No. 424.] THE BLOOD FLOW IN LUMBRICUS. 325 

each somite through the parietals must be carried to the dorsal 
by the ventro- and dorso-intestinals. It does not seem possible 
for all of the blood to take this course, owing to the small size 
of the intestinal vessels. It has been impossible for the pres- 
ent writers to arrive at a definite conception of the blood flow 
according to Harrington’s scheme. He represents the blood 
as flowing in different directions in different parts of the same 
vessel, and as changing its direction in the same vessel from 
time to time. Our experiments give no confirmation whatever 
to either of these conclusions, except that the direction is per- 
manently reversed in the front end of the ventral vessel and in 
the dorso-intestinals in front of the hearts as above described. 
Temporary reversal of the flow in limited regions might easily 
occur as a result of the animal's movements, and such phe- 
nomena may have led Harrington to erroneous conclusions, 
since he worked only by studying the pulsations through the 
skin with a hand lens. His scheme of the circulation is full of 
inconsistencies, due either to an attempt to avoid overloading 
small vessels and consequent breakdown of the scheme, or to 
the incomplete study of several genera, resulting in a patch- 
work which is not true for Lumbricus at least. It is but 
proper to say that these inconsistencies might have been 
removed had the author lived to see his work through the 
press. 
Bourne states that the lateral vessels carry blood backward, 
Harrington that they carry it forward. Our results positively 
support Bourne's view and, together with the results on the 
parietals, suggest a new interpretation of the morphology of 
the lateral vessels. Bourne considers them to be the sub- 
neural ends of the parietals in this region, dwt assigns to them 
an opposite function. The following facts seem to show that 
the lateral vessels are formed by anastomoses between succes- 
sive parietals, which carry the blood from the head region to 
the dorsal in somites X and XII, and that the lateral system 
therefore represents the parietals in the head. 
I. Absence of true parietals in front of somite XII. 
2. Connections of the laterals with the subneural in several 
somites (VIII-XII) which correspond perfectly to the ventral 
