404 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
winds, freshening towards evening, gives a long, regular curve. The air curve shows j 
the effect of the clouds of yesterday p. m. until 6 hours, when the same contrast with ' 
the New York and Boston curves appears as was shown in the case of yesterday. 
August 20, 1889. (Table 15 and Plate 15.) — We reached the shore end of line J j 
(J 3, No. 81) at 6 hours, went to the westward and commenced line K with K 3, No. 82, 
and then proceeded southward along that line. The fact that we were going inshore 
from warm to cold water and then out again from cold to warm water is well shown 
by the curves between 5 and 12 hours. The influence of the warm water upon the air | ; 
after 12 hours is clearly shown, as it was intensified by,a thick mist between 13 and | 
18 hours, during the time in which the elevation of the curve became so pronounced. | j 
The Boston curve is abnormal, but no explanation is at hand for this fact. | i 
August 21, 1889. (Table 16 and Plate 16.) — The air and water curves follow one 1 
another very well except at 7 hours, where the sudden rise was due to the rays of the j 1 
sun striking the side of the thermometer box directly. This was also the case between | 
10 and 18 hours, where the greatest increase in the temperature is noticed. Curiously | : 
enough this part of the air curve follows the New York curve closely. The water | ‘ 
temperature was very high during all this time, reaching a maximum of 76.2° F. at | 
15 to 16 hours, while we were going eastward from line Kto line H. At station K12, | 
No. 91, the specific gravity was 1.0259, and at station H 12, No. 92, it was 1.0258. An I | 
almost total lack of wind, the warm water, and the impossibility of protecting the I 
thermometer box from the sun’s rays while moving to the eastward is a sufficient I 
explanation for this abnormal air curve. r 
August 22, 1889. (Table 17 and Plate 17.) — The sudden increase in the air tern- I 
perature was again due to the sun’s rays upon the thffimometer box. From 8 hours 
onward we were surrounded by a thick haze, and from 12 hours onward we had no | j 
wind to speak of. From the fact that we supposed we were passing up line H (see <« 
map) and found that Gr 7 was our position at 9 hours this a. m., and then that the | 
plotted log would not bring us far enough to the westward to reach the position we ; ' 
occupied in the p. m., we have reason to believe that there is some probability in the 
theory advanced in connection with the work of August 18. The descent in the j 
water curve as we approach the cold waters of the shore is again apparent. We went j 
into Wood’s Holl to refit and replenish our stores and were there for three days. On I 
August 27 we started again and were forced back to Newport by the rough water. 
We made two attempts to renew our work, but were not successful until August 31. 
August 31, 1889. (Tables 18 and 19 and Plate 19.) — We made a beginning upon 
lineK, but were forced to give it up and run back at midnight on account of the high 
wind and rough water. The long-continued heat, as shown by the New York curve, 
is noteworthy. 
Septi mber 1, 1889. (Table 20 and Plate 20.) — The attempt this a. m. was made j 
upon line J with the same result. Not enough work was done to make records of any 
ralue. Tbe meteorological records and plates are given simply because of the sta- 
■"ions recorded in the course of the work. 
September 2, 1889. (Table 21.) — Nothing could be done until evening, when we 
started upon line H. We reached open water at 19 hours, and our meteorological 
record began at this time, giving too short a record to warrant a plate. The first 
station, HI, No. 114, was reached at 23 hours. 
