48 BROOKLYN MUSEUM SCIENCE BULLETIN 3. 3. 
well-marked mutual differences." However, as only the larvae of the 
European P. nigra, affims and micans were known to Dr. Boving, he is 
of the opinion "that at some day intermediate forms between them and 
the other Donaciinae larvae will perhaps be found, so that this genus, 
too, cannot be maintained as to the larvae." 
In consideration of all these facts and being unable to point out 
better and stronger characters, Plateumaris is treated as a subgenus of 
Donacia. 
The vaginal plates of the females of the subgenus Donacia are 
rarely exposed and I did not find the time to extract and study them, 
except of a few species, but as far as I have seen they seem to offer good 
additional characters for separating at least some of the species. They 
are generally of softer texture than those of the subgenus Plateumaris 
but by extracting them are easily distorted. Both plates are usually 
parallel and of equal width and length, similar to those of harrisi figured 
on plate V, fig. 16, and are generally arcuate or truncate at apex. Ap- 
parently they cannot be used as an ovipositor and the females of most 
of the species very likely lay their eggs in the flowers, on the leaves or on 
the stems of plants. 
Mr. Leng in his revision, published thirty-three years ago, separated 
the species of Donacia into four groups. These groups with the exception 
of group A and excluding harrisi from group C are natural and it was my 
intention to accept them. However, the relative width of the mesosternal 
process between the middle coxae used in separating the groups B and C 
is variable and not uniform in all the species of both groups and differs 
also in the two sexes of some of the species. Though I have looked for 
other characters for separating the species into groups I did not succeed 
in finding them. The species of Mr. Leng's group D belong all in 
Plateumaris. The species of his group A are widely different from each 
other but also from our other species. 
The late Edmund Reitter in 1920, following Prof. Jacobson, divided 
the European species of the subgenus Donacia into seven groups. Some 
of our species would fit in groups four and five but the rest scarcely in 
any of the others, besides the groups are by no* means natural. Our 
North American Donacias are apparently much more variable and differ 
more in form, sculpture, etc., from each other J;han the European species. 
In regard to variation the prothorax of all the different parts of the 
insect is the most variable in shape and sculpture, in some species ex- 
ceedingly so, in others less. Of certain species two or three specimens 
could be selected, which, without the intermediate forms, would be 
