OF ARTS AND SCIENCES. 63 



There is some indication that the greater changes of temperature 

 are near the apex of the curve, as would be expected. We are for- 

 tunate in having two records close to the apex that show the greater 

 part of the temperature fall to have been very sudden ; one is at 

 Hensonville, N. Y., as follows : — 



Time 9.13 9.20 9.25 9.27 9.30 9.35 9.40 9.50 10.8 13.30 



Temperature 85° 82° 81° 71° 70° 71° 72° 73° 74° 92° 



Wind — — Strong. — — — Light. V. light. — — 



The second is from the City Engineer's office, Providence, R. I., 

 where a self-recording thermometer (pattern of Richard Freres) gives 

 a curve showing a fall of 13° in thirty minutes, the greater part of 

 the change being accomplished in a small fraction of this time. A 

 second fall at 1G.35 was caused by the large afternoon storm of this 

 date. From the same source we have the automatic record of a Dra- 

 per's mercurial barometer, showing the abrupt increase of pressure 

 characteristic of thunder-storms, from 12.33 to 12.40 the pressure 

 rising from 29.695 to 29.730; and also an automatic wind record, 

 showing the sudden wind-squall at Hope Reservoir, Providence, 

 blowing at a rate of three miles in the five minutes from 12.33 to 

 12.38; before and after the squall, the velocity was ten to twelve 

 miles an hour. Another barometric record, from a Richard Freres 

 aneroid, was obtained from Plymouth, exhibiting an abrupt rise of 

 one millimeter = 0.04 inch at a few minutes after 13.00. It is greatly 

 to be hoped that self-registering instruments — especially barometers 

 — may be more generally kept at our stations: considering the great 

 value of their records, their cost is not relatively high. Detailed 

 observations of the clouds of this storm were so few that no generali- 

 zations can be made from them. 



The afternoon storm of July 21 (Fig. 8) came from New York, 

 crossing the Hudson mostly below Albany. The temperature in 

 New England had risen over 90°, with southwesterly winds, in many 

 places after the passage of the noon squall ; the heat became very 

 oppressive before the larger storm arrived. The thin front of the 

 high cirro-stratus was visible in the northwest at least three hours 

 before the rain began, and as it obscured the sun the temperature fell 

 slowly. The front edge of the lower clouds stretched S.W. to N.E. 

 in Connecticut, where the short-lived squall-wind attained a destruct- 

 ive strength, and was accompanied with hail at a number of points ; 

 but, unfortunately, our observers there were few, and many records 

 had to be gathered simply from the current reports of the newspapers. 



