186 
NATURE 
| DECEMBER 20, 1906 
research during the past year amounted to 425]., while 
the corresponding grant in the United States of America 
(salaries and administration expenses being excluded in 
each case) was more than 150,000l. It is true that in both 
countries the local authorities also spend some money on 
agricultural experiments, but the same disproportion would 
probably be found between the respective amounts if the 
figures could be arrived at. 
Are we to take it, then, that these figures represent the 
relative importance of the agriculture of the two countries, 
or does the larger figure indicate the greater need of the 
American farmer for experiment and investigation? The 
exact contrary is the case; in the British Isles we have 
to farm with dear land, dear labour, and a number of 
charges due to the proximity of a high civilisation. Farm- 
ing in consequence can only pay when there is a consider- 
able monetary return per acre, and the bigger yield neces- 
sary involves intensive cultivation, the purchase of 
fertilisers, and the employment of skill, which are all need- 
less to our competitors on a virgin soil. But each increase 
in the expenditure and skill necessary for the crop means 
a greater opening for knowledge and investigation; science 
can do little to save money for the man who merely stirs 
the surface of a virgin prairie, scattering in the seed mean- 
while, and then leaves it to take its chance until harvest. 
Compare with such a farmer the highly technical routine 
of the hop-grower who spends sol. per acre before he 
harvests his crop, his repeated cultivations, his manurings, 
his sprayings for various ends; it is with this kind of 
crops that science can find profitable employment. 
Looking at the average yields of the various countries 
of the world, we find that Great Britain is the most 
intensively farmed country; it obtains the biggest crops 
per acre, it has to spend the most to obtain them. Further- 
more, the biggest crop the greater are the risks of disease 
and blight, the greater are the difficulties in securing high 
quality. Here, then, in Great Britain exists the greatest 
need for knowledge and investigation; we cannot even 
always beg knowledge from wiser countries, for many of 
our problems are special, and brought about by the very 
conditions of high farming which prevail here. England 
was the first country to start an experimental station, yet 
Rothamsted still remains the only institution solely devoted 
to agricultural research in the British Isles, if we except 
the farm of the Royal Agricultural Society at Woburn. 
The income of the Rothamsted station, derived solely from 
private benefaction, is about 2600]. a year; in the United 
States each of the fifty-three States possesses a station 
receiving 3000]. a year from the Federal Government, 
besides what the State itself may contribute, in addition 
to the great central department of agriculture to which 
reference has already been made. 
SOME NEW METHODS IN METEOROLOGY.2 
PROF. BIGELOW has here collected six studies. The 
first four deal with diurnal periods :—(i.) of tempera- 
ture; (ii.) of barometric pressure; (iii.) of vapour tension, 
electric potential, and coefficient of dissipation; (iv.) of 
terrestrial magnetism; (v.) treats of the variable action 
of the sun and its effects upon terrestrial weather con- 
ditions; whilst (vi.) is a general review of the status of 
cosmical meteorology. 
The immediate occasion, the author tells us, for these 
studies was the necessity of deciding upon the best lines 
of work for the new ‘‘ Mount Weather ’’ Observatory, 
at Bluemont, Va., which is intended to serve as a centre 
for research in connection with the U.S. Weather Bureau. 
This observatory is to have on its staff experts in various 
departments, and there is to be an advisory committee, of 
which Prof. Bigelow is described elsewhere as chairman. 
Several of Prof. Bigelow’s views as to the prosecution of 
the higher meteorology have much to recommend them, 
as, for example, the following :—‘‘ If cosmical meteorology 
is to be established then all rough and ready methods 
must be abandoned, and the work of computing and dis- 
cussing the data must be placed in the hands of physicists 
1 “Studies on the Diurnal Periods in the Lower Strata of the Atmo- 
svhere.” Reprints from the Monthly Weather Review, 1905. By Prof. 
Frank Hagar Bigelow. (Washington: Weather Bureau, 1905.) 
NO. 1938, VOL. 75 | 
| 
and astro-physicists who possess scientific instincts and 
training ”’ (p. 48); or again:—‘‘ We must waste nothing 
by using bad methods of work and unskilled men ”’ (p. 51). 
