14 DEPARTMENTAL REPORTS. 



the fact that no attempt has ever been made to constrnct any nor- 

 mals, or to determine the seasonal variation of the precipitation as 

 depending upon this element. These computations will necessarily 

 involve a careful treatment of the wet and dry bulb temperatures and 

 a consideration of the troublesome psychrometric problems that are 

 involved. 



LOSS OF LIFE IN THE UNITED STATES BY LIGHTNING. 



In Bulletin No. 30 the information collected from all parts of the 

 country during the past ten years has been brought together and 

 summarized by Prof. A. J. Henry. It is shown in this publication 

 that destructive lightning strokes occur with greater frequency in 

 some parts of the country than in others ; that the region of greatest 

 frequency is in the Ohio Valley, the lower lake region, and the mid- 

 dle Atlantic States, and that, considering the sparsity of the popula- 

 tion, the number of fatalities in the middle Rockj^ Mountain region 

 and the upper Missouri Valley is surprisinglj^ large. A study of the 

 data has also enabled the Bureau to formulate a few simple precau- 

 tions against danger from lightning stroke that are here reiterated: 



It is not judicious to stand under or near trees during thunder- 

 storms, in the doorways of barns, near chimneys and fireplaces, or tim- 

 bers that lead directlj^ to the room. Neither should one stand near 

 the point of entrance of telegraph and telephone wires. (The latter 

 should invariably be provided with lightning arresters and ground 

 wires.) It is not advisable to huddle under wagons, thrashing 

 machines, or under frame structures surmounted by a flag pole. A 

 wire clothesline should not be attached to a dwelling house under any 

 circumstances; rather suspend it between two neighboring trees or 

 posts. 



WIND VELOCITY AND FLUCTUATIONS OF VS^ATER LEVEL ON LAKE 



ERIE. 



Strong westerly winds on Lake Erie pile up the water in the harbor 

 of Buifalo, at the eastern end of the lake, the rise in level at times 

 being so great as to be detrimental to navigation and injurious to 

 wharf i^roperty. The establishment under the direction of the Chief 

 of Engineers, U. S. Army, of self-recording water-level gauges in 

 Buffalo Harbor and at the western end of the lake, and the hearty 

 cooperation of that ofl&cial with the Bureau, has made it possible for 

 Professor Henry to study the relations between the force and direction 

 of the wind and sudden changes in the level of lake waters. 



It was found that with westerly wind velocities of less than 50 miles 

 per hour at the eastern end of the lake the changes of level in Buffalo 

 Harbor were not great enough to menace navigation; when, however, 

 the velocity of westerlj^ winds passes beyond 50 miles per hour, wharf 

 property is always more or less exposed to danger from flooding. The 

 height to which the water will rise depends partly upon the strength, 

 duration, and suddenness of the westerly winds and partlj^ upon the 

 season of the year. The winds of the warm season seldom prevail long 

 enough to cause an overflow. It was also found that while the crest 

 of the rise in lake level and the maximum velocity of the wind gen- 

 erally coincided in point of time, the water would begin to fall as soon 

 as the crest was reached, regardless of the force of the wind, and that 

 it would continue to fall and then rise again, in a series of oscillations 

 up and down, until the normal level was restored. 



