a Lightning-Con luctor " {Phil. Mag. 5th 

 series, vol. x. p. 427 et scq., December, 18S0) revives thi, im- 

 portant practical question. The old rule, first enunciated by 

 M. Charles, which makes the radius of the protected circular 

 area around the base of the rod equal to twice its vertical height, 

 has never been satisfactorily verified either on theoretical or 

 experiaiental grounds. This rule was adopted in the Report of 

 the Commission of the French Academy of Sciences drav.n up 

 by M. Gay-Lussac in 1S23 {.-iiiii. de Chim. et de Phys. 2nd 

 series, t. 26, p. 25S), and also in two ether reports drawn up by 

 M. Pouillet, one in 1854 (Complcs re>ul:is, t. 39, p. 1 142), and 

 the other in 1S67 (Comptes rutins, t. 64, p. 102). But still mr.re 

 recently the Committee appointed by the Prefet de la Seine to 

 superintend the construction of lightning-c inductors in the City 

 of Paris, in their Report in February, 1876, reduced 'Has radius 

 of the protected area to I "45 times the height of the rod. I am 

 ignorant on what grounds the Commission adopted this pr.cise 

 number. 



In this state of the problem Mr. Preece's paper was both 

 apposite and welcome. The rule which he deduces certainly has 

 the merit of deftnlteness ; but it seems to me that it fails to be 

 practically satisfactory. For it is very evident that his investi- 

 gation is exclusively applicable to " Bltiitt-Conductors" since the 

 " Powir of Points " is entirely left out of consideration. His 

 deductions might apply to the ^/««^conductors which crowned 

 the Royal Palace of George III., but are scarcely applicable to 

 the fainted rods now employed ! His investigation assumes that 

 the distance of the earth-connected objects from the electrifi.-d 

 cloud is the on.'j' clement which determines the direction of the 

 discharge. It seeois to me that the well establi-hed "poner of 

 points " to discharge, or rather to neutralise the electricity of 

 charged conductors, is an essential element in the problem of 

 the protected space. 



It is a well-known fact that when an electrified cloud ap- 

 proaches a fainted lightniug-C'inductur which is in good c 'n- 

 ducting connection with the earth, the sharp point becomes 

 charged by induction with o^ posite electricity of high tension 

 long before the distance between them approximates that re- 

 quired for a disrnftire di charge ; so that electricity of the 

 opposite kind from that of the cloud escapes from the point 

 in the form of a connective discharge or electrical glow, and 



neutralises that of the cloud, and thus silently disarming it, 

 averts the disruptive stroke of lightning. This neutralisation, 

 due to the power of points, constituting the prevenUve action 

 of lightning-conductors, is justly regarded as the most important 

 function of such rods ; although, under certain extraordinary 

 circuoistances, they may be forced to carrj' disruptive discharges. 

 Under any circu instances, however, it is obvious that pointed 

 conductors mast enlarge the proteced area as compared with 

 blunt conductors. 



It is very difficult, if not impossible, to estimate in a precise 

 xsarmerhiw \.\ai power of points would modify and distort the 

 ejuipotenlial surfaces in the intervening electric field. The 

 problem is evidently one of great complexity. The following 

 circumstances must obviously influence, to a greater or less 

 extent, the magnitude and direction of the resultant electromotive 

 force, which determines the path of discharge, comective or dis- 

 ruptive, viz. : (i) Distance of thunder-cloud from the point of 

 the conductor ; {2) variable dielectric properties of the intervening 

 a-j: ; (3) «';t- of the cloud ; (4) the variable Av;w« of its electric 

 charge, especially under the neutralising action of the pointed 

 rod; and (5) the fi'/.'ciVj/ with which the thunder-cloud ap_)roaches 

 the point of the conductor. The last Cinsideratijn is very im- 

 portant, and at the same tine most difficult to formulate ; for 

 the convective neutralisation is a gradual process requiring time. 

 It is evident that a heavily-charged thunder-cloud rapidly driven 

 towards the point of the conductor might give ri^e to a disruptive 

 s oark, while, if slowly approaching the same, it would have been 

 silently neutralised, and the stroke averted. In fict the strength 

 and direction of the resultant force is influenced by so many 

 variable conditions that it would tax the resources of a po.verful 

 calculus to indicate a formula which would satisfy, even approxi- 

 mately, the demands of practice in the ccistruction of lightning- 

 conductors. 



Nevertheless, it is quite certain that Mr. Preece's rule, which 

 makes the radius of the protected circular area equal to the 

 height of the rod for blunt conductors, is perfectly safe for 

 pointed rods ; for there can be no question as to t'ae fact that the 

 " power of points " enlarge: the protected area. 



The late Prof. Henry frequently witne-sed the efficacy of 

 convective discharges from the point of the lightning conductor 

 attached to the high tower of the Smithsonian Institution. 

 During violent thunder-storms at night, at every flash of lightning 

 he observed that "ajet of light, at least five or six feet in length, 

 issued from the point of the rod with a hissing noise." 



It is proper to add that while the circumstances influencing 

 disruptiz'e discbarges of electricity have been experimentally in- 

 vestigated by a number of physicists, the laws of convective 

 discharges from points do not seem to have received attention 

 from any experimenter. Thus I have not been able to find a 

 satisfactory answer to the following elementary inquiry, viz.- 

 Under given conditions, at what distance vcWl a pointed conductor 

 connected with the earth begin to neutrali.-e the electricity of an 

 insulated conductor by the convective discharge of the opposite 

 kind of electricity from the point ? 



In short, the whole subject of the "power of points," although 

 one of the best-establi.shed and most conspicuous phenomena in 

 electricity, is sadly in need of experimental investigation. This 

 class of electrical phenomena is pretty much in the same condi- 

 tion in which Franklin left it more than a century ago. 



Berkeley, California, January 1 John Le Con'TE 



[Mr. Preece has shown by considering the area between the 

 conductor and the charged cloud as an electric field mapped out 

 in equipotential surfaces and lines of force, t'oat "a lightning- 

 rod protects a conic space whose height is the leugth of the rod, 

 whose bae is a circle having its radius equal to the height of 

 the rod, and whose side is tlie quadrant of a circle whjse radius 

 is equal to the height of the rod." — Pkil. Mag., December, 

 1880.- Ed.] 



Localisation of Sound 



My friend the Rev. H. J. Marston, Second Master of the 

 School for Blind Sons of Gentlemen at Worcester, has com- 

 municated to me some very singular instances of the power of 

 localising siuud possessed by bUnd boys. 



One of the games in which his pupils most delight is that of 

 bo7vls A bell is rung over the nine-pin* ju^t as tlie player is 

 ready to throw the bowl, when, totally blind as he is, he delivers 

 it with considerable accuracy of aim. Mr. Marston vouches for 

 the fact that it is no uncommon feat for a boy to strike down a 

 single pin at a distance of forty feet three times in succession. 



