19G STATE HORTICULTURAL SOCIETY. 



of eacli other. They are also apt to recur in the same locality. They are 

 most frequent in the hottest part of the year and of the day. 



That the tornado is a small centripetal storm is indicated by several facts. 

 Trees and other objects are thrown down in the directions indicated by the 

 arrows in diagram (2.) At the outer edges of the path their trunks are paral- 

 lel to the path, in the center they lie across the path. Light objects are fre- 

 quently seen to be taken up into the air and to whirl upwards in a spiral 

 direction. 



The preceding are the local winds observed in Michigan. In the prediction 

 of them the science of meteorology cannot help us at present. The meteorologi- 

 cal stations are too scattered to be of use in the study of so local phenomena. 

 A tornado may form in the center of Michigan, travel many miles, and do 

 great damage without there being any indication of its presence in the instru- 

 ments of the observers at Detroit, Port Huron, Grand Haven or Chicago. 

 Were we able to multiply observers until we could have several in each county, 

 we might detect severe local storms in their formation and forewarn those 

 endangered by them. It is at least an experiment well worth undertaking. 



II. CYCLONIC WINDS. 



The cyclone is a storm-area with a centripetal system of winds of large 

 dimensions. In the weather reports it is usually referred to as an area of low 

 pressure, and it has been found that the centripetal winds pour in toward such 

 areas. Each cyclone is accompanied at a distance by one or more anti-cyclones 

 or areas of high pressure. 



e — Cyclones. The winds of the temperate zone have long been described as 

 variable, and it is only recently that it has been found that their variability is 

 subject to a definite law and their changes can be predicted for a day or two in 

 advance. A cyclone can be most easily conceived of as a large and mild tor- 

 nado. It is an area of low pressure, usually oval in form, with its long diame- 

 ter nearly north and south, and is surrounded by concentric gradients of suc- 

 cessively higher pressure. The existence of such dishes of low pressure is suf- 

 ficiently interesting, but the value of the knowledge of them becomes greatly 

 enhanced when we know that they come to us from the west and travel from 

 us toward the east in tolerably uniform paths and at a velocity on the average 

 of 30 miles per hour, and that they carry with them definite conditions of wind 

 and weather. They usually come from the Pacific or the northwest, travel 

 eastward on a curved path convex toward the southwest and disappear over the 

 mouth of the St. Lawrence. They are storm-areas and are accompanied by 

 clouds and a higher temperature. 



But the feature of interest here is the centripetal system of winds which 

 accompanies the cyclone. It is like that of the tornado, but while the tornado 

 system of winds turns indifferently with or against the hands of a watch, that 

 of the cyclone in the north temperate zone turns in a direction opposed to that 

 of the hands of a watch: It also differs from the tornado in its dimensions. 

 The former controls the wind only for a few rods on each side, while the latter 

 reaches out hundreds of miles from its center. While this storm-center is in 

 Iowa or Nebraska or at Quebec or near James' Bay the winds of Michigan may 

 still be controlled by it. The paths of the centers of the cyclones usually lie 

 south of Michigan in summer and over northern Michigan in winter. For 

 south Michigan we will therefore have, in summer, winds beginning at south- 

 east and passing through the east to the northeast; in winter beginning at 

 southwest and passing through west to northwest. 



