REPORT OF COMMISSIONERS OF INLAND FISHERIES. 137 



appendage" (D'R') is the mirror "image of the nearer" one (D'L'). 



The claws sketched at the right of the diagram may make clearer 

 the significance of these relations. The arrow in each figure indi- 

 cates the plane of section shown in the corresponding figures at the 

 left. 



One of the important points in this diagram is that a definite 

 morphological relation distinguishes the dorsal (Fig. II) and ventral 

 positions (Fig. Ill) from each other. Namely, that when a pair of 

 extra processes arise on the dorsal or smooth border of the dactyl, 

 they turn their toothed or ventral borders toward each other (Fig. II) ; 

 but on the other hand, when the abnormal processes occur on the 

 ventral or toothed surface of the dactyl, they turn their smooth 

 borders toward each other (Fig. III). 



We may now compare our actual specimens with the theoretical 

 relations just described. In specimen No. 1 the abnormal pair of 

 process arise from the dorsal surface, and consequently their rela- 

 tions should correspond to those given for the dorsal position in the 

 diagram. A glance at the specimen quickly shows that such is 

 actually the case; the toothed borders oppose each other; the two 

 extra processes are mirror images of each other, and the one nearer 

 the normal dactyl is also the mirror image of this segment; and, 

 finally, it is readily seen that the process nearer the normal segment 

 may be interpreted as an extra left dactyl (D'L') and the other process 

 as an extra right dactyl (D'R'), and consequently that these two 

 abnormal structures evidently morphologically represent an extra and 

 complementary pair of dactyls. In specimen No. 2 the abnormal 

 part arises on the toothed border of the dactyl, and the relations should 

 correspond to those shown for the ventral position in Fig. II of the 

 diagram. In this case the abnormal part is not spatially separated 

 into two distinct processes, nor is there even a slight bifurcation at 

 the tip to indicate that it is a double structure. On the other hand, 

 however, this abnormal process is itself symmetrical in form, and, 

 accepting Bateson's interpretation that "whenever an extra part is 

 itself symmetrical it always may be a double structure," we may 



