\ 



SCIE 



[Entered at the Posi -Office of New York, N.Y., as Second-Class Matter.J 



A WEEKLY NEWSPAPER OF ALL THE ARTS AND SCIENCES. 



Eighth Year. 

 Vol. XVI. No. 3i 



NEW YORK, July 11, 1890. 



Single Copies, Ten Cents. 

 3.50 Per Year, in Advance. 



■TOENADO LOSSES AND INSURANCE. 



The destruction of life and property by a tornado must 

 ever be regarded as the most important fact from a practical 

 standpoint, and in many respects this feature of the subject 

 is involved in greater confusion and uncertainty than any 

 other. As has already been said, such a violent storm 

 ought to be exceedingly well defined, and the amount of 

 loss determined within a small fraction. One reason for 

 this uncertainty has been the lack of persons desirous of 

 obtaining a complete list of property loss and damage by 

 tornadoes. We have an excellent illustration of an ojipo- 

 site state of affairs in the estimates of loss of property by 

 fire. Here there are hundreds interested in obtaining the 

 exact loss; and it is believed that in this country the loss by 

 fire, whether of insured or uninsured property, is known 

 within four per cent of the actual loss. A most serious dif- 

 ficully in making such estimates has arisen from a vague 

 and indefinite idea of what a tornado is, and how it should 

 be described. We are told, for example, ''if we care for the 

 name 'tornado' to define a distinct class of local storms, then 

 the funnel-shaped cloud, as shown by a distinct rotary 

 movement of the wind, or by peculiar destruction of prop- 

 erty, should be made the condition of classification. Both 

 for the purpose of study and practical results, this manner 

 of distinguishing the tornado is desirable. It leaves no 

 doubt as to where the line should be drawn, and recognizes 

 a peculiar and important class of meteorological phenomena, 

 independent of their effect upon life and property, which it 

 is quite well known that they can destroy if given an oppor- 

 tunity." We must take most serious exception to these 

 views. It is manifestly only by the effects displayed by 

 these storms at the earth that we can classify or discuss 

 them. The most violent commotions, the most surprising 

 shapes and appearances of clouds meeting or rolling over 

 each other, might have a passing interest, but surely they 

 would be practically of no account if they did not reach the 

 earth and there affect life and property. 



Descriptions. 



We have already seen that in the earlier history of these 

 outbursts they became familiarly known by a definite name. 

 For example: the New Brunswick (New Jersey), the New 

 Haven (Connecticut), the Stow (Ohio), tornadoes are definite 

 phenomena and extremely localized. In later times the 

 most terrible tornado but one that has occurred in this coun- 

 try was that at Grinnell, lo. To call this the Poweshiek 

 County tornado, and to say that it had a path 450 to 600 

 feet wide, takes from it almost entirely its definiteness and 

 extreme destruction. In another instance a tornado is de- 



scribed as causing a loss of $4,000,000, — the greatest in this 

 country, and, for that matter, in the world. This tornado 

 is given a path 2,640 feet wide, and passing through Rock, 

 Hennepin, Ramsay, and Washington Counties, Minn., and 

 through St. Croix, Polk, Barron, Chippewa, and Price 

 Counties, Wis. It is not intended, of course, to convey the 

 idea that there was a clean sweep a half-mile wide through 

 these nine counties; but, to one familiar with the more ter- 

 rible storms, such a description would convey an idea of a 

 most appalling disaster. A most careful study of this nine- 

 counties tornado has revealed the most astounding result, 

 that there were only two towns injured. The principal loss 

 was at Clear Lake, Wis., $139,100; and the other town was 

 Marine, Minn., with a loss of about $65,000. The total direct 

 loss by this tornado in all the counties was not over $250,000, 

 or one-sixteenth the published loss. It should be insisted on, 

 by all means, that every tornado should have a definite lo- 

 cal application and a name. In some cases where the coun- 

 try is thinly populated, and houses are destroyed here and 

 there through many townships, it will be necessary to group 

 all such hamlets in the name of a county or of the principal 

 town visited, but this should be rarely resorted to. 

 Indirect Loss. 

 Much confusion has arisen from grouping together losses 

 by the tornado proper, and by floods and hail which accom- 

 panied it; also by considering losses to orchards, crops, 

 fences, stock, etc., in connection with that to houses and 

 buildings. Certainly, in calculating risks for tornado insur- 

 ance, and in studying the definite losses, we should cata- 

 logue only the definite and direct loss to structures. It is 

 not a little remarkable, that, while the descriptions of torna- 

 does have tended to vagueness and indefiniteness as to ex- 

 tent of path and destructiveness, no such difiBculty has been 

 encountered in photographing the effects of a tornado. The 

 most completely destroyed houses and blocks, and scenes 

 exhibiting the severest violence of tlie storm, have always 

 attracted the photographer. This country has had over two 

 thousand tornadoes since 1873, but we would be very much 

 deceived if we thought that the scenes of desolation depicted 

 in half a hundred photographs were experienced at more 

 than two per cent of these. The worst of these photographs 

 are taken from only two or three tornadoes. 



Tornado Lists. 



These are being published from time to time, and in the 

 main are very untrustworthy, from a lack of care in colla- 

 tion, and a desire to exhibit some preconceived idea as re- 

 gards the tornado. A careful sifting of the lists has shown 

 that many of these tornadoes were of very slight account, 



