/ 
. 
FEBRUARY 5, 1914] 
NATURE 
631 
tiania; Comptes rendus, vol. clvi., pp. 450, 536, 1913). 
In particular he finds certain remarkable periodic 
trajectories in the form of circles the plane of which 
is perpendicular to the axis of the magnet and the 
centre of which is at some point on that axis. If this 
point coincide with the centre of the magnet we 
obtain circular orbits in the equatorial plane of the 
magnet. Further, there are other trajectories which 
never get outside closed toroidal spaces in the case of 
stability, or which approach asymptotically the circle 
in question in the case of instability. It appears prob- 
able that similar results would be obtained in the case 
of a ring of electrons, and that the outstanding 
problem of the stability of such a rotating ring when 
only electrostatic forces are considered might in this 
way be overcome. Experimentally such stable rings 
have been obtained by Birkeland by employing a 
magnetised sphere inside a vacuum tube. 
Some of the orbits calculated by Stérmer are also 
suggestive in connection with the wide angle scatter- 
ing of a particles investigated by Rutherford and by 
Geiger and Marsden. If the nucleus produce a mag- 
netic field, Rutherford’s estimate of its radius may 
require modification. aioe ALLEN. 
Wheatstone Laboratory, King’s College, London. 
I HAVE read the letters of Dr. Bohr and Mr. Moseley 
_ with great interest, and would like to make a few 
remarks in reply which may serve to render the mean- 
ing of my first letter more clear. Dr. Bohr says that 
we have no right to consider nNe?, m, r, and h as 
independent variables and that we must eliminate r, 
in which case we find his formula. I am not con- 
vinced that this is necessary a priori, as Dr. Bohr 
would seem to consider it. In some cases it leads to 
conclusions which are obviously erroneous. Suppos- 
ing, for instance, that we calculate the period of a 
pendulum by this method. If we eliminate h we 
find t=const. ge but if we eliminate 1 we find 
roy 
Ay 
t=const. \/ Bs We have just as much or just as 
oO 
little reason, a priori, to eliminate h or r, or any of 
the quantities involved in one case as in the other. 
In the case of the pendulum, h can only appear as a 
I-73 
where E is the energy. Possibly the same is true in 
atomic models. 
I suggest that Mr. Moseley’s frequencies, which 
can be represented by various equations, do not prove 
that one must necessarily adopt the formula obtained 
by eliminating 7. But even if it be admitted that r 
must be eliminated a priori, the fact that we then 
always find a formula which, as Dr. Bohr admits, 
only differs from his in the constant, seems to me to 
justify my view that the fact that the frequencies 
agree with the formula does not necessarily confirm 
Dr. Bohr’s special assumptions. The support to be 
derived from an agreement in the matter of the con- 
stant, however, is not very strong, as, according to 
Dr. Bohr’s theory, it contains a factor of the form 
(1/7,2—1/7,*) which obviously gives us the choice of 
an infinite number of values between O and 
277(N —o,)?. 
Mr. Moseley also adduces arguments only in favour 
of what he calls the hk hypothesis, not of Dr. Bohr’s 
special assumptions. The reasons, however, do not 
appear to me absolutely convincing. Thus he says 
v~(Fr?)?, where F is the resultant electrostatic force 
on one electron, and concludes that as M?.L?T-! is 
constant, ML?T-? is constant. He thus introduces 
NO. 2310, VOL. 92] 
corrective term, perhaps of a form similar to / 
various hypotheses, such as that the same number of 
electrons oscillate in every atom, that there exist 
no other forces than electrostatic, and so on. If one 
liked, the fact that v~ N* might just as well be inter- 
preted as v~ Fr”, assuming N electrons to be attracted, 
whence we could deduce ML?.L/T=const., i.e. a 
universal velocity times a universal moment of inertia. 
Mr. Moseley says no independent natural unit of 
length is known. It is very easy to imagine atomic 
models in which one occurs, as, for instance, that 
proposed by Sir J. J. Thomson at the last meeting of 
the British Association. 
There are one or two other points which do not 
seem to confirm Mr. Moseley’s interpretation of the 
phenomena which he has observed. Mr. Moseley him- 
self found, I believe, several lines in the characteristic 
platinum radiation, which are not where they should 
be according to his hypothesis, i.e. about in the region 
of wave-lengths two octaves shorter than copper. 
M. de Broglie has shown by means of the ingenious 
method for photographing X-ray spectra described by 
him in the Comptes rendus de l’Académie des Sciences, 
November 17, 1913, and completed December 22, 1913, 
and January 19, 1914, that platinum antikathodes 
emit at least ten independent lines. Although the 
whole spectrum was photographed, including the 
shortest wave-lengths, and although a continuous 
spectrum was observed in the region in which the 
lines were to be expected, the lines themselves were 
not present. Unless we ascribe all the strong lines 
observed to impurities and introduce a special hypo- 
thesis to account for the fact that the expected 
platinum lines are not observable, this seems to con- 
stitute a grave difficulty for the theory of Mr. Moseley. 
I have misgivings further as to the ring of four elec- 
trons being able to emit such strong lines as those 
observed, as the radius of the ring is about one 
hundred times smaller than the wave-length, but no 
doubt Mr. Moseley has considered this obvious objec- 
tion, and satisfied himself that it is unfounded. 
To recapitulate. It seems to me that Dr. Bohr pos- 
tulates the h hypothesis, and that Mr. Moseley derives 
it by introducing a hypothetical model: That the h 
hypothesis does not entail Dr. Bohr’s model. That 
Dr. Bohr’s constant as applied by Mr. Moseley con- 
tains a factor which varies from o to 1, and that ? the 
value chosen is entirely arbitrary. Therefore my view 
is that all that can be said of Mr. Moseley’s observa- 
tions is, that they do not contradict Dr. Bohr’s 
assumptions, not that they confirm them. 
F. A. LInDEMANN. 
Paris, January 25. 
Systems of Rays on the Moon’s Surface. 
Ir is a strange fact that those who have little ex- 
perience of volcanoes notice a rough resemblance be- 
tween the irregularities of the lunar surface and 
terrestrial volcanic vents. However much one juggles. 
with diminished gravity and magnifies volcanic 
energy in the past history of our satellite, there are 
still several facts which are overlooked by many 
theorists. Mr. C. H. Plant points out in Nature of 
January 15 (p. 550) that the ‘‘volcanic action of the 
moon was of enormous character’’—this would need 
be so to produce on such a small globe craters of 
80 kilometres or more in diameter. 
Now all large craters are the result of explosive 
action, and, in explosive action, only fragmentary 
ejecta are thrown out by the amount of volatile con- 
stituents of the magma, which, if sufficient to ex- 
cavate a crater, are also sufficient to break up all 
the igneous magma into scoriaceous or pumiceous 
materials, and not allow it to issue continuously as a 
lava stream. When lava rises, subsequent to an ex- 
