216 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
this furrow is a complete transverse series of non-segmental organs, and posterior to it another and 
the metameric. organs, including a pair of eyes. Ventrally this somite bounds the mouth posteriorly 
and is represented by a single annulus. Somite iv is also biannulate. Tire two annuli are of nearly 
equal width, or the first, slightly the wider. That the latter represents «1 and «2 seems evident from 
the presence of traces of a furrow passing anterior to the eyes and lateral sensitise, and by the presence 
of an additional series of organs anterior to this furrow. The dorso-median sensillte have moved 
forward slightly anterior to the traces of this imperfect furrow. Somite in is very imperfectly 
biannulate, the faint cross furrow not continuing to the margins. The non-segmental sense organs, 
though reduced in number, form two complete transverse rows. This somite is further interesting 
because it bears two pairs of eyes, the outer belonging to the dorso-lateral series and being the largest 
and most anterior of this series. They are crowded toward the posterior margin of the anterior incom- 
plete ring. The second pair of eyes represent the dorso-median sensillse and have moved forward to 
become included in a common pigment mass with the eyes on somite n. Whitman (’92) has described 
a similar condition in Clepsme ( Placobclella) . The eyes of this sixth pair are of small size and were 
detected only upon the examination of sections. Somites i and n are only imperfectly separated by a 
partial and slight furrow, which in many cases appears to be wanting altogether. The first pair of 
eyes is situated on the posterior part of the annulus representing n, but the other metameric sensible 
are very small and difficult to distinguish from the scattered organs. 
Considering the anterior end as a whole it will he seen that the transition from the uniannulate 
or even coalesced somites of the prostomium and lip to the complete quinque-annulate somite is a very 
gradual one. The appearance of a new ring is first heralded by the gradual separation of an additional 
row of non-metameric sense organs, then by. the appearance in the dorso-median region of a faint fur- 
row, which travels toward the margins, becomes gradually deeper, and creeps around to the ventral 
side toward the median line, where it finally becomes complete. These processes are always more 
advanced 1 in the posterior than the anterior portion of the somite, and in every stage of development, 
from the beginning biannulate somite m to the nearly complete somite vm, the portion of the somite 
posterior to «2 is more fully developed than the anterior portion. 
At the posterior end (pi. 12, fig. 3) the series is run through more rapidly, and owing to the 
crowding of the annuli the arrangement of the non-metameric organs could not be satisfactorily worked 
out. Somite xxiii is complete. On xxiv the two posterior annuli (65 and 66) are short and the 
dividing furrow faint or sometimes obliterated in the middle part,. Somite xxv is quadri-annulate, 
65 and" 66 being represented by «,3; 61 and 62 are also shorter and not fully developed on the ventral 
sides. These two somites, therefore, are more differentiated anteriorly than posteriorly; that is, like 
the anterior somites they are developed from the end toward the center of the body. There is a 
sudden change from the four rings of xxv to the two of xxvi, and there is here no definite clue to the 
whereabouts of the rings (61 and 62) which, as such, are wanting in somite xxvi and in xxvii, which 
is precisely similar. These two somites also differ from those which immediately precede them in 
their stronger posterior development. They may, therefore, be compared with the anterior bian- 
nulate somites hi, iv, and v, in which the easy transitions show conclusively that the annuli 61 and 
62 belong potentially to that portion of the anterior and usually larger annulus which lies anterior to 
the metameric sensitise. The almost universal position of the latter on the posterior part of the annulus 
gives added force to this view. The resemblance of such biannulate somites to the complete somites 
of Microbdella has already been pointed out (Moore, 1900), and it, need only be added that the persist- 
ence of this type of somite at both ends of the leech’s body, under mechanical conditions which appear 
to stimulate neighboring somites to growth in opposite directions, is significant of a probably phyletic 
meaning. 
Three and sometimes even all four pairs of the dorsal sensillae may be traced serially on the pos- 
terior sucker; but they are very irregularly developed, sometimes multiplied by division, sometimes 
reduced in number by concrescence or suppression and always variable in position, so that they have 
little value in the determination of metamerism (pi. 12, fig. 3). 
Color . — The smaller examples only present a distinct color pattern, which becomes more obscure 
and diffuse with increase in size. Some specimens have been preserved in formalin, and the colors 
are described from these; but they probably have undergone some alteration, so that, only the pattern 
is significant. Of a specimen measuring 22 mm. in length the ground is a clear reddish clay color, 
1 It is assumed for descriptive purposes that the processor development is a progressive one by increasing complexity 
from before backward. 
