The Ea rth worms of Ireland, 1 1 9 



decaying trees. It has often been confused with another closely allied 

 species which Eisen first differentiated under the title of Allolobophora 

 subrubiainda. This worm is very widely distributed, and when once seen 

 is not easily mistaken for any other, notwithstanding the fact that its 

 girdle or clitellum occupies almost exactly the same position as that of 

 one or two other species. It is true that the Gilt-tail {Allolobophora sub- 

 rubicunda, Eisen) is by no means confined to woodlands, but its affinities 

 are entirely with the Dendrobcence, and it specially delights to live among 

 fallen and decaying leaves, dead branches of trees, and similar vegetable 

 debris. I have found it depositing its egg-capsules quite under the bark 

 of decaying trees. 



When Eisen established the genus Dendrobccna, it is remarkable that he 

 did not place therein his new species Allolobophora arborea. It is described 

 as an arboreal or dendrobaenic species, and its characters were in many 

 respects so similar to those of his type of the new genus that at first we 

 are astonished to find the two placed under different genera. The fault 

 lay in the fact that Eisen placed too much stress upon one character, to 

 the exclusion of the rest. With him, any worm whose prostomium cut 

 the peristomium in two was a LuDibriciis, whatever other characters it 

 possessed. In Dendrobmia the prostomium occupied about three parts of 

 the peristomium, while in Allolobophora the prostomium onl)'^ slightly cut 

 into or divided the buccal segment. It is now found that this is far too 

 arbitrary and unnatural an arrangement, and that while undoubtedly 

 every true Lumbricus has the peristomium completely divided by the 

 hinder process of the prostomium, yet not every worm with this feature 

 is a true Lumbricus. Want of attention to this fact has led to further 

 confusion in the case of a recently discovered worm which Levinsen has 

 described as Lumbricus eiseni. This worm, which was first described 

 from specimens found at Copenhagen, has been obtained by Rosa in 

 Italy, and by myself in various parts of Great Britain ; and is a true 

 Dendrobcena, notwithstanding the fact that it has the buccal arrangements 

 of a typical Lumbricus. In colour and in the disposition of the setae it 

 somewhat closely resembles Lumbricus purpureus, Eisen, but there the 

 resemblances end. The true Lumbricus has five or six girdle segments, 

 in this worm there are eight or nine. In Lumbricus the tubercula pubertatis 

 stretch across the four inner segments of the girdle ; here they are 

 absent, or if present their position is abnormal. In Lumbricus there are 

 two pairs of spermathecEe, in this worm they are entirely wanting. 

 Lumbricus emits no yellow fluid ; this species does, though not always. 

 Lumbrictis is a true earthw^orm, this is as truly dendrobaenic. Surely 

 these are characters which cannot be ignored, and show conclusively 

 that the mere shape of the prostomium is an insufiicient generic 

 character unless accompanied by others which are permanent. 



We are now in a position to consider the several British species of the 

 subgenus Dendrobcena which have so far been observed and described. 



Genus Allolottophora, § Dendrobaena = Group No. 3 of Rosa's 



Classification. 



I. A. (Dendrotoaena) celtica, Rosa.— Prostomium only partially 

 dovetailed into the peristomium. Individual setae somewhat widely 



