10 
METEOROLOGICAL RESULTS OF LAST CRUISE OF CARNEGIE 
Table 13. Comparison of 12-hour waves of atmospheric pressure at approximately same latitudes on 
islands and at sea 
(Carnegie values from mean of observations on all days when ship’s position at local mean noon was 
within 5° north or 5° south of given Jabba) 

Station 





2 mm mm 
Lerwick 60.1 N 2 years May -Aug. 122 0.102 
Carnegie 17 days July -Aug. 143.3 - 21.3 0.230 - 0.128 
Jersey 49 N 10 years May-Aug. 139.2 0.250 
Carnegie 50 N 40 days May -Aug. 147.9 - 8.7 0.129 +0.121 
Ponta Delgada 37.8 N 7 years May-Sep. 148.1 0.384 
Carnegie 40 N 35 days May-Sep. 138.5 + 9.6 0.212 +0.172 
Taiwan 25.0 N 10 years May-June 161.6 0.633 
Carnegie 30 N 38 days Aug.-Sep. 147.8 +13.8 0.420 +0.213 
Port-au-Prince 18.5 N 6 years May, Aug.-Oct. 161.3 0.778 
Guadeloupe 16.0 N 9 years May, Aug.-Oct. 153.9 0.668 
Means 17.2N 157.8 0.721 
Carnegie 20 N 32 days May, Aug.-Oct. 155.2 + 2.6 0.625 +0.096 
Manila 14.6N 12 years May, Aug.-Nov. 161.2 0.895 
Jaluit 5.9 N a May, Aug.-Nov. 166.1 0.848 
Means 10.2 N 163.6 0.870 
Carnegie 10 N 44 days May, Aug.-Nov. 160.6 + 3.0 0.761 +0.109 
Batavia 6 Ss 2 Nov., Jan.-Apr. 158.5 0.965 
Samoa 13.8 S 6 years Nov., Jan.-Apr. 160.0 0.720 
Means 9.98 ce 159.1 0.842 
Carnegie 10 S 45 days Nov., Jan.-Apr. 153.7 + 5.4 0.809 + 0.033 
Mangarewa 23.3 S 2 years Nov., Jan.-Mar. 168.2 0.676 
Mauritius 20_S a Nov., Jan.-Mar. 162.3 0.722 
Means 21.6S 165.1 0.698 
Carnegie 20 S 33 days Nov., Jan.-Mar. 160.6 + 4.5 0.759 - 0.061 
Easter Island 27. 6S 1 year Nov.-Feb. 166.3 0.493 
Carnegie 30 S 22 days Nov.-Jan. 150.0 +16.3 0.479 +0.014 
Mean\difference}incamplitude)(island=sea)uc---s-s-sesceeeeeseeeeee ete ee neat eee ne Eee aee eee +0.063 

2 Not given in reference. 
these seventeen days were cloudy to overcast, with fre- 
quent light mist, or otherwise affected by meteorological 
conditions which would lead one to expect a small ampli- 
tude [15]. In addition, the probable error of the seven- 
teen observations of the Carnegie in latitude 60° north is 
estimated to be 0.073 mm, whereas the difference be- 
tween the values for Cg as computed from the data at 
Lerwick and those of the Carnegie is only 0.128 mm. 
For mean latitudes other than 20° south, the days in 
which observations were made by the Carnegie are more 
symmetrically distributed among the months used in the 
computations; therefore it can safely be assumed that 
the value of -0.06 mm, computed as the mean difference 
between island stations and oceanic locations, is proba- 
bly near the true value. 
Simpson [5, p. 12] maintains that the phase of the to- 
tal semidiurnal wave at any position on the earth can be 
quite closely determined by 
tan A = 
0.937 cos3¢ sin 154°+ 0.137 (sin2¢ - 1/3) sin(105° - 2A) (5 
0.937 cos3¢ cos 154°+ 6. 137 (sin2¢ - 1/3) cos (105° a 
As in equation (4), the righthand members of equation 
(5) contain as variables only the latitude ¢, and the lon- 
gitude A. For the purposes of comparison, the mean 
longitudes corresponding to the Carnegie ranges of lati- 
tude were supplied in this formula and the phase angles 
for the mean positions computed. The computed and 
observed phases given in table 14 indicate no systematic 
difference between the observed phase angles and those 
calculated after Simpson’s formula. Unfortunately, lo- 
cal mean time instead of apparent or ship’s time was 
used in all computations for the Carnegie. According to 
the equation of time this error could not be greater than 
8° in phase (16 minutes of time) and in most cases it 
would be considerably less than this. 
In this connection it is interesting to compare the 
mean yearly phase angles of the 12-hour wave at Easter 
Island, Samoa, and Jaluit with those computed after 
Simpson’s formula per equation (5) [14]. These results 
(table 15) indicate that in each case the observed phase 
angle is greater than the calculated; for example, the 
maximum amplitude occurs earlier than is indicated by 
Simpson’s results. When the mean yearly phase angles 
for the island of Jersey (49° N, 2° W) and for Lerwick 
