In other experiments pieces of one species were cut from the side of the body and grafted upon the cut-surface of the anterior end (or elsewhere) of another species. In one of these experiments a piece from the side of _A. terrestris_, that extended over five or six segments, was sewed upon the anterior cut-surface of _L. rubellus_ (from which the anterior five segments had been removed). In about a month new tissue appeared on the ventral side between the two pieces, and a little later a complete head developed, whose dorsal side was made up of the small piece (Fig. 53, _C_). The grafted piece was dark, and the new, regenerated part light in color and continuous with the brown color of _L. rubellus_, from which the new part had arisen. It is important to notice that the four segments of the graft are completed by four segments of the new part. After three months the new part had assumed the red-brown color of _L. rubellus_. The color of the grafted piece had not changed. We see in this case that even the presence of a part of another worm in a regenerating region does not have any influence, at least so far as color is concerned, on the new part, even though its segments supplement some of those of the new part. The new tissue seems to have come entirely from the major component, and to have carried over the color characteristics of the old part. It has been shown that when two posterior pieces are united by their anterior ends the combination must sooner or later die, since it has no way of procuring food. The question arises: What will happen if one of the two components is cut in two near the place of union? Will a head then develop on the exposed aboral surface, _because_ a head is needed to adapt the worm to its surroundings, or possibly, if it occurred, because the major component exerts some sort of influence on the short, attached piece, as happens in hydra and in tubularia? Both Joest and I carried out an experiment of this sort, and found that a tail and not a head regenerated, as shown in Fig. 16, _F_. The experiment is, however, insufficient to answer the question, since the region in which the second cut was made is a region from which only a tail (and not a head) arises, even when the oral end of a piece is exposed. In order to avoid this difficulty I carried out another experiment. Two worms had the first five or six segments cut off and the exposed anterior ends of the worms united, as shown in Fig. 16, _D_. Then one of the components was cut off, leaving three or four segments attached to the anterior end of the other component. Although regeneration began in one case, it did not go far enough to show what sort of a structure had developed, but Hazen, who took up the same experiment, succeeded in one case in obtaining a definite result. At the exposed aboral end of the small piece a head and not a tail developed (Fig. 16, _E_). At first sight it may appear that