THE LATITUDE-VARIATION TIDE. 
115 
ft. in. mm. 
^ = 0.051 = 0.61 = 15.5, 
= 98° ± 24°, 
Period = 437.6 ± 2.0; 
aud I reserved the second term for further consideration. 
Thus far the assumed period, in conformity with which the 
observations had been distributed into groups, was the then 
Chandler period of 427 days. In this distribution (see the 
formulae) r = l h , v = 1281, m = 8, N = 6. I computed and 
applied all the corrections for imperfect elimination of other 
tides (the C" and S" terms in (3) and the y" and <s n terms 
in (6)1, finding them sensible for the solar annual and semi¬ 
annual, the lunar declinational diurnal, and the lunar semi¬ 
diurnal. 
On communicating these revised results to the superin¬ 
tendent, he directed me to make two other distributions of 
the same series. I selected the periods 440 days and 456 
days. In the 440-day distribution t = l h , v = 1320, m — 8, 
N = 5, and I had to apply corrections for the solar annual 
and semiannual, the lunar declinational diurnal, and the 
larger lunar elliptic diurnal. In the 456-day distribution 
t = l h , v = 1368, m = 8, N = 4, and I had to apply correc¬ 
tions for the solar annual and semiannual and the lunar 
semidiurnal. The results for a and £ from the several inde¬ 
pendent sections of each distribution are brought together 
in the following table: 
No. of section. 
427-day distribu¬ 
tion. 
440-day distribu¬ 
tion. 
456-day distribu¬ 
tion. 
a. 
e. 
a. 
e. 
a. 
E. 
1. 
ft- 
0.0784 
O 
77 A 
ft- 
0.0644 
O 
187.3 
ft. 
0.0812 
528 
o 
150.8 
2. 
413 
134.4 
668 
194.1 
51.2 
3. 
307 
274.4 
239.0 
431 
180.0 
458 
113.1 
4. 
710 
752 
198.2 
465 
10.1 
5. 
0.0306 
285.0 
079 
*[33.7] 
122.1 
628 
— 11.8 
6. 
0.0785 
279 
— 160.2 
7.. 
0.1250 
— 230.3 
* Rejected for indetermination. 
