20 



DR. J. F. GEMMILIi ON 



[May 12, 



press them widely apart. The growth in bulk of this mass which 

 causes this pressure can only have occurred at a relatively late 

 stage. Almost perfect coalescence of the medullee has thus been 

 allowed through secondary fusion, while the upjDer parts of the 

 spinal cords have been kej)t markedly composite. Further, there 

 is reason to believe that nearly all the composite mesenchymal 

 structures occuriing at the transitional region have been produced 

 through secondary fusion. 



The monstrosities grouped under Class III. (p. 16) do not 

 exhibit symmetrical lateral union, owing to their marked ventral 

 convergence, and, accoitlingly, what has been said above I'egarding 

 the behaviour of different organs at the transitional region does 

 not apply to this Class. 



In all cases except those of pure ventral union (for an example 

 of which see p. 18), the change from the double to the single 

 condition is effected at the expense of the inner or adjacent 

 elements, while the outer elements become the right and left sides 

 of the single region of the body. This rule has long been recog- 

 nised as holding good in similar double monstrosities among the 

 higher vertebrates. For the sake of easy reference, it may be 

 useful to tabulate here the piincipal composite structure, produced 

 by fusion of adjacent elements, and mentioned in the descriptive 

 pai't of this paper : — 



As a rule, where adjacent muscles persist, there are also 

 corresponding motor nerves. But the small muscles described as 



