660 
NAL OIE 
[AucuST 29, 1912 
but are an indirect result of the germinal instability 
occasioned by crossing in the ancestry. 
It is to be hoped that further study of this new series 
of forms, with particular regard to the manner of 
origin of the mutant types, together with crossing 
experiments with the O. Lamarckiana series, will 
throw further light upon the nature of the mutation 
processes in Oenothera. R. R. Gates. 
14 Well Walk, Hampstead, N.W. 
William Herschel and his ‘‘ Desertion.’’ 
In the valuable discourse on Sir William Herschel 
delivered at the Royal Institution on April 26 by Sir 
George Darwin, the well-known story of the desertion 
of the young bandsman from the Hanoverian Guards 
has been alluded to (NaTuRE, August 15, p. 620). A 
week or two after the delivery of this discourse the 
“Scientific Papers of Sir William Herschel” were pub- 
lished by the Royal Society and the Royal Astronomical 
Society, and in the introduction to that work there is 
given a detailed account of how Herschel left the army, 
written by himself and corroborated by the still exist- 
ing official discharge, signed by the colonel of the 
Guards in 1762. As many readers of NATURE may not 
come across that work, it may be of use to give a 
summary of the facts here. 
After the battle of Hastenbeck (July, 1757) young 
Herschel (eighteen years of age) left the army and 
went home to Hanover, on the suggestion of his father. 
But on his arrival there he found that as a non- 
combatant he was liable to be pressed into the army 
at any moment. He therefore at once (or very soon) 
returned to his regiment, putting on his uniform again 
(not taking it off, as stated) when he had passed the 
sentries at Herrenhausen. He remained with the army 
till the following September, when he finally left it, as 
his father pointed out to him that there could be no 
objection to his doing so, since he had not taken the 
oath when he joined the band as a boy of fourteen. 
He then went straight to Hamburg without going 
home first, and proceeded to England, where he had 
spent five or six months in the previous year and 
where he wished to settle. In March, 1762, he ob- 
tained a formal discharge, which is now printed in 
my above-mentioned introductory memoir. The story, 
originally published by: Airy on the authority of the 
Duke of Sussex, that George III. in 1782 handed 
Herschel a formal ‘‘ pardon,” must therefore have been 
due to some misunderstanding or other. 
L. E. Dreyer. 
Armagh Observatory, August 23. 
The Disintegration of Metals at High Temperatures. 
DurinG experiments on the disintegration of metals, 
particularly those which are not supposed to combine 
directly with oxygen, such as certain metals of the 
platinum group, I have found the disintegration to be 
due to the direct formation of an oxide. The loss of 
weight of a hot platinum wire, for instance, is zero in 
nitrogen, in hydrogen, and in a vacuum. By means 
of an expansion apparatus, all metals tried are found 
to give nuclei when oxygen is present, but not when 
it is absent, either in other gases or in a vacuum. 
The occluded gases come off in a vacuum in molecular 
aggregations, but there is no evidence that they bring 
particles of the metal with them. The loss of weight 
cannot be due to volatilisation, as it diminishes with 
diminution of pressure of surrounding oxygen. 
By weighing experiments, the weight of oxygen 
absorbed and of platinum lost correspond to the forma- 
tion of a hitherto unknown oxide of. platinum. This 
oxide is. deposited upon the walls of the containing 
NO. 2235, VOL. 89] 
vessel as a black powder; on being heated it turns 
to the metal, producing a platinum mirror. Micro- 
scopic examination does not reveal any evidence of 
crystals, either in the black powder or in the mirror. 
If, however, a piece of glass having such a deposit, 
and having been heated in different places, is boiled in 
aqua regia, the parts where the metallic mirror has 
been formed by heating become clear very quickly, 
whilst the black powder, where it has not been heated, 
remains unaffected. J. H. T. Roperts. 
University of Liverpool, August 20. 
September Meteor-showers. 
Tue following meteor-showers become due during 
the month of September :— 
Epoch September 4, 19h. 30m. (G.M.T.), nineteenth 
order of magnitude. Principal maximum, September 
4, 6h. 10m. ; secondary maxima, September 3, 7h. 4om., 
and September 4, 18h. 
Epoch September 7, 3h. 30m., approximately first 
order of magnitude. Principal maxima, September 6, 
2h. 15m., and September 7, 21th. 5m.; secondary 
maxima, September 7, 12h. 30m., and September 8, 
20h. 4om. 
Epoch September”7, 2h., approximately first order 
of magnitude. Principal maximum, September 6, — 
6h. 30m.; secondary maxima, September 5, 11h. 20m., 
and September 6, 23h. 55m. 
Epoch September 9, 15h. 30m., approximately 
seventeenth order of magnitude. Principal maximum, 
September 7, 22h. 50m.; secondary maximum, Sep- 
tember 9, 13h. 
Epoch September 9, 14h., sixteenth order of magni- 
tude. Principal maximum, September 8, toh. 15m. ; 
secondary maximum, September 8, 2h. 35m. 
Epoch September 12, 2oh., thirty-fifth order of 
magnitude. Principal maxima, September 9, 18h. 25m., 
and September 11, t4h. 30m. 
Epoch September 8, 17h. 30m., approximately 
seventeenth order of magnitude. Principal maxima, 
September 9, 22h. 45m., and September 11, 18h. 45m. ; 
secondary maximum, September 9, 8h. 34m. 
Epoch September 16, 8h., sixteenth order of magni- 
tude. Principal maxima, September 13, 1oh. 30m., 
and September 15, 6h. 35m.; secondary maximum, 
September 15, 14h. 30m. ; 
Epoch September 14, 22h. 30m., eleventh order of — 
magnitude. Principal maximum, September 13, 
22h. 25m.; secondary maxima, September 12, 
12h. 50m., September 13, 18h. 30m., and September 14, — 
8h. 50m. & 
Epoch September 16, gh. 30m., twelfth order of 
magnitude. Principal maximum, September 14, — 
16h. 50m.; secondary maximum, September 16, 
4h. 55m. 
Epoch September 15, 4h. 30m., tenth order of 
magnitude. Principal maximum, September 14, 
20h. 45m.; secondary maximum, September 13, 
oh. 40m. 
Epoch September 14, 15h., eleventh order of magni- 
tude. Principal maxima, September 15, 18h. 25m., 
and September 17, 14h. 25m.; secondary maximum, 
September 17, 2h. 35m. 
Epoch September 19, 2th. 30m., approximately 
seventh order of magnitude. Principal maximum, 
September 21, 2h. 35m.; secondary maxima, Septem- 
ber 20, 8h. 45m., and September 21, 6h. 30m. 
Epoch September 21, 9h. 30m., third order of 
magnitude. Principal maximum, September 
22h. 35m.; secondary maxima, September 23, 2h. 30m., 
and September 24, 22h. 25m. 
Epoch September 25, 15h. 30m., fourteenth order of 
magnitude. Principal maximum, September 23, 
22, 
