THE BLOOD-CORPUSCLES OF THE ANNELIDES. 489 



into direct communication with the blood-vascular system, there 

 was any likelihood of, as there could be no need for, a lacuna exist- 

 ing to connect the two systems. But where, as in Branchiobdella > 

 the perivisceral cavity has not been differentiated even into lateral 

 and median vessels, not to say not into capillaries, and where the 

 blood-vascular system proper itself remains in a condition of com- 

 parative simplicity, failing to develop a capillary system upon the 

 walls of the intestine, and having its two great trunks connected 

 by simple arches of anastomosis, the probability that a lacunar 

 should exist, as in so many other cases, in supplementation of the 

 missing capillary intercommunication, appears, as my above-given 

 quotations abundantly show, to be very great. 



If BrancMobclella is to be considered a leech, it is very strange 

 that it should differ from its allies in not having its perivisceral 

 system freely and openly communicating with its blood-vascular, 

 the acknowledgedly existing difference that its perivisceral system 

 is peritoneiform and not vasiform being a very much smaller dif- 

 ference, as the lacunae of one genus are often enough replaced by 

 vessels in an allied one. If BrancMobclella is to be considered one 

 of the Ann elides in the restricted sense, as it has, so far as I know, 

 been for the first time proposed in Gegenbaur's 1878 edition of his 



* Grundriss der Vergle^chenden Anatomie ' to consider it, I should 

 still consider the question an open one so far as authority goes, 

 following in this Professor Gegenbaur, who (1. c. p. 179) repeats 

 the words used in the earlier edition of it (p. 198) and in the 



* Grundziige' (ed. 1870, p. 234): — 



'Dieser aus einem lacunaren system hervorgegangene Gefassapparat ist auf die 

 Hirudineen beschrankt, denn bei den Anneliden ist die Scheidung des Gef ass-systems 

 von der Leibeshbhle fast durchgehend entwickelt. Wo sie fehlt, sind nicht Weiter- 

 entwickelungen, wie sie die Differenzirung der Leibeshohle der Hirudineen bot, 

 sondern Riickbildungen im Spiele.' 



Now, I submit that the parasitic habits of Branchiobdella are, on 

 general grounds, likely to have produced such a readily-produced 

 ' Uiickbildung ' as that of leaving a vascular system open at one or 

 more points. And as a third warning against rashly stating that 

 the vascular system of Branchiobdella • most certainly does not com- 

 municate ' with its perivisceral, I will draw attention to the fact 

 that in Malacobdella, which, whether it be nemertine or leech, had 

 anyhow been supposed to possess a closed system of vessels, whether 



