402 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
to represent easterly and nortlieasterly weather, as contrasted with the westerly M 
weather which was shown in yesterday’s curves, they are hard to explain. Even this 
explanation might not hold good for the conditions nearer the coast, because, as will 
be seen from the water curve, we were passing through a very distinctly warm baud m 
of water from 9 to 13 hours (73.4° F.), and the specific gravity was also high (Station H 
D 9, sp. gr. 1.0260). This, taken with the fact that the wind was blowing with about J 
a force ot 4 from the northeast, complicates the problem very much. 
August 9, 1889. (Table 11 and Plate 11.)— We completed line D in the a. m., but | 
after starting line E with three stations the water became so choppy that we were 
forced to head for the shore after finishing Station Ell. The covered sky, warm 
water, and westerly winds account for the height of the air curve, which follows the 
curves of New York and Boston pretty well, the disturbance set up by the westerly 
wind being most strongly felt in New York, next on the Gramptis, and last of all in 
Boston, as was natural. The rise in the water-temperature curve between 8 and 11 » 
hours as we were moving westward in the Gulf Stream is interesting, although it was ' 
undoubtedly accentuated by the atmospheric conditions. j 
We went in to Wood’s Holl, where we were detained 4 days by stormy weather. j 
August 17, 1889. (Table 12 and Plate 12.) — We left Wood’s Holl on August 16, ' 
but were forced to put back into Tarpaulin Cove on account of rough water. j 
The wind was very light in the a. m. of August 17, and it took ns until noon to 
reach Station G2. We passed into warm water again between G3 and G4 at 15 |j]| 
hours, after which time the air and water curves follow one another very closely. In 
the contrast of air curves all is normal until 19 hours, when the air curve of the ' 
Grampus stops in its downward course and the influence of the warm water upon the , , 
air temperature places the curve in strong contrast to those of New York and Boston. _ 
August 18, 1889. (Table 13 and Plate 13.) — The irregularities observable in the 1 
contrast of the air and water are partly accidental ; for example, that at 8 hours was j 
caused by the fact that our thermometer box was directly exposed to the strongest ij 
glare of the sun across the deck in the direction of the louvre openings in its sides, j 
This condition was corrected upon noting the reading. Then there are two 
depressions, one at 8 hours ^ud the other at 12 to 14 hours, which show the result of 
cloudless sky followed by clouds at these hours. The maximum water temperature 
was reached at 15 hours, station G 12 (74.5° F.) accompanied by a specific gravity 
of 1.0257. The air and water curves as contrasted in the p. m. show the result of 
heavy clouds upon the temperature of the air. The contrasts of the air curves upon 
the Grampus and those of New York and Boston are very nicely shown when 
allowance is made for the variations noted above. The Grampus curve would present 
a comparatively slight range, while the land curves would give most decided maxima 
and minima. It will be seen by a glance at the column containing the list of stations 
that we were still upon line G, in continuance of our work of the previous day. Upon 
comparing our plotted line (by log record) with our observed latitude of yesterday at 
16 hours, station G 4, No. 61 (see also map of stations), we found that we had drifted 
to the westward. The wind force only enabled us to creeii along and we were thus 
fully exposed to the action of the currents as we slowly worked across them. The 
drift to the westward was such as might have been expected, for we were undoubtedly 
in a part of the ocean where the cold current would take us in that direction, but we 
