neio Genus of Olujochceta. 11 



instance of how impossible it is, in this group of animals, to 

 determine affinities by the outside only; alike as they arc 

 superficially, it is not much, if any, exaggeration to say that 

 few genera are structurally farther apart than Oordiodrilas 

 and Meyachcvta. If there were any scope for the action of 

 natural selection in this direction, some might regard this 

 resemblance as an instance of the phenomenon known as 

 " mimicry." It is, however, difficult to see in what way one 

 kind of worm would be advantaged in resembling another, 

 as the characters are not so striking as to appeal to the eye- 

 sight of the natural friends or foes of either. 



The length of Gordiodrilus tenuis after preservation was 

 about 90 millira., the breadth 1 millim. ; the body was bent 

 into a spiral, which I have never noticed except in the long 

 and thin aquatic worms. 



External Characters. 



An examination of a portion of the body-wall mounted in 

 glycerine shows the reason for the curious way in which the 

 worm clings to the finger when handled. Some of the setce 

 are very large ; if the body-wall be examined with a lens 

 only, there appear to be only two pairs to each segment. A 

 more careful examination, however, shows that the normal 

 four pairs are present in each segment of the body, with the 

 exception, of course, of the first ; the lateral pairs are so 

 small as to escape observation, owing to the eye being accus- 

 tomed to the large ventral setae, unless a moderately high 

 magnifying-power is used. The seta3 are strictly paired, and 

 the lateral setai are about a quarter of the bulk of the ventral 

 setEGj the latter are, however, not absolutely of the same 

 size ; the innermost seta of each pair is rather larger than 

 the outermost. There is nothing unusual in the shape of 

 these setai; they have the sigmoid form so generally met 

 with among the Oligochasta, but the tip is perhaps slightly 

 more hooked than is ordinarily the case ; this is no doubt 

 partially a cause of the tenacious way in which the worm 

 clings to the finger. Among worms which have sigmoid 

 sette in both dorsal and ventral bundles [Phreoryctes, Lum- 

 briculidffi, nearly all "earthworms"), it is by no means 

 common to find such a great difference in size as that which 

 has been recorded in Gordiodrilus tenuis. There are, how- 

 ever, several forms where a like difference does occur. In 

 Phreoryctes^ for instance, whicli has besides a certain super- 

 ficial similarity to Gordiodrilus^ there is connnoidy a similar 

 inequality in size between the dorsal and ventral seta?. In 



