359 
Tue general view held by philosophers seems to be 
that when temperature rises the gravitative attraction 
changes (if at all) by a function of temperature only. This 
leads to the results stated in the beginning of their letter 
by Dr. Lindemann and Mr. Burton. But itis commonly 
conceded that at present there is no trustworthy theory 
of gravitation, so that one seems entitled to suggest 
that any increment in the force may be a function of 
both mass and temperature. The simplest formula, 
that of the weighted mean temperature, brings the 
facts as at present known into line. It seems 
that we must await further data; for, of course, a 
fact, however slight, may shatter a theory lying in 
its course. 
If M, m, » are the masses of earth, kilogram, and 
milligram, we have the cases cited thus. When m 
alone rises in temperature the increment of its force 
on p» is GmpaT/d*, and on M is GmmaT/d?. The 
first is sensible, the second insensible, compared with 
the forces when cold. Again, when , alone rises in 
temperature the increment of its force on m is 
GppaT/d?, but the increment of the force of M (con- 
sidered as an aggregate of p particles) on m is 
GppaT/d*?, multiplied by 
M M 
ee : p+ _GMmaT/d?. 
wh M+m im 
This is identical with the increment of force on m 
due to M (considered as a whole), and is sensible com- 
pared with the force when cold. The series forms a 
consistent whole. 
If the increments are due to radiation and reson- 
ance (see Nature for July 13 last), there is 
an acoustical parallel. A medium forle will set 
up resonance to a slight degree in a large fork, to 
a greater degree in a small fork, all the forks having 
one frequency. 
I cannot understand how the idea arises that action 
and reaction would ever, on this theory, be unequal. 
Let the force of m on M, both cold, be F. When m 
only rises in temperature the force is (F+f,), and when 
M only rises it is (F+f.), f2 being of a higher order 
than f,. But in each of. the three cases the attraction 
is mutual and equal between M and m. 
As regards Cornu’s results, I stated that we could 
‘deduce nothing’’ from them, so there is a mis- 
apprehension. I made no attempt to reconcile his 
results with mine. 
I do not consider that the indirect results obtained 
from Prof. Boys’s research can be laid by the side of 
mine for comparison. In his experiment any rise of 
temperature would involve both masses. Supposing 
the increment is due to molecular or other agitation 
of m due to radiation in my case, in his case there 
would be an enormously greater agitation in m due to 
direct rise in temperature. I see no reason why this 
should not be one hundred times as great as the effect 
due to radiation. The weighted mean formula was 
used to account for the effects observed when either 
mass is heated. The case when both masses are 
heated is different, and has not been at present con- 
sidered, the data at hand being indirect and inconclu- 
sive, 
The probable result from my experiments is 
a@=+(1-2+0-05)x 10-°, or, as amended recently at 
Newcastle, a=+(1-3+0-05)x 10-5. Thus, while indi- 
vidual results differ among themselves, in some cases, 
by as much as the whole effect, yet the collective result 
is not weak. 
I wish to express my appreciation of the generous 
terms used regarding my experiments by the authors. 
, f ; P. E. SHaw. 
University College, Nottingham, December 13. 
NO. 2462, VOL. 98] 
NATURE 
[JANUARY 4, 1917 — 
THE USE OF METEORITIC IRON BY ~ 
PRIMITIVE MAN. 
Mae as they reach our planet, are 
of three different kinds, namely, siderites, 
composed chiefly of nickel-iron; aerolites, com- 
posed chiefly of stony matter; and siderolites, 
composed of a mixture of iron and stone. The 
first and third only are of interest to us in this. 
inquiry. 
Great diversity of opinion has always prevailed 
among archzologists concerning the source of iron 
used in antiquity before it was intelligently pro- 
duced from the ore. On the whole these opinions 
are about equally divided, though perhaps the 
majority are inclined to the verdict that meteoritic 
iron was not used by primitive man, for the fol- 
lowing reasons :—First, because nearly all irom 
implements of antiquity extant, at least from the 
Old World (including the piece found in the Great 
Pyramid of Cheops), are of terrestrial origin; 
secondly, because it was believed to be non- 
malleable; thirdly, because it was said to be too 
scarce; and fourthly, because it was argued that 
small fragments could not have been detached 
from meteoritic masses by the means available to 
primitive man. 
There is, to the casual observer, a great deal 
of truth in these four arguments, and they prove 
that there were several sporadic sources from 
which the early iron was drawn; perhaps in one 
locality they were chiefly terrestrial, while the use 
of extra-terrestrial or meteoritic iron predominated 
in other localities. 
The fact that iron, in the earliest ages and from 
whatever source, was extremely rare is beyond 
all doubt. One might say that it was at least as 
rare as, and perhaps more valuable than, gold. It 
might therefore be asked: If meteoric iron was 
so rare and valuable, why was it not more care- 
fully preserved by the people of antiquity? For, 
whereas we have plenty of gold objects of ancient 
workmanship in our museums, those of meteoritic 
iron are conspicuous by their absence. To this 
we may safely reply that the value of the meteoritic 
iron lay in its actual employment in the form of 
tools, implements, and weapons, and to have 
hidden such objects in treasure stores would have 
nullified their value, and that is one reason at 
least why so few were preserved. This disposes of 
argument number one. | 
We must bear in mind also that in the New 
World this argument does not apply with the same 
force, as we shall see that objects of meteoritic 
iron are not so rare there as in the Old World. 
The reason for this is that, whereas iron was pro- 
duced in the Old World from the ore more than 
three thousand years ago, it was not introduced 
into the New World until the discovery of that 
continent about four hundred years ago, and it 
is obvious that objects more than three thousand 
years old have become rarer in our day than those 
comparatively new ones of only four hundred’ 
years ago. 
A paper on this subject was contributed by the: 
present writer at the autumn meeting of the Irom 
