THE BIONOMICS OF ENGLISH OLIGOCH^TA 107 



and Notts, by the Dodder at Ballsbridge, Dublin, and by the 

 stream at Swords ; and have received it from Epping Forest. 

 The forms at Kew were large, with correspondingly large 

 cocoons, while those at Malvern were small with small cocoons. 

 It is in many ways a most curious worm, and seems, like 

 O. gracile, to be gradually working westward. 



Constancy and Variation. — This reference to the two forms of 

 H. oculatus Hoffm. leads me naturally to some remarks on the 

 tendency to change in some worms, and the evidences of 

 stability in others. The most stable English worms are the 

 four species of Lumbricus and the three species of Octolasium. 

 Out of the thousands of specimens which I have examined 

 during the past quarter of a century, it has rarely been my lot 

 to see any varieties of either. Some years ago I recorded a 

 short-tailed form of Lumbricus for Calverley near Leeds, and 

 some Continental writers affirm that the girdle of L. terrestris 

 extends over more than six segments, but I have never seen a 

 single case of this kind in England. 



It might here be remarked that normally the girdle in the 

 genus Lumbricus extends over six segments, while the tubercula 

 pubertatis occur as a band on the innermost four. Further, 

 there is a regular gradation in the matter which is peculiarly 

 interesting. This may be shown by the following chart, in 

 which the figures show the segments covered by the tubercula : 



1. L. rubellus Hoffm. . 28, 29, 30, 31. 



29, 3°, 3i, 32. 



2. L. castaneus Sav. , 



3. L. melibceus Rosa 



4. L. tyrtaeus ? . 



5. L. studeri de R. 



6. L. terrestris L. 



7. L. papillosus Friend 



3°. 3i» 32, 33- 



31, 32, 33, 34- 

 32, 33, 34, 35- 

 33, 34, 35, 36. 

 34, 35, 36, 37- 



8. L. festivus Sav. (= rubescens Friend) 35, 36, 37, 38. 



No. 4 is doubtful, but in view of the regularity here displayed 

 it seems impossible to believe that there is not a true form to 

 fit the niche. But while the tubercula are constant it is curious 

 to observe that the girdle is variable in one or two instances, 

 and these become instructive accordingly. Why is it, for 

 example, that the Irish worm L. papillosus has only five girdle 

 segments instead ot six, and has a pair of large papillae on each 

 side? L. melibceus similarly has only five girdle segments. 



The three species of Octolasium found in England are like 



