NI 
[@) 
rs) 
NATURE 
[May 26, 1923 

Letters to the Editor. 
[The Editor does not hold himself responsible for 
opinions expressed by his correspondents. Neither 
can he undertake to return, nor to correspond with 
the writers of, rejected manuscripts intended for 
this or any other part of NATURE. No notice ts 
taken of anonymous communications.) 
Gravitation and Light-Pressure in Spiral Nebule. 
Pror. LINDEMANN’s idea that the spiral nebule 
may be clouds of particles small enough to be repelled 
by light is of considerable interest. But we must re- 
member that light carries with it another potential 
influence which it exerts when it encounters matter, 
namely, the power of ejecting an electron with an 
energy of the same order as that of the electron 
responsible for the light. Star-light, therefore, 
should be able to eject electrons with enormous 
energy ; and this kind of induced radio-activity may 
have several partly unforeseen results. A stellar 
variety of spectrum is one of them, if a continuous 
spectrum can be composed of a multitude of fine lines, 
with gaps only where the specific exciting radiation 
was absent. 
Unpolarised self-luminosity is surely more likely 
than mere reflection of incident light. The reddened 
_ light from the centre, observed by Mr. Reynolds, 
might well be a sunset effect, due to vision through a 
number of small blue-reflecting particles ; the pheno- 
menon does not harmonise so well with the idea of 
borrowed light. 
I suppose that Dr. Jeans’s spiral polar arms might 
occur in a Lindemann cloud as in any other enormous 
quasi-gaseous mass. 
The fact that some few of these nebule are approach- 
ing the galaxy, instead of rapidly receding, may be 
accounted for by the suggestion that in these few 
the particles have aggregated into larger groups (as 
they may under some kinds of electrification), so 
that gravitation once more predominates over light- 
pressure. 
The excessive transparency of space seems limited 
to our own extensive neighbourhood, for in remote 
regions opacity will set in sooner or later, and all 
stray radiation—however enfeebled by distance— 
will sooner or later be re-absorbed, with perhaps 
exciting and generative material consequences. The 
birth as well as the death of matter seems not hope- 
lessly beyond our scope. OLIVER J. LODGE. 

Breeding Experiments on the Inheritance of 
Acquired Characters. 
ALTHOUGH I agree with Dr. Kammerer in holding 
the opinion that somatic modifications do, sooner or 
later, affect the gametes or reproductive cells in such 
a way as to produce an inherited development of a 
corresponding change of structure, I regret that the 
evidence presented in his lecture printed in NATURE 
of May 12 is in some respects open to the objection 
that it is not in accordance with the present state of 
biological knowledge. Another objection, which may 
be partly due to the fact that the lecture is only a 
brief summary, is that the evidence does not include 
sufficient detail, or precise comparison with controls. 
For example, Dr. Kammerer states that ‘‘ Thanks 
to its enclosing membrane, the ovary of the Sala- 
mandra can be removed from the surrounding tissue 
as a whole,”’ which, according to the context, is not 
the case with the ovary of birds. I have never 
heard hitherto of the existence of an enclosed ovary 
in any amphibian. Unless I am altogether mis- 
NO. 2795, VOL. 111] 

taken, the germinal surface of the ovary is exposed to 
the coelom in Salamandra as in other Amphibia, and 
the ova escape through this surface as they do in 
birds, and not into an internal cavity of the ovary as, 
e.g., in teleostean fishes. If the latter were the case 
it would be very difficult to understand how ovarian 
transplantation could be carried out as in Kammerer’s 
experiments. 
The fact that Mendelian segregation occurs when 
naturally spotted Salamandra is crossed with 
naturally striped Salamandra, but not when it is 
crossed with experimentally striped specimens, 
suggests that in the former case the striped character 
is gametic, while in the latter it is not. But if the 
experimentally striped character is not gametic, what 
becomes of the heredity ? Kammerer says doubtless — 
both are inherited, but the long-established character 
obeys the Mendelian law, the new character does not. 
But T. H. Morgan has shown that new gametic 
characters in Drosophila obey Mendelian laws from 
their first appearance. Newness or oldness has 
nothing to do with Mendelism. A slight degree of 
heredity is possible, then the experimental striping 
causes only a slight change in the gametes. Then I 
presume the natural striping has been caused by 
exposure to yellow surroundings (soil) for thousands 
of generations and has become*completely gametic, 
or almost so. 
The difference in Mendelian behaviour then would 
be due to the fact that the natural striping is almost 
entirely gametic, the experimental striping almost 
entirely somatic. Such a result would agree with 
the result of my own experiments on the production 
of pigment on the lower sides of flat-fishes, supposing 
it to be true that spotted salamanders occur in 
Nature on dark ground, striped (yellow) specimens 
on light or yellow ground. 
On the other hand, in the experiments on the 
results of ovarian transplantation Kammerer puts 
forth the extraordinary conclusion that the soma of 
the naturally striped female has no influence on the 
ova derived from a spotted female, but the artificially 
striped soma makes the ova derived from a naturally 
spotted female behave as though they came from 
a striped female. Here we have a complete gametic 
change due to somatic influence, while according to 
the Mendelian experiments there was little or no evi- 
dence of gametic change. Such contradictory results 
may be true, but it would require a great deal of corro- 
boration to prove them. 
Kammerer states that the case of Ciona intestinalis 
affords an experimentum crucis. He certainly ex- 
hibited photographs of living Ciona in the parents 
of which the siphons had been several times am- 
putated. In these young specimens the siphons 
were ‘‘ monstrously long,’’ and had been so “ from 
birth.’’ Putting aside the fact that Ciona is not, I 
believe, viviparous, where were the controls? I have 
a very strong suspicion that all young Ciona when 
extended under favourable conditions (e.g. supply 
of oxygen and food) have ‘“ monstrously long ” 
siphons. The evidence required is a large number 
of exact measurements, under the same conditions, 
of the siphons in the young of parents which were 
subjected to amputation, and in those of uninjured 
parents. . T, CUNNINGHAM, 
East London College, Mile End, E., 
May 12. 

Vertical Change of Wind and Tropical Cyclones. 
THE first step towards forming an opinion about 
the physical processes which operate in the formation 
and maintenance of tropical cyclones is a clear 
