WAITE : PLEXI OF NECTURUS. 77 



Xerve xxii has here topographically the same relations as nerve xxi 

 ill group A, but its strength relative to the rest of the plexus is much 

 less, and in two cases (in seven) it fails to send a branch into the plexus. 



The position of the first haemal arch is in every case on the 23d 

 vertebra. 



At first sight the whole plexus in group B seems to have moved with 

 the girdle one segment caudad as contrasted with the position in group 

 A, a condition which might be explained by the interpolation of a pre- 

 sacral segment; but for several reasons I do not believe such an expla- 

 nation to be sufficient. If a presacral segment had been intei-polated, 

 thus entailing a change in position to the extent of an entire segment, we 

 should expect still to see variations in the position of the first hiemal 

 arch. In one half the cases in group A (see Table, p. 81 ) the first haemal 

 arch was on vertebra 23, three segments posterior to the main nerve of 

 the plexus. Hence in group B, if the interpolation hypothesis be true, 

 the hasraal arch ought to occi;r, at least occasionally, in the same rela- 

 tive position, i. e. on the 24th vertebra, instead of maintaining as it 

 does a constant position on vertebra 23, only two segments posterior to 

 the main nerve (xxi) of the plexus. The specimens examined by Parker 

 ('96, p. 712) show the same result, i. e. a constancy of position of the 

 first hsemal arch on the 23d vertebra in all specimens bearing the girdle 

 on the 20th vertebra. Bumpus ('97, p. 473), however, among thirty-five 

 specimens having this position of the girdle finds three (9%) in which 

 the first haemal arch is on vertebra 24. This shows that such a position, 

 though possible, is rare. 



Again, we should not expect the tendency of the last trunk nerve 

 (xviii in group A, xix in group B) to enter the plexus to be so much 

 greater in group B (43%) than in group A (15%), nor should we expect 

 to find its strength in group B distinctly exceeding that of the ordi- 

 nary spinal nerves, as it does, since there is no indication of such an 

 excess in group A. 



If interpolation has occurred, nerve xxii should show about the same 

 average relative strength as the element (xxi) corresponding to it in 

 group A. This is not found to be so, for xxii in group B is relatively 

 and absolutely much weaker than xxi in group A, and by completely 

 failing in two cases (28%) to enter the plexus shows an inconstancy 

 not seen in the corresponding nerve (xxi) in group A. This leads 

 to the conclusion that in group B this most posterior nerve (xxii) 

 is not an essential element of the plexus. 



Nerves xx and xxi in group B correspond, in position relative to the 



