* 
Magnetic Dip in the United States. 89 
parison of the observations eighty eight values of the same ele- 
ment, the accordance of which with each other will enable us to 
judge of the confidence which may be placed in a single obser- 
vation. It will be seen that the dip is usually the greatest when 
the face of the instrument is east. Subtract then the dip ob- 
served with the face west from that found with the face east 
with the same position of the needle. For example, in the first 
observation 72° 47’ from 72° 5’ gives — 42’, and so of the rest. 
We thus obtain the following table, which exhibits the observed 
values of twice the zero error. : 
42 +14lgri}— ‘14 7+ 2+ ‘1420 +1134 4 |4+ 8 
+19 |— 34 3/4 3+ 8+13)— 2}4+15 |+ 7 |+ 7 |4+12 
+6 |+ 7-24 54 7— 4)- 74+ 9 [+ 5.5/- 3.5419 
+ 5 |-10+ 44 7410+ 1]—12/4+ 9 |413.5)-13 |4+16.5 
+4 + 4—- 2+ 5 0+ 3/— 14 4 |4+ 1 /410 |+ 8 
+1 |— a+ aly 9417+ 5 O+ 3 |+ 9 [+ 5.5/4 3.5 
+ 3541114 6+ 34 54 s|— 4/412 |4 8 |+10 |+ 6 
+ 25/4 84104104 1/410/2 4l- 3.514 6.5- 2514 65 
The mean of all these observations is +4/.5, which may be 
taken as equal to twice the zero error. The difference between 
this and the preceding observations will show the errors of the 
observations, which when classified are as follows : 
‘ J / ! 
+15.5)4+-8'5] 4.5.5|4-3.5/+2.5|+1.5|—0.5|—1.6|—-3.5/—5.5|— 8.5 
14.5) -7.5| 6.51 3.5] 2.5) 1 | 1.5) 3.5) 5.5) 8.5 
14.5). 7.5) 5.5) 3.51 2.5). 1 .5| 1.5} 3.5) 6.5) 8.5) 
125, 7 | 5.5} 3.5] 251 .5| .5] 1.5] 3.5) 6.5) 11.5 
12 | 65] 45, 3.5] 2 a 6 Cid) 3.0 6 148 
10.5] 6.5 4.51 3.5! 2 5 1.0) 2. | 4.5) 7.5) 16.5 
9.5) 5.5, 4.5] 2.5) 1.5] 9.6) 1.0| 2. | 45) 8 | 17.5 
| 9 | 5.5} 4:5] 25} 1.5) 2B) 1.5) 2.5] 5.5] 8 | 465 
By far the greatest error here is —46/.5, which was obtained 
ftom the first observation. The difference between the readings 
With the face of the instrament and needle both east, and that 
With the former west and the latter east, instead of being — 42’, 
should be +.4/.5. This does not inform us which reading is most 
in error ; if, however, we apply the correction to 72° 5’, making 
tread 72° 61/5, the resulting dip would be 71° 54/.1, corre- 
‘ponding nearly with the other observations at the same place. 
Vol. xx, No. 1.—Oct.-Dec. 1840. 12 
