fibre, cotton, and grass-down, and lined as usual with fine grass-roots.  Sometimes these nests are long and purse-like, and sometimes globular, either attached to, or pendent from, two or more twigs.  One nest before me, a sort of deep watch-pocket, suspended from five twigs of the jhao (Tamarix dioica), measures externally 2.75 inches in diameter, is a good deal longer at what may be called the back than the front, and at the back fully 5.5 long.  Internally the diameter is about 1.5, and the cavity, measuring from the lowest portion of the external rim, is 2.5.  This is a very large nest.  Another, built between three leaves, has an external diameter of about 21/2 inches, and is externally not above 3 inches long.  It is unnecessary here to describe the beautiful manner in which, when it makes use of leaves, this bird sews them together, as this has already been well described by others where O. sutorius is concerned, and P. stewarti is, in some cases, when forming a nest with leaves, fully as neat a workman.

The nests vary so much, and I have heard so much, discussion about them, that having seen at least a hundred and having taken full notes of some twenty of them, I shall reproduce a few of these notes:—­

“Agra, July 17th.—­Two nests—­one nearly globular, composed entirely of fibrous roots, hair, wool, and thread, and lined with fine grass, suspended by a few fibres and hairs between the fork of a branchlet in a little dense bush of Indian box; the other, suspended from the tendril of an elephant creeper, was principally formed by one of the leaves of this, to which, to form the remaining third of the exterior, a second leaf of the same plant was carefully sewn.  Interiorly there was a little wool, and at the bottom fine grass.

“July 20th.—­On a furash-tree (Tamarix furas), beautifully made of fine soft wool, shreds of tow and string, very fine grass and grass-roots, and the bottom neatly lined with very fine grass-roots.  In shape the nest is like one half of a long old-fashioned silk purse, round-bottomed and very compact, with a long slit-like opening on one side towards the top.  It contained five eggs.

“July 26th.—­Two nests, one formed almost entirely in a single mango-leaf, the sides of which are curled round so as nearly to meet, and then laced by a succession of cross-threads of cobweb, carefully knotted at each place where the margin of the leaf is pierced.  The intervening space is closed by fine tow, wool, and the silky down of the cotton-tree, with just the top of a small mango-leaf caught in from above so as to form an arched roof.  The other nest was rounder in form, having less of a leafy structure.  It had, however, the leaf of the Phalsa forming the back and sides (partly), whilst the whole of the front was composed of soft wool, tow, dry grass-roots, thread, and a few pieces of the soft tree-cotton.  It had a neighbouring leaf just caught in on one side.  This contained four fresh eggs.