Fuly 23, 1885 | 
polarisation, and examination between crossed Nicols failed to 
elicit any appearance indicating strains in the glass. The 
spectra seemed to be identical with the colours of thin plates, 
with which they agree in being much brighter in reflected than 
in transmitted light. The colours are also complementary in 
the two cases, as can easily be shown by the use of a micro- 
meter eyepiece; the colour of the band lying on a certain 
division being noted, the moving of a suitably arranged screen 
cuts off the light from the mirror under the stage and allows that 
from a condensing lens above the stage to fall on the glass; the 
Fic. 3. 
colour is then seen to change into its complementary, passing 
through white if the two lights are of suitable relative brightness. 
Ifthe surface layers of the glass near the scratch are separated some- 
what from those below, the interval forms a thin plate, and this 
would suffice to account for the spectra, as the separation would 
necessarily reach a maximum at the scratch, and we there find 
the spectra nearest together ; moreover, along the edge of the 
outermost spectrum, the black of the first order is quite distinct 
wherever the spectra are broad enough to allow their colours to 
be distinguished. The splitting perpendicularly to the cleavage 
plane would then be quite analogous to that seen in other brittle 
- DB = { 
Se 
Fic. 4.—a = the scratch; =the cleft under the torn-up surface layer; cc 
= the splits. 
substances, and as every one must have repeatedly observed it in 
cutting such substances as camphor, paraffin, or ice (Fig. 4 
shows the condition of the glass on this supposition). That a 
tendency really exists in scratched glass for the surface layers to 
tear up we know from the fact that splinters of glass near the 
scratch wil] often keep peeling off for days together. 
Bo Fy. Love, 
Demonstrator of Physics in the Mason College 
Birmingham, July 14 
Prof. Sylvester's Articie on ‘‘A New Example of the 
Use of the Infinite and Imaginary in the Service of 
the Finite and Real” 
I SHOULD like to be allowed to mention that Mr. Buchheim 
has drawn my attention to the fact that one of the theorems 
in the postscript to my recent article in NATURE contained in 
the formula — 
PEO 
NLP SOO 
is virtually given under aslightly different notation in the second 
edition of Grassmann’s ‘‘ Ausdehnungslehre,” p. 141. 
July 12 J. J. SYLVESTER 
OG 
NATURE 
271 
Rainfall of N.W. England 
THE rainfall on the coast-line of the Dee and the sea from 
Chester to Llandudno in the spring and summer months presents 
some interesting features. As a rule the amount of rain is the 
greatest at the Chester end in the spring months (April to June), 
and at the Llandudno end in the summer months (July to Sep- 
tember). This is shown by the following table :— 
1882 1883 1884 
= Chester eee Chester caer Chester 
; In. In. In. In. In. In. 
April. 3*on 29 Se59 0°68 0o'79 132 ... of92 
May ..2 1°54) seen 1°86 1°09 1°37... 0°94 
JRE co SNH 2°15 2°04 Ielh :.. 2°06 
7°94 ... 9°70 4°69 ... 4°82 3°84 ... 3°92 
July =| Se208e2 '3s52 TeEG (4 .2/.1'86 4°43 --. 4/08 
AMG: one 3hA7) one 2c52 ZNO 7.1 2764: 1235... 1x05 
Sept. ... 2°45 ... 1°80 4°34 5/07 1°67 ... 1°95 
OFZ ee eOr 7°66 ... 8°57 745 ... 7°08 
That this result is not a mere coincidence will be gathered from 
the following statement of the rainfall last month (June, 1885) 
at Chester, Colwyn Bay, and Bagillt, this last being between 
the other two places, at a distance of fifteen miles from the former 
or twenty-five miles from the latter :— 
Chester Bagillt Colwyn Bay 
June In. n. In. 
5 ed o'l2 0'07 os 
6 ee 38 me ‘21 ; *20 
7 6 "33 55; "29 gi "24 
8 a ‘II 285 05 ao ‘Ol 
16 oF 18 cee 15 oa 03 
18 a i fo) a 06 Bee 05 
19 a 04 om ‘02 re ‘02 
20 Be "16 ee ‘09 a 03 
23 Si "79 ne "46 oy 32 
24 ES 13 4 ‘08 are "02 
29 Pe — *o2 — 
2°34 1°50 bp Co) 
It will be seen that on every day, except one, the rainfall was 
highest at Chester, Bagillt coming next, and Colwin Bay being 
the lowest. There must therefore be some definite law which 
governs this gradual decrease from east to west; and in the 
hope of drawing out the opinions of more skillful meteorologists 
on the subject, I venture to suggest the following explanation. 
It is generally admitted that atmospheric currents travelling 
across the Atlantic from the south-west reach our islands charged 
with aqueous vapour almost to saturation. These, meeting the 
mountains of Wales and Cumberland, are driven up into a 
higher and colder region where the moisture is condensed into 
rain, causing a very heavy rainfall on the western slopes. Now 
if a line be drawn south-west from Colwin Bay, it will be found 
to pass over some of the highest mountains in Wales, so that 
in spring, when the air above these mountains is still cold, the 
moisture is so effectually condensed that there is little left to 
fall on the north-east side. But if a parallel line be drawn 
south-west from Chester it passes over no very high region, and 
the moisture is therefore less completely drained from the atmo- 
sphere. But, as the summer goes on, the mountain-tops become 
warmer and the condensation on them less complete, and then 
more rain is left to fall on the north-east side, ze, at Colwyn 
Bay. And as the seer day temperature is higher at Chester 
than Colwyn Bay, less rain is condensed at the former place. 
Chester, July 18 ALFRED O. WALKER 
““Foul Water ” 
My attention has just been called to Mr. Shrubsole’s letter 
under above title. Having been for several weeks past engaged 
in dredging off the North Wales coast, I have continuously 
noticed the profuse amount of gelatinous bodies diffused through- 
out the sea, evidently of the same character as observed by Mr. 
Shrubsole off Sheerness. They appeared here early in June. 
The little bodies are distinctly visible on holding a bottle of 
sea-water up to the light. They vary in size from 1/16th to 
* Colwyn Bay is about five miles east of Llandudno. 
