WAITE : PLEXI OF NECTURUS, 79 



the movement of the eutire plexus is but the summation of the move- 

 ments of its elements, the individual nerves. (2) Nerve xxii is so incon- 

 stant and contributes so feebly to N. ischiadicus, that, in compensation 

 for this deficiency, all of xxi is taken iip to form N. ischiadicus, and thus 

 it can contribute nothing to N. cruralis. Again, I have shown that, in 

 group B, XXI is relatively weaker, and xx relatively stronger than the 

 corresponding elements of the plexus in group A. Hence a contribution 

 from XX to N. ischiadicus (in group B) serves to make up the deficiency 

 arising from the inconstancy and stronger posterior tendency of xxii 

 (cf. Fig. 4). (3) It is to be noticed that in group A (Figs. 2 and 3), 

 XXI always entered N. ischiadicus and that nerve only, while xx was 

 (see Table) in many cases (a type) divided between N. ischiadicus and 

 N. cruralis. This gives to xx the definite relation of a component part 

 to N. ci'uralis in group A, a relation which might be expected to be 

 retained after migration, i. e. in group B. But in group A xxi in no 

 case had any relation to N. cruralis. It is therefoi-e less probable that 

 in the new condition xxi should form a new connection entering JST. cru- 

 ralis, than that it should give its entire strength to N. ischiadicus, as it 

 does, with which it had the definite relation of a component part in the 

 old condition. The general backward movement of the parts of the 

 plexus, and the former (a type of group A) branched condition of xx, 

 may account for the fact that this nerve sends a branch to N. ischi- 

 adicus in this post-migration condition. 



These considerations lead me to believe that the y8 type is the only 

 condition to be expected in group B, if the plexus has reached its posi- 

 tion there by means of migration of the locus of its strength-centre. If 

 however, interpolation of a presacral segment has occurred, we should 

 expect to find about the same variations in type of topography in group 

 B as in group A. These do not occur, therefore the persistence of the 

 /? type of topography in group B ; the inconstancy of nerve xxii, and 

 the added activity of xix in their contributions to the plexus; together 

 with the slight variation in the position of the first haemal arch shown 

 in this group ; — all combine as evidence against the theory that the 

 position of the pelvic girdle in group B is the result of interpolation of a 

 presacral segment. 



Group C includes three individuals (10% of the total number ex- 

 amined), in which the attachment of the girdle is unsymmetrical. 



Of these one (Plate 2, Fig. 5) bore the right sacral rib on the 19th 

 vertebra, the left on the 20th, while the first hsemal arch was on the 



