142—8 INDIAN TERTIARY AND POST-TERTIARY VERTEBRATA. 
and in the form of the neural spine ; altliough differing by its shorter centrum, and 
the other characters in wliich it differs from Leptoptilus. The writer has not found 
any otlier birds whose vertebra approaches so nearly to the fossil as those mentioned 
above ; and all the other genera have vertebrce very much smaller tlian the specimen 
under consideration. 
Distinctness and affinities. — The foregoing comparisons show that the j)resent 
specimen probably indicates a Siwalik stork (or allied form) exceeding in size the 
largest Tjeptop)tiliis argala, and specifically distinct from any living bird. The 
circumstance that the limb-bones of L. fatconeri are practically indistinguishable 
from those of L. argala renders it almost certain that the present vertebra does not 
belong to the former. Whether the characters pointed out as distinguishing the 
fossil vertebra from the existing genera with which comparisons have been made 
should bo regarded as of generic or of specific value appears extremely doubtful ; 
and it, therefore, seems advisable not to attempt to give any name to the present 
large Siwalik bird. The distribution and relations of the genera Leptoptiliis, Mgcteriay 
and Balceniceps are such as to render it highly probable that an allied extinct genus 
should have formerly existed in India. 
Order : ANSERES. 
Family: ANATIDJE . 
Genus : MERGES, Linn. 
Distribution . — The genus has a wide distribution, being found in the Paliearctic, 
Nearctic, and Neotropical regions. It has apparently been hitherto unknown in the 
fossil state.^ 
Merges (?) sjd. 
History. — The specimen on which this determination rests is noticed here for 
the first time. 
Cervical vertebra. — In jdate XIV., figs. 3, 3a, 3b, there are given three views of 
an early cervical vertebra of a bird obtained by Mr. Theobald from the Siwaliks of 
Asnot, Punjab. Fig. 3 represents the hasmal, fig. 3a the anterior, and fig. 3b the 
neural aspect. The specimen lias lost both pairs of zygapophyses, a portion of the 
neural spine, and the posterior part of the lateral expansions of the centrum. 
This vertebra is characterized by the extreme flatness and 'width of its haemal 
surface (fig. 3) ; and in this respect comes nearer to Mergus than to any other genus 
with which the writer has been able to compare it. It is still shorter and wider 
than the corresponding vertebra of M. serrator, but indicates a bird of about the 
same size. The writer has been unable to compare the specimen with M. merganser.^ 
M. albellus, or 31. castor. 
Although the generic determination cannot be considered as certain, the present 
1 Thii so-ca.)lodL Mcri/its rortzoHi=Sula ronzoni. 
