6i8 JOURNAL, BOMB AY NATURAL HIST. SOCIETY, Vol, XXVIII. 
her snare with a system of branching spokes. All her radii are not, as in the 
snare of the Araneus, clear straight lines undivided from end to end. On the 
contrary, many of them after a greater or less distance break into two, and thus 
reach the circumference as a pair of spokes. Let us investigate how this is done. 
The NepMla, unlike the Araneus, does not first complete her radii before com- 
mencing at her temporary spiral. Both stages occupy her at the same time. 
Now she lays out a few radii, now she takes a few turns at the spiral, then she 
resumes the radii again. Furthermore, while constructing her radii, she does 
not work from the centre, but rather from the last turn of the temporary spiral 
which she has laid down. If this statement is clear, then it should be obvious 
that, on the spider’s return journey after fixing one end of the second radius 
at the circumference, she does not continue all the way to the centre, but halts 
at the outermost turn of the temporary spiral and there anchors her thread. 
Thus at the point where the radius and the temporary spiral meet there the 
radius will appear to divide. As a consequence of this peculiar mechanism 
the radii are seen to branch in all parts of the snare. Some divide near the 
centre, others near the circumference, and the point of division is always where 
a radius and a spiral meet. Thus we cannot count the number of the radii, 
at least the number will be much less near the centre than near the circumference 
of the snare. An idea of this branching wiU be gained by making the circular 
count. An inch from the centre they number 28 ; four inches from the centre 
they are 80, and at the circumference 112. Indeed it is clear that, in a strict 
sense, they are not radii or spokes at all. 
I pass now to explain why these peculiarities exist, since there is always a 
good reason for everything we see. The great increase in the number of the 
radii depends of course on the immensity of the snare. They are the scaffolding 
of the architecture and must support it at every point. Twenty are sufficient 
for the little web of the Araneus ; over a himdred are required for the Nephila's 
sheet. Moreover the radii are much closer together than they are in 
the ordinary snare. This likewise depends on the vast area of the sheet. Com- 
pare it with a mechanical wheel. A small wheel can be made with only a few 
and widely-separated spokes ; they are sufficient to give it rigidity and strength. 
But increase the circiunference, and many more must be provided, and they 
must not be far apart. The rim must be supported at many and closely-con 
nected points, otherwise the structure will coUapse. The mechanism is essen- 
tially the same in our comparison between the two snares. 
At each journey the Nephila completes two radii, while the Araneus manu- 
factures only one. This deviation has a value ; it makes for economy in the 
architecture, an economy not only in time, but, since each radius is a single 
thread, an economy also in the expenditure of silk. The saving of time must be 
of some importance, since the laying of the radii is a tedious work, and a much 
more prolonged operation than in the case of the ordinary snare. So also is it 
a service to the spider not to waste her precious silk, especially in so immense 
a snare which contains such a multitude of lines. Possibly the strength of 
the architecture is diminished by the confinement of each radius to a single 
thread. But compensation is made for this by the great increase in the 
number of the spokes. 
The emplojTuent of branched instead of unbroken radii is another ingenious 
method of securing economy in material and of adapting the system of radiat- 
ing spokes to the extensive area of the snare. Indeed, without the adoption of 
some such mechanism, it would, I think, be quite impossible for the spider to 
extend so wide a sheet. She may be an inimitable architect. But it is beyond 
even her constructive powers to lay out over a hundred lines that diverge 
equally from a common point. At the point of divergence they would be so 
massed and crowded that the spider would be utterly unable to manipulate 
the individual threads. She could never separately distinguish them nor 
