B04 

emigrants. But Mr. Oakenfull goes beyond his 
brief when he dwells on the superiority of Brazil 
to the British Colonies, particularly in relation 
to the price of land. It must be remembered that 
Brazil is as yet only partially developed, and the 
present economic situation may be gauged by the 
value of the national securities. While Canada 
3 per cents. are now quoted at 83, Brazil 4 per 
cents. of 1910 stand at 49: No doubt this is 
owing to temporary causes, but for the intending 
emigrant the contrast is significant. 
Improved Four-Figure Logarithm Table, Multi- 
plication and Division Made Easy. By G. C. 
MeLaren.. Pp. 27. (Cambridge: At the 
University Press, 1915.) Price 1s. 6d. net. 
TuesE tables aim at popularising logarithms. 
Like other four-figure tables they give the 
logarithms of numbers from 1000 to 9999. While 
the ordinary tables give these in one opening 
of two pages by means of difference columns, 
Mr. McLaren gives all the gooo entries independ- 
ently, and so of course without difference columns. 
His tables consequently occupy nine openings or 
eighteen pages. Rapid reference to the various 
openings is made possible by ‘“thumb-indexing,” 
and for speed there is probably nothing to choose 
between these tables and the customary ones. 
There is a gain in accuracy, both because 
the use of difference columns is not trustworthy in 
the last figure, and because of Mr. McLaren’s 
ingenious device of showing the last figure to the 
nearest third. Whether these tables seriously 
reduce the skill required for their use as com- 
pared with ‘the customary tables we have some 
doubt. We think their appeal will be chiefly to 
thosé calculators who require slightly greater 
accuracy than the customary tables allow and 
who at present use five-figure tables. D. M. 
Joseph Pennell’s Pictures in the Land of 
Temples. 4o illustrations. (London: William 
Heinemann, 1915.) Price 5s. net. 
THE sub-title to this attractive volume very well 
describes its contents. It runs: “ Reproductions 
of a series of lithographs mace by him in the 
land of temples, March—June, 1913, together with 
impressions and notes by the artist.” The illus- 
trations start at Taormina, proceed around 
Sicily—thence to Italy, and are continued in 
Greece. The book is dedicated to Mr. R. M. 
Dawkins, late director of the British School at 
Athens, who showed Mr. Pennell where he would 
find the temples. The artist says with becoming 
modesty that having seen the pictures Mr. Daw- 
kins expressed the opinion that they had 
“something of the character and romance of the 
country.” It is unnecessary here to praise the 
work of so distinguished an artist; it is enough 
to say that the pictures convey just the impression 
which the temples made upon Mr. Pennell: “the 
great feeling of the Greeks for site in placing 
their temples and shrines in the landscape—so 
that they not only became a part of it, but it 
leads up to them.” 
NO. 2380, VOL. 95 | 
NATURE 

[JUNE 10, 1915 


LEITERS TO THE EDITOR. 
[The Editor does not hold himself responsible for 
Opinions expressed by his correspondents. Neither 
can he undertake to return, or to correspond with 
the writers of, rejected manuscripts intended for 
this or any other part of NatuRE. No notice is 
taken of anonymous communications.] 
The Continuous Spectra of Gases. 
In spectroscopic literature there are many casual 
references to a continuous background in the vacuum 
tube spectra of various gases, such as oxygen, 
chlorine, etc. Usually these observations appear to 
have been confined to the visible region, and I can 
recall no comments on continuous spectra in the ultra- 
violet except in the case of hydrogen. Schniederjost 
(Zt. f. Wiss. Phot., 1904, p. 265) and Friederichs 
(Bonn Diss., 1905) observed such a spectrum at low 
pressures, which extended to a wave-length of about 
2100. The latter attempted to use the uncondensed 
discharge through a small capillary tube at about 
2 mm. pressure as a source for the photography of 
absorption spectra, but found that the results were 
unsatisfactory, even with exposures varying from 
twelve to twenty-four hours. 
In photographs of the hydrogen spectrum ;obtained 
with a large two-prism quartz spectrograph I have 
frequently observed this continuous spectrum.  Al- 
though the resolving power of this spectrograph in the 
extreme ultra-violet is greater than that of a five-inch 
grating in the first order, there is no evidence of reso- 
lution into lines or bands. The spectrum appears to 
be uniformly continuous, and it seems likely that its 
gradual fading out in approaching the wave-length 
2100 is due rather to the absorption of the thick quartz 
system than to the lack of these wave-lengths in the 
emitted light. It appears to be due to pure hydrogen, 
for successive improvements in purity due to the 
removal of oxygen, water vapour, and nitrogen cause 
no noticeable change; nor does the addition of a trace 
of oxygen to hydrogen previously freed from that gas 
so far as possible cause any appreciable difference. 
It seems very unlikely that a continuous spectrum 
can arise from free vibrations within the atom or 
molecule, hence it has been usually ascribed to mole- 
cular collisions. In comparing different gases at the 
same pressure, the number of collisions would depend 
mostly on the mean velocity of the molecules, so that 
the number of collisions would rapidly diminish as 
the molecular weight increases; hence we might 
expect that the continuous spectrum of a light gas 
would be stronger than that of a heavier gas. This 
was found to hold good for hydrogen, helium, and 
neon. Photographs were obtained of the spectra of 
these three gases in vacuum tubes prepared by Hilger. 
The pressure was about the same in all. With a two- 
minute exposure, the continuous spectrum of hydrogen 
was very intense, that of helium about half as strong, 
and that of neon about one-third as strong. They all 
extended to about the same limit—that set by the 
transparency of the quartz. In all these cases the 
uncondensed discharge of a medium-sized induction 
coil was used. The introduction of a condenser almost 
completely obliterated the continuous spectrum. When 
a condenser is used the radiation probably comes from 
dissociated ions, with free periods little disturbed by 
molecular collisions. 
Nitrogen, krypton, and xenon did not show any 
continuous spectrum. 
Some tests showed that hydrogen tubes may render 
excellent service as sources for the study of absorption 
spectra in the ultra-violet. It was not found advisable 
to use capillary tubes, or to work at such low 
pressures as Friederichs did. The best results were 
