312 MONTHLY WEATHER REVIEW. J m , 1902
publication of meteorological observations in daily, monthly,
and annual reports; the forms for the record of the observa-
tions; the symbols for meteorological phenomena; the self-
registers and their reduction to standard, and the commence-
ment of the meteorological year on the first of December. It
WBB further recommended that observations be conducted on
earthquake phenomena, that the atmosphere be explored with
balloons, that the ozone of the air and the formation of clouds
be studied. Provision was made for the second congress in
the following year. This congress met in the same place on
December 17,18,19, and 20,1901. The prospectus has already
been published in the MONTHLY WEATHER REVIEW, page 519,
November' 1901, and a resume of the proceedings will be
found on phge 132 of the REVIEW for March, 1902.
All meteorologists will be gratified to see these evidences of
activity in Mexico, and especially will they appreciate the fact
that the movement to establish a National Mexican Service is
going forward along the most approved lines. , It js evident that
the leaders are planning to conform to the resolutions of the
International Meteorological Congress generally, and also to
keep in touch with the practical system of the United States
Weather Bureeu, as far as possible. It is extremely important
that the Mexican Plateau should be placed under a strictly
scientific regime as promptly as can be clone, and that a com-
mon network of stations and telegraphic exchanges be insti-
tuted between the United States and Mexico, such as has long
existed between the United States and Canada.
NOTEl ON THE OSCILLATION PERIOD OF LAKE ERIE.
By R. A. HARRIS, U. R. Coast and Geodetlc Survey, dated Juue 2 i , 1902.
In a paper recently issued by the Weather Bureau entitled
Wind Velocity and Fluctuations of Water Level on Lake
Erie, the author, Prof. Alfred J. Henry, finds the theoreticnl
period of oscillation for the lake to be about eighteen hours
he notes that observations made at Buffalo and Amherstburg
indicate a period of fourteen hours, or a little more. In deter-
mining this 18-hour period, the lake is assumed to be isochronnl
with a rectangular body of water 60 feet deep and 246 statute
miles long. The object of the present note is to point out how
the observed period may be made to harmonize with a plausible
theoretical period.
I n any statement of this question which regards the depth
of the lake as uniform, one can hardly assume the average depth
to be so small as 50 feet; probably 60 or 65 feet is a good
value.
It would be a difficult matter to ascertain mathematicallJ
the free period of a body so irregular in outline and PO vari.
able in depth as Lake Erie. Nevertheless, the following approsi.
mation appears to be useful. Consider a square area oecillat.
ing in the manner shown in the accompanying fig. 1. W e
can imagine thin partitions to be erected along the liner
of motion and the oscillation will go on as before. That is
any one of the pointed areas will have a free period of oscillatiol
the same as that of the square. They are isochronal with e
rectangle whose length is equal to a side of the square
although their common least length is the square's diagonal
or 4 9 times the length of a side. I f , therefore, Lake Erie b4
represented by a leaf-like figure composed of several of thf
pointed areas, Maumee Bay marking one end and Buffalc
the other, the free period of such a body would be only 1/49
or 0.7071 times the period of a rectangle whose length is thit
lxtreme length of the lake. (See U. S. Coast and Geodetic
lurvey Report, 1900, pp. 686-689.) We can readily suppose
hat as a matter of fact the lake lies between the two hypo-
hetical bodies. With a length of 350 miles and a depth of 60
eet, the mean of the theoretical period for the leaf-like figure
,nd that for a rectangle is fourteen and one-quarter hours,
vhich is about the obRerved period of the lake.
West of Sandiusky the average depth of the lake is about
10.feet. This is partially separated from the eastern or main
)ortion by several islands and shoals. If this partial boundary
vere made sufficiently complete it would constitute the west-
:rn boundary of the oscillating body, and from this region a
lerived wave would progress to Amherstburg, the time of
,ransmission being about 1.7 hours. As the highs or lows at
hiherstburg are on an average, but little later than the lows
)r highs at Buffalo, it is probable that the oscillation extends
,he whole length of the lake, although its period may be
ilightly influenced by the partial barrier that actually exists,
tnd l)y the shallowness of the western end. The great depths
'ound between Dunkirk anc1I;ong Point must also have some
blight effect upon the free period of the body.
FIG. 1.
Questions connected with the oscillntions of lakes can be
studied experimentally by means of looclels suitably con-
3tructed. For, by ignoring frict.ion the method of dynamical
3imilaritmg can be applied. In practise the vertical scale of the
model inust generally be greater than the horizontal in order
to obtain depths suficiently great for the purpose. The only
restrictions are that the inazimum depth in the model shall be
but a small fraction of the length, and that wherever the mo-
tion is considerable, the slopes of the bottom along the lines
D f motion mast be small. If '11 denote the ratio of any hori-
zontal distance in the model to the actual distance, and if m
denote the ratio for heights (so that ,i l i /j j is the ratio of the verti-
cal to the horizontal scale of the model), then the period ratio
w i l l be n/,/Z.