Aquatic Oligochaeta from Japan and China. 
95 
specimens, drawn under the camera lucida, which explains how the marked upturn- 
ing of one lip may co-exist with a flange-like expansion of the margin (there seems 
however to be really no such expansion at the posterior margin of the end of the 
tube). 
Secondly, I stated that the dorsal vessel was ventral in position, lying near and 
to the left side of the ventral nerve cord, and was only actually dorsal in the first 
eight segments. Nomura finds that it is near the ventral vessel in the segments con- 
taining the genital organs, but ' ' behind these segments it reassumes its position on the 
dorsal side." And to this may be added that Nomura makes the supraintestinal 
vessel originate from the dorsal vessel in segment v, and open into it again at the 
hinder end of the body ; in my previous species the supraintestinal was (as usual) 
confined to a few segments, from v to ix (though another vessel could be traced on 
the right side of the intestine as far back as xxi). As I was particularly interested 
in the circulatory system of the Oligochaeta, and especially of the Microdrili, at the 
time, I think these statements may be accepted (see also 19). 
Notwithstanding that Nomura does not bring forward the real points of differ- 
ence at all, I believe he is right in saying that the worms are the same. I have care- 
fully examined a number of the Kyoto specimens, and I have no doubt that they at 
least are the same as my L. socialis ; the dorsal vessel is ventral throughout, except 
in the most anterior segments (fig. 6); the end of the penis-sheath shows a strongly 
upturned margin on one side (fig. 7) ; and there is no trace of a supraintestinal vessel 
in the middle of the body (fig. 6), though a section is perhaps scarcely conclusive 
evidence on this point. 
Now the species which I received from Kyoto is apparently common, — the speci- 
mens were bought in the market, where they were sold as food for goldfish, — and it 
must certainly have been represented among Nomura's three species, — the only spe- 
cies of Limnodrilus in his ' ' fairly extensive collections made in different localities in 
Tokyo." Of these three, the one in which the length of the penis-sheath is 30 — 33 
times its breadth, and the one in which it is 3 — 4 times, cannot enter into considera- 
tion; hence I believe I am justified in assuming that the form I have received from 
Japan is his form B (penis-sheath 10 11 times as long as broad) to which he restricts 
the name L. gotoi. 
The last question concerns the nomenclature of this species. The rule is that 
when a species is divided into two or more restricted species, the name of the origi- 
nal species must be retained for one of the restricted species. I take it that the 
rule refers to cases where the original account or diagnosis is a valid description of 
some group of forms, which by a further refinement of observation is shown to be 
divisible; the original description includes both. In the present case the original 
description is (according to Nomura's supposition, which I have accepted) not a vaHd 
description of any form or group of forms whatever, and includes neither species 
(since taking some characters from one, some from another, it is at variance with 
both). The only course is therefore to drop the name (unless it should hereafter be 
shown that a form corresponding to the description does actually exist). 
