OP ARTS AND SCIENCES. 81 



there was no squall connected with the storm, the wind remaining 

 fresh southerly or southwesterly during the rain, except at Haverhill 

 and Newburyport, on the northern side of the storm, where it was 

 reversed to northeasterly. 



d. The fourth storm probably consisted of several parts : it extended 

 over central Connecticut in the middle afternoon, but its motion cannot 

 be well determined for lack of observations. Fine festoon clouds 

 were seen on the after side of its cirro-stratus cover in the southern 

 part of the State. 



e. The last and most extensive storm of the day began in southern 

 New Hampshire about 14'\ and moved to the southeast about twenty- 

 five miles an hour. It does not seem to have been a well united 

 storm, but consisted of numerous loosely connected parts, from which 

 showers of varying strength fell. At Lowell storms c and e together 

 yielded about four inches of rain, one gauge reporting 3.93 and another 

 4.20 inches; and of this the greater part fell from 15\15 to 16^.15, 

 without much wind, althougli the clouds were much agitated as the 

 storm approached. Violent winds appeared only in a limited district 

 north of Worcester, about Holden and Princeton, and hail fell a little 

 farther north, at Westminster. -' 



Mr. Clayton has constructed several synoptic charts for storms c 

 and e, as far as the observations would allow, showing the extent of 

 the rain area and the direction of the wind at certain even hours. 

 Owing to the failure of many observers to carry out their instructions 

 literally, it was found impossible to make these charts very complete; 

 but the later ones show a tolerably distinct cyclonic motion of the 

 winds within and around the oval rain area, implying that the thunder- 

 storm area possessed a gentle spiral rotary circulation on a small scale, 

 as has been determined for storms of this kind in Europe. On this 

 date the very moderate gradients gave good opportunity for the forma- 

 tion of such " secondary depressions." Especial effort will be made 

 to accumulate data for such synoptic charts during the coming summer, 

 as they promise to be of particular interest in connection with the 

 smaller local storms. 



July 30 possessed no well-defined centres of high or low pressure ; 

 its temperature was lowered (mean max. 75°) by the passage eastward 

 of the area of low pressure that controlled the weather of the preced- 

 ing day. A few scattered reports were received, the only storm of 

 any severity being in central New Hampshire about 8^. 



July 31. This was a day of generally equable distribution of press- 

 ure, of rather high temperature, and with winds of indeterminate circu- 



