412 
NALURE 
[ AucusT 31, 1899 
{ 
india-rubber about 3-inch thick, firmly attached by a slot and 
screwed bar to each roller, completes the ar rangement. 
The rollers being wound through about one entire revolution, 
and the india-rubber being thus stretched tight, layers of cloth, 
clay, paste or other giving material, are laid upon it. The 
handle is then turned in the reverse direction, and the india- | 
rubber gradually released. Folds are in this way shown slowly 
growing—the broad elastic band simulating the contraction of a 
and 3, cloths are seen 
portion of the earth’s crust. In Figs. 2 
Fic. 2. Fic 3. 
folded thus—first, without superincumbent weight, and second, 
with a weight of 30 lbs. 
That the larger folds are those generated at the surface, and 
the smaller and more numerous those produced under pressure 
(z.e. at great depths), is here made evident. | 
By substituting blocks of stone or wood for ordinary weights 
above the cloths (Fig. 4) and repeating the experiment, some of 
the relations between folding and faulting are clearly shown. 
| the best score. 
Fic. 4. 
If clay be used instead of cloths, all the results of Favre’s 
well-known experiments (Arch. @. Sctences Phys. et Nat., 
1878, and also NATURE), and many of those described by 
Cadell, Bailey Willis and others, can be obtained, and with the 
exercise of a little ingenuity it is easy to vary the experiments so | 
as to reproduce a large number of the fold-forms known, and to 
illustrate their consequences—thrusts, faults, &c. 
This machine was made for me in 1880 by the late Mr. C. D. 
Austen, of Newcastle-upon-Tyne, from my designs. 
G. A. LEBOUR. 
The Durham College of Science, Newcastle- 
upon-Tyne, August 18. 
Scoring at Rifle Matches. 
IN his letter to NATURE of August 17, Mr. Mallock appears 
to assume that there is such a thing as abstract ‘‘ accuracy ”’ in 
estimating the value of a marksman’s score. The method in 
use at Bisley is, as I understand him, to be regarded as a 
rough approximation to the accurate method, whether the best 
available approximation or not. Is it not rather the case that 
the standard of accuracy is itself arbitrary, and what the 
authorities at Bisley have established is not an approximation | 
to an ideal standard, but is to be regarded as a real standard of 
cellence ? | 
In result Mr. Mallock’s ‘‘ accurate” method is this: in | 
his notation any two scores for which R* +p? is the same are of 
NO. 1557, VOL. 60] 
e 
equal merit, or that one for which R*+ p? has the least value is 
Now, if ‘‘a” be the distance of any shot mark 
from the bull’s-eye, # the number of shots, R?+p?=3a2/z, 
Mr. Mallock’s standard, then, is that the best score is that for 
which the sum of the squares of the distances from the bull’s- 
eye is minimum. I see no reason why this method should be 
regarded as accurate par excellence, except the analogy of the 
method of least squares. But the analogy is misleading. 
Where the method of least squares is applicable, the object 
is to find the most advantageous value of an unknown quantity 
to be deduced froma number of observations. An accurate 
value of the quantity does exist. And of two or more results 
deduced from the observations, that which is nearer to the 
accurate value is always better than one more remote, however 
near to the truth either may be. 
In rifle shooting, on the other hand, there is generally some 
finite *space—e.g. the port-hole of an enemy’s ironclad, such 
that all shots which pass through it are of practically equal 
value, and all shots which do not pass through it are of little or 
no value. 
This is much more accurately represented by the Bisley 
method than by the method which Mr. Mallock would sub- 
stitute for it. S. H. Burspury. 
THE only remark I should wish to make on Mr. Burbury’s 
letter is that every shot on the target is truly the record of an 
observation, and that there is every reason to treat these records 
as far as is practicable by the methods which apply in obtaining 
the best means of a number of observations. Of course, it is 
only in the case of ‘‘centre of target” competitions the 
“R24? a minimum ” test applies. Prizes might well be given 
for close grouping, with a penalty depending on the mean 
distance of the group from the centre of the target. 
August 22. A, MALLOCK, 
Spectrum Series. 
Stk NorMAN Lockyer’s lectures on ‘‘Spectrum Series”’ 
seem to show very clearly the important fact that there is a 
close connection between the valency of an element and the 
lines in its spectrum. 
The connection indicated is as follows :— 
Nonvalent elements yield spectra with single lines. 
Monovalents yield doubles, 
Divalents yield triplets. 
On turning to the list given in NATURE (vol. lx. p. 370), it 
| will be seen that helium, by yielding doubles as well as singles, 
and cobalt, by yielding doubles only, are practically the only 
discordant cases in Sir Norman Lockyer’s list, since aluminium 
| and indium are trivalents, and their anomalous behaviour in 
yielding’ doubles only can perhaps be explained. 
August 26, W. SEDGWICK, 
Magnetic ‘‘ Lines of Force.” 
IN some text-books and by some lecturers (e.g. Prof. A. 
Gray, as reported in Nature of August 17, p. 379), the 
lines of magnetic force are said to be the curves along which 
iron filings are marshalled when sifted over a piece of card laid 
over a horizontally placed magnet. 
Surely this is hardly correct. The true lines of magnetic 
force must be represented, like those of all other radiant forces, 
by radiating straight lines drawn through the points of action of 
the resultants of all the forces residing in the individual mole- 
cules of a given magnet (such points, though varying in position 
with the position of a magnetic body in the field, being often 
referred to as fixed ‘ poles”). 
The symmetrical figures traced out by iron filings merely show, 
of course, the directions in which a line joining the poles of a 
very short magnet will lie in different parts of a magnetic field, 
under the influence of the true lines of force. IBS IR ae 
August 29. 
Critical Pressure.—A Suggested New Definition. 
THE critical pressure of a substance is commonly defined as 
““the least pressure that will suffice to reduce that substance from 
the gaseous to the liquid state when at its critical temperature.” 
But this definition contemplates the matter solely from the stand 
point of what occurs at the critical temperature, and I think it 
