488 
NATURE 
| Marcit 21, 1907 
I should accordingly like to ask Mr. Galton whether 
he would indicate what, in his opinion, are the chief con- 
siderations to be taken into account in giving preference 
to the mean or the median as the better measure of the 
“ average’? It is a point upon which there is consider- 
able difference of opinion; the recognition of the median 
is rapidly extending, and some statisticians incline to think 
that there is a growing tendency to quote it in cases where 
the ordinary arithmetic mean is preferable. 
March 16. R. 
H. Hooker. 
Rothesay Rainfall and the Sun-spot Cycle. 
Tue rainfall of Scotland has been thought to show, in 
its variations, an influence of the sun-spot cycle of cleven 
years (in the sense of most rain about maxima). Evidence 
of this was furnished not long ago by one of our ablest 
meteorologists, Dr. Buchan, in a paper to the Scottish 
Meteorological Society (Journal, 3 ser., Nos. XVili.—XiX., 
p- 117)- 
For such an inquiry the record of Rothesay, in Bute, 
is singularly valuable, extending back as it does to the 
year 1800 in unbroken series. 
The relation to the sun-spot cycle may be traced, I 
think, not only in the total annual rainfall of Rothesay, 
but also, with more or less distinctness, in the amounts for 
certain sections of the year, and even individual months. 
THE WEATHER REPORTS OF THE METEOR- 
OLOGICAL OFFICE. 
HE commencement of the new year was marked 
by the introduction of a number of changes 
in the weather reports of the Meteorological Office. 
Two notable events have contributed to bring about 
modifications in the daily report. Arrangements 
have been made for regular telegraphic reports from 
Iceland, and for occasional reports by wireless tele- 
graphy from the ships of the Navy. — 
Thanks to the Danish Government and the Great 
Northern Telegraph Company, the cable to Faeroe 
and Iceland, long desired by meteorologists and 
fishery associations, was laid in the summer of last 
year. There is a touch of sadness in the reflection 
that Adam Paulsen, director of the Danish Institute, 
who led the way so assiduously towards this meteor- 
ological Canaan, only got a distant view of the 
promised land. In August, 1906, he issued a 
circular on behalf of the Danish Government, de- 
fining the terms of subscription for the service of 
, Phe es EGE ia a pelolaiveres 
Gop Or, 08 12/6 "Do 3.728 32-36 “40744748 SBS Co 6469 72 76 Bo 8 BS F246 00 
= ae a SS el Tiga —- 
[etlieglonieclsaeelne ig emp 
meteorological telegrams from Iceland, but, as 
already reported in Navurr, he died before the 
arrangements were completed. 
Reports from  Thorshavn in 
nog Faeroe and from Seydisfjord, on the 
Taal east coast of Iceland, have been re- 
15- aa | ceived in London, in a_ provisional 
23 tis ry | way, since the end of October, but 
“Ay 06 fi saa|NS | \ the meteoroiogical telegrams from 
fs ise | { J ait Reykjavik, on the west coast, com- 
ala Sy aay Val ry menced on Friday, Webruary 15, as 
HA INN neh eae part of the full system which in- 
in i Wey) se cludes messages from Blénduos and 
fo 13}— eral pe i a Akureyri, where the land line 
Peale = \ fh touches the northern fjords, and 
Peas I /\ Grimstadir, between the last-named 
ae Sencha iN place and Seydisfjord, where the 
# E , aN el iE cies cable lands. The meteorological 
ales : pan oy arrangements are not complete even 
\ / / \4 \ yet, for the reports do not conform 
x) to the established international 
te] 
a, Rainfall July, Rothesay, 1300-1906 ; smoothed with sums of five ; B, sun-spot curve. 
Thus it is met with in the rainfall of and 
especially that of July. 
I have prepared a curve of the July rainfall (a), in 
which, by a familiar method, each year-point represents 
the rainfall of five contiguous Julys (i.e. 1800-4, 1801-5, 
and so on). Below is the sun-spot curve (8). The amount 
of correspondence between these two seems remarkable, 
and not easily explained by fortuitous coincidence. 
Avex. B. MacDowa tt. 
summer, 
The Relationship between Diamonds and Garnets. 
In an able paper entitled *‘ The Diamond Pipes and 
Fissures of South Africa,’’ read before the Geological 
Society of South Africa rather more than a year ago, Mr. 
H. S. Harger refers more than once to the significance 
of the fact that diamonds have been found embedded in 
garnets. Perhaps the fact that the converse is also true, 
namely, that the garnet sometimes occurs embedded in the | 
diamond, may not be without its share of interest. I have 
here at the present time a fragment of a Wesselton 
diamond, weighing a little more than a carat, containing 
a small, irregular garnet of about one-tenth of a carat. 
Originally the fragment seems to have formed a part of 
a shapeless diamond of perhaps two carats, which evidently 
enclosed either two or three small garnets, or garnets and 
diamonds. J. R. Surton. 
iXimberley, South Africa, February 20. 
NO. 1951, VOL. 75| 
model, either in uniformity of the 
hour of observation or the extent 
of the information transmitted; but 
those who have seen what the new 
information means for the weather 
map of north-western Europe, what light it throws 
upon the meteorological situation of the northern 
Atlantic, wiil appreciate the satisfaction that is felt 
with the result of the negotiations even in their pre- 
sent stage. Paulsen has indeed carved for himself 
a memorial aere perennius upon the winds and 
weather of the stormy northern island. 
It is to the Lords of H.M.. Treasury that we owe 
the realisation of this long-cherished project so far as 
this country is concerned. It need hardly be said 
that the cost of the mew service is very considerable. 
Their lordships have undertaken to ask Parliament 
to increase the grant for meteorology from 15,300l., 
the figure at which it has stood since 1882-3, by 
200l., and the greater part of our share of the ex- 
penses for Iceland telegrams is thus provided for. 
In order that the new information may be incor- 
porated in the daily weather report the area of the 
charts has been extended to a more western longi- 
tude than hitherto, and the occasion has been utilised 
also to take in an area as far south as Gibraltar, and 
to meet a wish, often expressed, that a barometric 
chart of the 6 p.m. observations of the previous even- 
ine should be given. This appears as an inset chart 
on the same scale as ‘“ yesterday’s ’? 8 a.m. chart for 
the whole of Europe, side by side with the 8 a.m. 
chart for ‘to-day.’ But six o'clock observations 
