362 SURVEY OF THE INTERTROPICAL 
182a carried away by the wind, break short without 
July 24. splintering. 
We passed Capes Gloucester and Upstart 
25 . during the night and early part of the next morn- 
ing. Between the latter cape and the low projec- 
tion of Cape Bowling-green, we experienced an 
in-draught of three-quarters of a knot per hour. 
This also occurred last year ; and it should be 
guarded against by ships passing by : for the land 
about the latter cape is so low that it cannot be 
seen at night. 
From the period , of our entering among the 
Northumberland Islands, the weather, although 
fine, had been more than usually hazy ; the wind, 
during the day, blew moderately from S.b.E. and 
South, and veered towards night, to S.E.b.E. and 
E.S.E. ; but, when we passed Cape Cleveland, 
it blew a fresh breeze, and was so very hazy, 
that we could not take advantage of our vicinity 
to the coast by verifying or improving any part 
of our former survey, except the outer or seaward 
side of the Palm Island Group, near which we 
passed in the evening. 
26 . The next morning we were off the southern- 
most Barnard’s Island, and, as the coast between 
Double Point and Fitzroy Island had not been 
satisfactorily laid down on the previous exami- 
