In all the combinations that have been so far described, the dorsal and ventral surfaces of both components were kept in the same direction, so that the ventral nerve-cord of one piece came in contact and fused with the nerve-cord in the other piece. Sometimes it may happen that the components are not quite in the same position, and the end of one nerve-cord may fail to abut against the other one. In such cases Joest thinks that regeneration is more apt to take place in the region of union, and he has carried out a series of experiments in which the pieces were intentionally united, so that they are not in corresponding positions. It is found that if one piece is turned so that the nervous system lies 90 degrees, or even 180 degrees (Fig. 52, _B_), from that of the other piece, the union takes place just as when the pieces have the same orientation, except that the ends of the nerve-cords do not unite. Subsequent regeneration from one or from both components generally takes place in the region of union. It is more difficult to unite pieces of different species of worms, yet Joest has succeeded also in making combinations of this sort. One union between the anterior end of _Lumbricus rubellus_ and the posterior end of _Allolobophora terrestris_ was permanent, and the new worm reacted as a single individual, and lived for eight months. Each piece retained its specific characters, and showed no influence of the other component. By means of a similar experiment we have a way of finding out if one component can influence regeneration taking place from the other piece. Although Joest made only a few observations of this sort, the results show that no such influence is manifested. [Illustration: FIG. 53.--After Joest. _A._ Small piece of _Allolobophora terrestris_ from posterior end grafted upon anterior end of another individual. Oral end free. Four weeks after grafting eight new segments formed. _B_. Same fourteen days later. A new part of thirty-seven segments had appeared at end of former eight segments. _C._ A piece of the body wall of _Allolobophora terrestris_ grafted upon the cut-end (anterior) of _Lumbricus rubellus_. Two months later, as shown in figure, a head had grown on major component. _D._ Anterior and posterior ends of _A. terrestris_ united to make a “short worm.” _E._ A piece of body wall of _A. cyanea_ grafted on side of body of _Lumbricus rubellus_. _F._ Piece of _L. rubellus_ grafted on side of body of another individual to produce a double-tailed worm.]