146 
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 zs 
taken of anonymous communications. | 
The Spectrum of Neutral Helium. 
At the end of his letter to NatuRE of January 13, 
p- 46, Dr. Silberstein appends a note to the effect that 
he has been able to express the diffuse series HeD’ in 
the form 
N =109723 {1/27 + 1/10? — £/g2 — r/m?* 
with errors of o-7A for the second line, and of between 
o-1A and 0-35A for the next ten. May I be allowed 
to offer the following remarks :— 
(1) A formula determined on a definite hypothesis, 
as here, ought to reproduce the wave-lengths within 
observation errors or at least be able to account for 
deviations from them. According to the data given 
the deviations amount to between too and 200 times 
the possible errors (7oo for the second). The usual 
empirical formula reproduces all the lines within these 
limits, except the first, the O-C errors being 0-000 for 
m =2, 3, 4 and the largest for higher values of m being 
0-02. Thelimit isdefinitely within + 0-1 of 27175-68 1/A, 
in other words, N(1/2? + 1/10% — 1/92) must have this 
value. This, of course, is possible by an empiric 
choice of N, but it would probably upset even the 
pcueh agreement when this is used in the last term 
/m?. 
(2) That the diffuse singlet series HeD’, and indeed 
also the diffuse doublet HeD’, can be represented 
roughly in the form A - N/m?, is due to the fact that 
for this special series the denominator in the empirical 
formula, m +0-996369 +0:002917/m, is necessarily very 
close to a whole number, and its deviations therefrom 
produce comparatively small effects when m becomes 
large. A similar arrangement in the cases of S’, S” 
or P” would be found impossible. 
(3) But the most fatal objection is that N(1/22 
+1/10? — 1/9?) must also be the first term of the p’ 
sequence, which is at least numerically represented by 
Pp’ (m) =N/(m +1-014593 - 0:004392/m)?._ Here again 
the denominator is nearly an integer (though further 
from it than in d’(m)), and no doubt it could also be 
represented by N/m?, with greater deviations than in 
the case of d’, but the first term would then be N/2? 
and not N(1/2? +1/10? — 1/92). é 
It is perhaps a difference in temperament, but to 
me Dr. Silberstein’s note appears rather to weaken 
than to give a ‘‘ much stronger support’ to his 
proposed theory. However, I am not here discussing 
his hypotheses, one objection to which I raised in a } 
letter to NATURE on September 2 last (p. 309) which 
Dr. Silberstein has not dealt with. 
W. M. Hicks. 
January 15. 

Some Experiments on Rate of Growth in a Polar 
Region (Spitsbergen) and in England. 
In a recent paper (Journal Mar. Biol. Assoc., vol. 12, 
1920, p. 355) attention was directed by me to the lack 
of critical evidence bearing on the theories offered to 
explain (a) the abundance of life in polar regions, 
and (6) the occurrence of several generations of a 
species living side by side in polar waters. Murray 
and Loeb and others have suggested that an explana- 
tion ot these phenomena may be found in a greatly 
NO. 2779, VOL. III | 
NATURE 

[ FEBRUARY 3, 1923 
retarded rate of growth which, it is postulated, must 
occur in the low temperatures prevailing in these 
regions. The present writer urges (a) that we know 
nothing about the rate of growth of organisms in 
polar regions, and (b) that the kind of metabolism 
of animals in polar regions—and in deep-sea situations 
—is not necessarily the same aS that in temperate or 
tropical regions. A given organism may be regarded 
as a machine, but it is perhaps derogatory to the kind 
of machine one is dealing with to assume that other 
life-machines existing under totally different conditions 
are necessarily governed by identical applications of 
the same laws; for example, it does not necessarily 
follow that because the rate of metabolism in tropical 
or temperate animals falls off rapidly with decreas- 
ing temperatures approaching 0° C., that metabolism 
in polar animals is necessarily of the slow rate of 
temperate animals at polar sea-temperatures. No 
reason has yet been shown that adaptation of meta- 
bolism cannot occur; on the contrary, there is 
every reason to expect such adaptation. 
The following experiments on the rate of growth 
in marine organisms at Spitsbergen—designed to 
obtain information on these problems—have given, 
however, mainly a negative result, but as in one 
case a positive result—yielding a much greater rate 
of growth than has ever been suspected—has been 
obtained, it is worth while recording the result now. 
It is hoped to write a fuller account later, giving 
details of the apparatus used, in the Journal of the 
Marine Biological Association. 
In 1921 simple experiments on rate of growth 
were carried out in 7 fathoms of water close to Anser 
Island in Klass Billen Bay, Spitsbergen, by the 
biologists of the Oxford Spitsbergen Expedition, and 
mainly under the direction of Mr. Julian Huxley 
and Mr. A. M. Carr Saunders. The present writer 
had hoped to carry out the experiments under 
personal supervision, with the promised help of Dr. 
Hoel of the Norwegian Fishery Board, but circum- 
stances nullified these arrangements. 
Two pieces of apparatus were used—a galvanised 
iron-wire network cage of 3-inch mesh and 5 feet 
by 4 feet by 9 inches was tarred and moored to the 
bottom of the sea after putting a large number of 
dried oyster shells inside it; and a floating tarred — 
wooden raft with strings of shells attached was 
anchored in the sea near the cage. The apparatus 
was put in the sea on June 27, 1921; the raft and 
shells were inspected by Mr. Huxley on July 16, and 
—owing to the illness of Mr. Carr Saunders—finally — 
hauled by Mr. R. W. Segnit, geologist, and Capt. 
Johannsen on August 24, 1921. ‘ 
On July 16 Mr. Huxley found practically no growth 
on the raft nor on the shells on the raft, but the cage © 
was not hauled. On hauling the cage on August 24 
the sea-urchins shown in Fig. 1 were found inside 
the cage. The door of the cage, which only covered 
the central portion of one long face of the cage, 
was found to be closed and /aced as had been previously ~ 
arranged on putting the cage in the sea. The 
astonishing sight of the relatively large sea-urchins 
inside the cage attracted attention at once, and a 
fruitless examination of the cage was made for any 
means of access greater than the mesh of the cage. 
The conclusion was therefore drawn that the urchins 
must have entered the cage while small, 7.c. of a 
diameter upwards to about 1-6 cm., and grown to 
the size observed, i.e. upwards to about 2-9 cm. in 
diameter—excluding spines—within 58 days. 
This result was regarded as very important, and 
a confirmatory experiment tried again at the same 
spot in 1922, under the direction and by the kindness 
of Mr. J. Mathieson, of the Scottish Spitsbergen 
Syndicate scientific staff. When Capt. Johannsen 
