| 
Cad 
, 
€ 
ON SEISMOLOGICAL INVESTIGATIONS. 
Forming groups of these we obtain the mean values of Table If. :— 
TaBLE II. 
No. in Group dese! |Mean Correction | (1913) (22) 
ac Ikon Ga a Cees AGG GP 
1 6-4 +14 | +4 | =e | 
4 13-2 +10 | + 8 | — § 
| 4 18°5 +3 | +38 — 8 
4 33:9 — 4 —10 | —13 
2 53°6 — 6 | — 7 — 6 
6 68:2 —15 —4 — 6 
| 5 72:1 —10 +1 — 8 
Observations of PR, and SR. 
But a knowledge of S—P will not suffice to correct the separate 
tables. We must find some other way of doing this, and recourse was 
had to the observations of PR; and SR. 
We shall first assume that these represent reflections at the mid-point 
without loss of time. Thus PR, is a P wave reflected at the mid-point 
(M) between the epicentre (E) and the observing station (OQ) so that 
EO=2EM=4, 
and the time for PR, is twice that for EM or A/2. 
In Table III. the first column gives A and the second gives the observed 
value of 3(S-+P)—PR,, or as we may write it }(S—P)+P—PR,, which 
is nearly constant. The observations were collected from the records 
TaBie III. 
a | Obsd | Tables | 0-C, | o—0, | 
| Ss. s. s. s 
23 77 98 ae | —15 | 
28 15 102 a7 —14 
33 80 103 —23 = 
38 83 100 =i ) —2 
43 88 95 | Ea +5 
48 91 88 +3 +10 
53 95 85 +10 +10 
58 98 83 +15 | +10 
63 102 82 +20 ie 
68 106 j 83 9s +11 
72 109 ! 87 +422 +11 
78 109 92 oy +9 
83 106 92 +14 +9 
88 105 | 96 + 9 aa 
102 | 101 eT +4 
of several observations, especially Pulkovo, in the years 1914 and 1915, 
and are fairly numerous for all but the small values of 4. The actual 
