306 On the Spiders’ Power of conveying their Threads [APRIL, 
to the upper end of the arch, and clenched to the middle of the 
piece, as shown in the figure. 
The weights were placed in a scale, F, suspended by a double 
rope, which was applied upon the circumference of the arch, as 
shown in the edge view of it at ff, and made fast to it at the top. 
By this means, as the rope, f f, always drew at a tangent to the 
arch, E, the effective leverage of the load, F, to bend or break the 
piece was in all cases very nearly in direct proportion to the weight 
applied in the scale; and in order to obtain the neat weight inde- 
pendent of the weight of the scale and the arch, a small line, g, 
was fastened at the lower end of the arch, and carried between the 
double ropes, f f, as shown in the edge view. This line being 
conducted over two pulleys, G, G, had a weight, H, applied, suffi- 
cient to counterbalance the weight of the scale and the arch. The 
arch, E, was divided into degrees and parts, which were numbered 
from the middle upwards. ‘lhe angle of curvature was pointed out 
by the edge of an index, k, made of a piece of plate fastened by a 
screw against a prop, K. This index was capable of a slight adjust- 
ment before the experiment began, in order to bring it to zero. 
Whilst the piece of wood continued without any load, and there- 
fore without any flexure, all weight applied in the scale after this 
having a direct tendency to bend the piece, the degree of flexure 
produced by any given weight was noted by the degrees and parts as 
shown by the index upon the divided arch. It would sometimes 
happen when the piece was cross-grained, and much loaded and 
bent down, that it would get a tendency to twist, and the arch 
would go over sideways. To prevent this, a line was made fast on 
each side to the upper end of the arch, and during the experiment 
two persons held the ends of these lines, and kept the arch upright 
by pulling, so as to counteract any tendency it might have to go 
sideways, though without increasing or diminishing the load which 
tended to bend or break the piece. 
SS a eT ee 
ARTICLE LY. 
On the Power that Spiders have of conveying their Threads from 
one Point to another, and of flying through the Air. By 
Carolan. 
(To Dr. Thomson.) 
SIR, 
As the following experiments tend to elucidate the method by 
which the geometrical spider conveys itself and its threads from 
one place to another, and exhibits some curious facts connected 
with that phenomenon, I thought that an account of them might 
not be uninteresting to some of your readers, especially as the 
subject has never been thoroughly investigated. 
In order to ascertain the nature of the transitive power which the 
2 
