290 Direction of the Wuid^ New Haven^ Conn. 



These conclusions appear to be contirniecl by a coin]>arison of the 

 directions of the Avind at New Haven and Wallingford during the six 

 wanner months of the year. In the months of April and September 

 the diurnal change of direction is much greater at New Haven than 

 at Wallingford, the wind being almost exactly North in the morning, 

 and nearly South at the hottest })art of the day. In May and August 

 the wind at both stations is nearly North in the morning and South 

 in the aftenioon, but with this diiFerence, that at Wallingford the 

 Westerly motion exceeds the Easterly, while at New Haven the 

 Easterly motion exceeds the Westei'ly. It seems probable that the 

 latter effect is coniined to places but little elevated above the level of 

 the sea. In June the curves at the two stations are quite similar ; 

 while in July the diurnal change is much the greatest at New Haven. 

 We conclude therefore that the New Haven observations are not 

 inconsistent with the explanation heretofore given of the winds at 

 Wallingford, and that the peculiarities of New ^laven result from 

 local causes, among which are to be enumerated its low position, and 

 perhaps also the shallow water of Long Island Sound, with Long 

 Island on the south of it. It is suspected that these local Avinds at 

 New Haven are of the nature of counter cun-ents, analogous to the 

 counter currents observed along the banks of rapid rivers, especially 

 where the banks are considerably indented. 



Velocity of the Wind. 

 In the year 18(50, a Robinson's anemometer, made by L. Casella of 

 London, was procured by Prof. Elias Loomis for Yale College. The 

 hemispheres are three inches in diameter, the distance between the 

 centers of the opposite cups is 1;V5 inches, and the distance traveled 

 by the wind is recorded up to 500 miles. The anemometer was 

 erected upon one of the towers of Graduates' Hall at an elevation of 

 65 feet from the ground, wdiere the exposure was entirely unob- 

 structed. In December, 1863, regular observations were commenced 

 l)y Prof. Loomis, and have been continued to the present time. The 

 observations were made at intervals of one, two or three days, 

 according as was found convenient, the object being simply to deter- 

 mine the average velocity of the wind for each month of the year. 

 It was soon found that the velocity indicated by the observations was 

 smaller than had been expected, and it was suspected that the instru- 

 ment was not entirely reliable. After the observations had been 

 continued for two or three years. Prof. Loomis decided to procure a 

 second instrument from a different maker. He accordingly requested 



