ely state
very briefly the few facts already established, with some of the
possible explanations.

The spinning _mammulae_ are placed in pairs at the lower part of the
abdomen, near its hinder end, and number four, six, or eight in
different species. They are little conical or cylindrical papillae,
closely resembling the pro-legs of caterpillars, and are composed of two
or three joints, the terminal one of which is pierced with a greater or
less number of minute holes, the sides of these, in some, if not all,
cases, being prolonged into tubes. Through these holes or tubes issue
the fine filaments, which, uniting as they dry in the air, constitute
the line from each spinner.

[Illustration: Fig. 4. Spinners.]

Now the _Nephila plumipes_ possesses at least three pairs of spinners.
Of these, two are much larger than the third, which indeed does not
appear till they are separated. From the _posterior_ of the two largest
pairs _seems_ to proceed the _white_, and from the _anterior_ the
_yellow_ silk, while from the small intermediate pair seem to proceed
very fine filaments of a pale-blue color, the use of which is to envelop
the prey after it has been seized and killed, being drawn out by the
bristles near the tips of the spider's hinder legs. Beside these six
papillae there is, just in front of the anterior pair, a single small
papilla on the middle line, the nature and use of which I have not
ascertained, though I feel quite sure that no silk comes from it. The
large median papilla, just _behind_ the posterior pair, surrounds the
termination of the intestines, and through it the excrement is voided,
the insect for this purpose turning back the abdomen as she hangs head
downward, so that neither the web nor the spinners shall be
contaminated. Now it has recently been ascertained that the minute
globules with which the circles are studded, and the number of which on
a web of average size is estimated at _one hundred thousand_, do not
exist in that form when the viscid lines are first spun by the spider,
but as a uniform coating of gum upon a thread; this gum, of itself and
according to physical laws, soon exhibits little undulations, and then
separates into the globules which have long been observed and supposed
to be formed by the spider. The fact of spiders selecting the night for
the construction of their webs, the difficulty of making any close
observations upon them while so engaged without disturbing them, and the