1894- EARTHWORMS AND THEIR ALLIES. 47 



these cysts, in fact, doubles the animal's chances of being conveyed 

 to distant places ; the cocoon is, of course, the other chance. The 

 cocoons of earthworms, being larger and often buried at considerable 

 depths in the soil, are not likely to be accidentally attached to the 

 feet of migratory birds, which, moreover, frequently affect the margins 

 of lakes and pools, where the smaller cocoons of the aquatic Oligo- 

 chaeta w^ould be very liable, one would imagine (there are no positive 

 facts known to me), to be caught up and carried away. Another 

 matter to be borne in mind is the more common occurrence here than 

 among the earthworms of the escape of several young worms from one 

 cocoon. The importation, therefore, of a solitary cocoon might serve 

 to found quite a large colony in a new country. These are some of 

 the facts and suggestions which arise from recent work upon the 

 Ohgochaeta as regards the general phenomena of distribution. 



But distribution has z bearing upon another branch of the 

 subject which has received much attention — the classification of 

 the group. The classification of the Oligochaeta affords a literal 

 instance of " Quot homines, tot sententiae." The workers are, un- 

 fortunately, at present few, though this is mending, and everyone has 

 his own individual scheme of arrangement ; indeed, many of us have 

 provided more than one scheme, though at fairly decent intervals. In 

 attempting to decide upon the conflicting claims of different families 

 to occupy a basal position in the series, it is reasonable to be in- 

 fluenced to a moderate extent by the facts of distribution. It is usually 

 held that a wide and discontinuous range indicates an archaic form. 

 Thus, for instance, Peripatns and the Tapir. Now, if we apply this 

 to the earthworms it is perhaps the genus Perich^ta which affords the 

 most striking example. The genus has complete circles of setae, from 

 twenty to upwards of one hundred in number, according to the species, 

 upon each segment of the body, and is found in Australia and India, 

 Central America and the West Indies ; it is feebly represented in 

 tropical Africa and South America. The family Perichaetidae, may 

 perhaps, be split up into four or five genera, but the differences are 

 not great and there are gradations. The principal claim urged in 

 favour of Perichcsta as the rhost archaic existing form of earthworm 

 was the condition of the excretory organs. In most text-books the 

 typical Annelid excretory system is said to be a pair of tubes in each 

 segment of the body, opening into the body-cavity by a ciliated funnel 

 and on to the exterior by a pore. Researches, beginning with the year 

 1885, have shown that among earthworms a very large number of 

 genera do not possess this arrangement at all. In Pevichata, for instance, 

 that system is represented by a dense tufting of the body-walls and 

 septa by innumerable coils of excretory tubes which open on to the 

 exterior by innumerable pores scattered irregularly over the surface 

 of the body. It has been stated by more than one naturalist that 

 these nephridial tufts communicate from segment to segment, so that 

 there can hardly be said to be any metamerism in an organ which in 



