these cases the same individual egg gives the extremes in breadth and length.
Habits very like those of Ixulus but Erpornis seems to keep much to the tops of very high trees. In the non-breeding season it is not usually found in heavy forest but prefers the thinner outskirts of big forests or the smaller forests which generally fill the ravines and pockets in the grass-lands. It was common in the thin deciduous forest in the North of the N. Cachar Hills, where we found it in small parties diligently hunting the smaller branches and twigs for insects. It is a very silent bird and I have not heard its note.
Subfamily LIOTRICHINÆ.
This subfamily contains a number of genera the placing of many of which is a matter of no little difficulty. Since Gates wrote the first volume of the first edition of the Avifauna we have learnt a good deal which has enabled us to eliminate several genera which are obviously non-Timaliine, but further examination of material anatomically may assist us to place yet others in more suitable positions than the present.
Of the 16 genera included by Gates in his Liotrichinæ, the following five have been removed to other families. Irena is now placed in Oberholser's new Family Irenida'; MelanocliJora has been transferred to the Titmouses, Paridce,; Le/ptopoecile and Cepludopyrus have been included in the Regilidæ, whilst Psaraglossa is a true Starling and will be found in the Sturnidoi.
Gf the remaining genera there are still some whose position is especially doubtful. Ciitia aud Ptemthius liave, it has been suggested, many affinities with the Campejihagidce and Harington claims that their nidification also proves this; to me, however, the nidification seems to point strongly to a position somewhere near J'ldiina, Lvidns etc. and, for the present, the reasons for their retention in the subfamily seem greater than for their rejection.
The position of Myzornis is problematical, and careful examination of pterylosis and anatomy and a correct knowledge of its breeding habits are urgently required. Chloropsis is in the same group as Acthorhync/ms and j3£githina and seems to be in many ways intermediate between the Timaliidn and Pycnonotidce, the fact that the sexes differ seeming to determine their position in the former rather than the latter. Hypocolins is a very curious bird with a vei*y sliort first primary and may eventually have to be placed in a family by itself.
The subfamily as now restricted differs from the previous subfamilies of the Timaliidæ and from the Pycnonotidæ in having the sexes differing in coloration; the young are very like the adults but rather duller; the wing and tail are generally not greatly different in length; the first primary, with the exception of Hypocolius, is about half the length of the second; the wing is fairly rounded but longer and more pointed than in the preceding
