ON SEISMOLOGICAL INVESTIGATIONS. 49 
and then the errors are distributed as in the last column. The mean 
is now +0°014m. or +0°8s. and the mean of the errors is + 0°31 m., 
very close to the mean of the errors 0°35 m. obtained above for the 
reflected wave. A’ considerable part of this mean error may be due 
to the errors of reading distances from the epicentre, and to the error 
of assumed position of the epicentre itself. 
Large Negative Readings.—The eight negative readings are prob- 
ably due to accidental air tremors just preceding the quake; these 
call for no special remark here except that they seem to be pretty 
clearly separated off from the normal readings; even making a generous 
allowance for accidental error in the latter. It will be seen that the 
numerically smallest (—2°3m.) is a full minute away from the out- 
side error (— 1'2m.) included among possible normal readings. The 
details may be given here in case the observatories care to examine 
the records :— 
Observatory pee Date es Ss wees 
m. m. Fa 
Lemberg B.O Jan. 11, 1913 — 33 — 88 
Lemberg B.O. Mar. 23, 1913 — 50 — 87 
Lemberg B.O. April 30, 1913 —14-2 — 79 
Lemberg B.O. June 22, 1913 — 42 —3-9 78 
Aachen . W. April 30, 1913 —248 — 80 
Paris. . —_ Mar. 23, 1913 — 34 —0-7 98 
Ksara — Mar. 23, 1913 — 23 — 90 
Batavia W. Mar. 23, 1913 - 33 — 47 
Coming now to S, two large positive errors have already been 
mentioned as associated with large positive errors in P, viz., +9°5 m. 
at OCzernowitz on 1913, March 23, and +7°4m. at Triest on 1913, 
January 11, as also one considerable negative error of —3°9m. at 
Lemberg on June 22, 1918. These are omitted from further notice. 
Two large negative errors are 
Observatory Machine Date P Ss A 
Tiflis . igs Ca G 1913, Jan. 11 +02 —81 ie 
Florence . . . A 1913, Jan. 11 —09 —7-7 gs° 
The former is due to some unknown mistake; the latter is probably 
a mistake of S for PR,. These are also omitted from further notice. 
Two positive errors of smaller amount as follows :— 
Observatory Date P Ss A 
IRiverviow «|... <« . & 1913, June 22 —0:3 +4:5 94° 
Heidelberg . . . . 1913, June 22 +01 +4-7 80° 
are omitted as quite anomalous. The remaining errors are grouped 
in Table II. 
1914. a E 
