282 Platybdella Anarrhichae 
appears to be eight adradially arranged bundles, that is to say, two 
dorsal, two ventral, and two lateral on each side. Near the elitelluni 
they break up into twelve or sixteen or even more bundles. In Piscicola 
the bundles are four in number and interradial in position. Near the 
clitellum, at a shorter or longer distance, the bundles are usually divided 
into four adradials. But in Abranchus brunneus numerous small 
bundles lead forwards without any definite arrangement wherever they 
have room. From my observations this appears to be the case also 
in Platybdella anarrliicliae (Fig. 4 c.g.d.b.), where the ducts, often in 
threes, run interspersed with longitudinal muscle fibres (Fig. 5). There 
are naturally eight spaces where mixed collections of ducts and muscles 
can run, two dorsal (one of which is indicated in Fig. 4), two ventral 
on either side of the ventral sinus, and two lateral on each side above 
and below the lateral sinusiis. But, far from calling these eight bundles, 
I consider them as ducts running forwards among the muscle fibres 
wherever they have room. I desire to lay stress on the considerable 
resemblances between Platybdella and Abranchus noted in this paper, 
which might lead to their inclusion under one generic name. 
Digestive System. The mouth is situated at the centre of the 
anterior cupuliform sucker, and the anus opens on the dorsal side 
between the last and last but one of the annuli immediately preceding 
the posterior sucker. There is an oesophageal bulb or pharynx into 
which opens a pair of well-marked salivary glands. The alimentary 
canal is straight, and of very nearly the same calibre throughout its 
length, while its walls are exceedingly thin, so that the contents can be 
easily seen, which gives the leech its reddish markings. The contents 
of the canal are always abundant, and consist of bright red nucleated 
blood corpuscles from the gills of the host. Van Beneden and Hesse 
state that there are no caeca, but this, however, is not the case. There 
is the usual single pair of caeca, ventral to the intestine, which cohere 
in four places, being the second stage in the development of cohesion 
from the primitive pair of caeca, and figured in my paper on Calliobdella 
(p. 214, Fig. 5 c. q.v.). Here there are connections not only in the sixth 
and fifth segments of the abdomen (as in Abranchus microstomus), but 
also in the fourth and third, thus only a short piece is unfused. I find 
that the food, i.e. blood corpuscles, tend to collect in these cross con¬ 
nections, which are obviously in a direction at right angles to the body 
axis, and these give to the animal the appearance of reddish streaks 
to which allusion has already been made (Fig. 6). 