But Prof. Bigelow possesses, apparently, a duality in his 
nature, and the following are examples of his second self :— 
“In the midst of this concatenation of forces the terrestrial 
magnetic field stands out as the best unifier or integrator. 
It is the most sensitive and delicate pulse which we 
possess, having one throb in the solar mass, and the other 
in its synchronism with the earth’s meteorological 
elements ’’ (p. 48). This seems not unworthy of Colonel 
Starbottle addressing a jury, but what exactly does it 
mean? Here, again, is what we are told of the sun :— 
“Recent computations indicate that at the centre... 
there is a nucleus which... is nearly as solid as the 
interior of the earth, with a temperature of about 
10,000° C.; the average density . . . is 1-43 times that of 
water, and this is located at half the distance from the 
centre to the surface ’’ (p. 39). 
Feeling doubts of our capacity to follow with advantage 
Prof. Bigelow’s highest flights, we have devoted more 
attention to his studies on the diurnal variations. The 
view to be taken of these must depend on whether they 
are intended as examples of the methods to be followed 
by the Mount Weather Observatory, or whether they are 
simply illustrations of the ‘‘ rough-and-ready ’’ methods 
the abandonment of which the author elsewhere recom- 
mends. Study i. deduces from continuous temperature 
records at Blue Hill Observatory, and from observations 
made during or in connection with kite ascents there, 
the diurnal variation of temperature at a series of heights 
for every month of the year. The final results are 
embodied in Figs. 14 to 25, the diurnal variation being 
assumed negligible at the height of 3400 metres the whole 
year round. The original data are not given, and the 
methods of manipulating them are only indicated generally. 
Of the probable value of the results no estimate seems 
possible. Study ii. gives some general, but not very 
lucid, information about the diurnal variation of baro- 
metric pressure. Of the amplitude of the 24-hour term 
it says, not incorrectly, ‘‘ it is very different at neighbour- 
ing stations.’’ Yet Prof. Bigelow obtains Fourier co- 
efficients for a composite diurnal inequality based on data 
from Boston, New York, Washington, Buffalo, and Cleve- 
land. Again, we are told in the general remarks that the 
amplitude of the 24-hour term is from one-fourth to one- 
half that of the 12-hour term. But in the composite case 
treated by Prof. Bigelow the 24-hour term is larger than 
the 12-hour term in the summer months, and the arith- 
metic means from the twelve monthly values of the ampli- 
tudes seem closely alike for the two waves. x 
In the calculations, the diurnal variation is assumed to 
be completely accounted for by three waves of periods 24, 
12, and 8 hours. If [n] denote the departure at hour n 
from the mean for the day, then the contributions to [n]} 
from the 12- and 8-hour waves are respectively 
E(x] + [2+ 12}} and 4{[727]+[72+8)+[z+ 16}, 
and what remains after subtracting these two contributions 
from [n] is assumed to represent the contribution of the 
24-hour wave. This method cannot be recommended even 
for rough preliminary work, unless the 24-hour term is 
largely dominant and the Fourier series is known to con- 
verge very rapidly. In the present instance the amplitude 
of the 8-hour wave is, according to Prof. Bigelow’s figures, 
about half that of the 24-hour wave from November to 
February. In these months the observational data would 
certainly give an appreciable 6-hour term. The same 
method is then applied to the diurnal variation of tempera- 
ture (with sign reversed) as deduced in Study i. for heights 
of 195, 400, and 1000 metres at Blue Hill. The results for 
the 8-hour wave at 195 metres during the summer months 
at once arrest attention. In July, for instance, no hourly 
value assigned to this wave is positive. This seems to 
be due, not to misprints—though these are somewhat 
numerous in the tables—but to error in the figures for the 
diurnal inequality itself. If the twenty-four hourly differ- 
ences from the mean are summed algebraically, there is 
in most months a substantial remainder, showing that the 
mean value for the day has not been correctly taken. 
Limits of space allow only of brief reference to other 
