324 
be useful steel. He failed entirely in this ; he never suc- 
ceeded in producing merchantable steel from ordinary 
English cast-iron by this method. 
The Bessemer process, as at present conducted, consists 
in first oxidising simultaneously all or nearly all the car- 
bon and silicon, and then adding to the decarburised 
iron a new dose of carbon, by means of a known quantity 
of spiegeleisen of known composition ; thus reverting to 
the old Sheffield principle of first bringing the cast-iron 
to the state of wrought or decarburised iron, and then 
adding carbon to convert it into steel. 
It is commonly represented that the failure of the early 
attempts at direct steel-making by the Bessemer process 
arose simply from the difficulty of determining the right 
moment at which to stop the blow, and thereby to regu- 
late the proportion of carbon ; and that the whole advan- 
tage of the spiegeleisen is the means it affords of doing 
this. Dr. Percy says:—“‘In attempting to produce 
steel by the methods specified by Bessemer, it has hitherto 
been found very difficult, if not impracticable, af least 
in this country, to ascertain with certainty when decar- 
burisation has proceeded to the right extent, and when 
therefore the blast should be stopped. Accordingly the 
plan now adopted is to decarburise perfectly, or nearly 
so, and then add a given proportion of carbon in the 
state in which it exists in molten spiegeleisen, the precise 
composition of which should of course be known.”* 
Neither in Dr. Percy’s nor any other account of the Bes- 
semer process do I find that the necessity of complete 
decarburisation as a means of completely separating the | 
silicon is fairly appreciated. 
If merchantable steel could be made from English pig- 
iron by simply stopping the blow before complete decar- 
burisation, Mr. Bessemer would surely have produced some 
good steel in the course of his long and costly efforts which 
preceded the idea of introducing the spiegeleisen, forit must 
be remembered that the quantity of carbon required in steel 
extends over a very wide range—that steel may contain 
from 0°40 to 2'00 per cent. of carbon, and that steel with 
every degree of carburisation within this wide range is in 
demand in the market at good prices, provided it be free 
from phosphorus, silicon, &c. Nothing is practically 
easier than to stop the blow at such a moment as shall 
ensure a degree of carburisation somewhere between this 
wide range ; and there can be no doubt that, in his early 
experiments, Mr. Bessemer, like other inventors of direct 
processes, made an abundance of iron that was duly car- 
burised within the above-stated limits, although he failed 
to produce useful steel. 
Dr. Percy's qualification, “at least in this country,” is 
rather curious. He has probably learned that steel has 
been directly made in Sweden (though he does not men- 
tion it in his work) by the Bessemer process, and he seems 
to attribute this to the superior ability of the Swedish 
operators, enabling them “to ascertain with certainty 
when decarburisation has proceeded to the right extent.” 
I differ entirely from Dr. Percy in this conclusion, being 
convinced that Mr. George Brown, the manager of the 
Bessemer Department at the Atlas Works, Sheffield, who 
was the first to work the Bessemer process with com- 
mercial success, is better able (on account of his much 
greater experience and thorough knowledge of the work) 
than any of the Swedish manufacturers, to determine 
when any required degree of decarburisation has been 
attained. It is not the superior skill of the Swedish 
operators that has enabled them to make steel directly by 
the Bessemer process; but the fact that they, like the 
Styrian workers, used a very superior charcoal-iron to 
start with ; and that the blowing out of all the carbon was 
not absolutely necessary for the sufficient purification of 
this quality of iron. 
W. MatTiEuU WILLIAMS 
* “Metallurgy,” “Iron and Stécl,” p. 814. The italics are my own. 
NATURE 
[ Aug. 18, 1870 
ON THE NATURAL LAWS OF MUSCULAR 
EXERTION 
HE experiments published by Mr. W. Stanley Jevons, 
in NATURE on the 30th June last, illustrate well two 
laws of muscular exertion which were established by ex- 
periments made by myself in 1862 and 1863. These laws 
may be thus stated :— 
Law 1. The work given out by a single group of muscles, 
in a single contraction, is constant. 
Law 2. When the same group of muscles is kept in con- 
stant action, the total work done by them until fatigue 
sets in, multiplied by the rate at which they are com- 
pelled to work, is constant. 
Mr. Jevons’ first series of experiments, in which different 
weights were thrown by the arm to various distances on 
level ground, illustrates the first law. In throwing weights 
in this manner, the arm, aftera little practice, instinctively 
pitches the weight at the angle corresponding to the maxi- 
mum range, and as the maximum range is proportional to 
the square of the velocity of projection, it may be used to 
replace that velocity squared, in estimating the work done 
by the arm. 
The total work done is the same as if the weight used 
and the weight of the arm were concentrated at the centre 
of oscillation of the loaded arm, regarded as a compound 
pendulum. 
Let us assume 
w weight held in hand ; 
oB weight of arm ; 
v velocity of centre of oscillation. 
By Law 1, the work done is constant and is represented by 
(w + x) v? = const. (1) 
Let 
V = velocity of hand ; | 
Z = radius of oscillation ; | 
a = length of arm. | 
then ew as (2) 
a 
It is easy to show (assuming the arm to be a uniform 
cylinder) that 
bind 2) ANG MUSE) (3) 
a) 3. (27ers) 1 
By means of (2) and (3), equation (1) becomes | 
; 5 \2 | 
(w+ x) (3 w+ 4) XR =e (4) — 
(@w + a) 
where & denotes the range (proportional to V?) and A 
denotes a constant, if Law 1 be true. | 
Mr. Jevons’ experiments give the following correspond- — 
ing values of wand A. | 
ww 
56 lbs. 
28 
R 
1°84 ft. 
370 
6°86 
10°56 
14°61 
18°65 
23°05 
ier S 27°15 5, 
We are required to assign certain values to 2 and A, 
which will make equation (4) best coincide with the eight 
simultaneous values of w and & found by observation. 
I find by trial that these values are 
14 
7 
4 
I 
2 
” owed 
+ = —8:0lbs. 
A = 262°2, 
If we solve equation (4) for 2, we find 
A(2w+2)? 
(5) 
(wa) (3 way" 
