152 
Mr. H. Seebohm on the 
between the third and fourth. On the other hand, out of ten 
examples in Dresser's and my own collections of A. streperus, 
eight have the second primary equal to the fourth, and in two it 
is intermediate between the fourth and fifth. Dresser has iden¬ 
tified S. arundinacea of SevertzofF with A. streperus, without, 
however, giving any description. I therefore take it for granted 
that the bill is about the size of that bird's (say culmen ’63 to '7). 
From f Stray Feathers' [loc. cit. clause 33) I get the addi¬ 
tional information that the tail is shorter than the wing, and 
the second primary equal to the sixth. The wing is too much 
rounded for either A. streperus or A. palustris; the bill is 
too large and the tail too short for A. agricola; but all the 
three items of information we possess point to Acrocephalus 
dumetorum (Blyth), with which species I am accordingly in¬ 
clined to identify it. 
Salicaria brevipennis (p. 83) is certainly not an Acrocephalus 
at all, the first primary being twice as long as the coverts. W e 
must look for this bird amongst the smaller and greyer species 
of Hypolais — opaca, pallida, rama, or caligata. To decide 
to which of these species it belongs we must have the length 
of the culmen. This is given as 3| lines. This is manifestly 
an impossible measurement. In Blanford's f Eastern Persia' 
(ii. p. 192) we find the following measurements of the culmen 
of the three smaller of these species given— H. pallida *72 to 
•68 inch, H. rama *68 to ’57, and H. caligata '55 to *5. Severt¬ 
zofF's measurements are undoubtedly those from the point of 
the bill to the beginning of the feathers. I shall shortly 
identify, from evidence independent of the length of the cul- 
men, the next species, &. capistrata, Sev., with Acrocephalus 
agricola (Jerd.). SevertzofF gives 4 lines as the length of 
culmen of S. capistrata. The true length of culmen of A. 
agricola is '55. We may therefore construct the following 
table for translating SevertzofF's length of bill in lines into 
true length of culmen in decimals of an inch:—3|=*51, 
3| = *53, 4= '55, 4| = *62, 5 = ‘69, which we shall find very 
useful as we go on. The bill of S. brevipennis being 3| lines, 
equal to culmen -53, there can be little doubt that this bird is 
Hypolais caligata (Licht.). 
