3 58 Rev. F^. Morice’s further Notes on 
and the present were the same. He gives no character 
for either sex that I cannot recognise in the Saleve insects; 
and liis statement that the species is attached exclusively 
to Lotus is borne out by all the facts that have come before 
me. 
2. Osmia manieata, Morice. 
In 1900 I could only record two examples of this species 
(both ^ ^), one from Algeria, the other taken by the late 
Sir S. S. Saunders probably in the Ionian Islands. I 
have now quite a long series of both sexes, and can record 
it from the following additional localities : Spain (Granada), 
South Italy (Taranto and Brindisi), Greece (neighbourhood 
of Athens and Olympia, both sexes common in May 1901), 
Asia Minor (Smyrna ^ and ^). Its range therefore extends 
over the Avhole length of the Mediterranean. 
0. mcinieata, in both sexes, is generally at once recog¬ 
nisable simply by its great size. Its length may extend 
to 13 or even 14 mm. (that of adunca only from 9 to 11), 
Its breadth is still more remarkable, quite twice (!) that 
of a normal cidunea in all my specimens. This regular 
difference in size, and still more in proportions, makes 
it perfectly easy to separate examples of the two forms; 
and, as shown in my former paper, the ^ ^ differ entirely 
in the structure of the concealed 6th ventral-plate. In 
the $ $, however, I have quite failed to recognise any 
points of detail on which a reliable “ character ” for their 
separation can be based. The calcaria, indeed, are usually 
(perhaps always) somewhat rufescent in manieatet (black in 
adunea), and the antennae also tend in the former species 
to show rufescence beneath, but the extent of this rufes- 
cence varies. The normal number of wing-hoohs seems to 
be greater in manieata than in adunca (13-14 against 
11-12); but, as we commonly find large and small exam¬ 
ples of a single species differing in this wa}^ 1 have some 
hesitation in suggesting that such a difference may be 
here “specific.” Still when, as in my collection, a long 
series of manieata and another of adunca from many 
localities are exhibited side by side, the general “ habit ” of 
the two forms is so obviously dissimilar that no amount of 
common characters can make them seem identical; and 
there can at least be no doubt that the ^ ^ differ markedly 
and regularly (for I have dissected many specimens of both) 
in the paradoxically developed 6th ventral-plate of the 
